Journal of Water Sanitation and Hygiene Development sample issue by IWA Publishing

Review Paper

04.1

2014

Review Paper Challenges for the future of urban sanitation planning: critical analysis of John Kalbermatten’s inﬂuence Ruth Kennedy-Walker, Barbara Evans, Jaime Amezaga and Charlotte Paterson

ABSTRACT During the 1980s, John Kalbermatten and his colleagues at the World Bank revolutionised urban sanitation planning. During the last 30 years urban sanitation planning theory has evolved from an engineering focus to a more participatory, multi-disciplinary and user-focused future, informed largely by the work of John Kalbermatten. This paper looks at a number of the most important urban sanitation planning approaches that have emerged post-Kalbermatten and seeks to trace the inﬂuence of Kalbermatten’s work on their theoretical underpinnings and characteristics. The extent to which other ideas, such as the sanitation value chain, have increasingly been incorporated into planning approaches is discussed and some of the challenges affecting successful urban sanitation which lie

Ruth Kennedy-Walker (corresponding author) Jaime Amezaga Charlotte Paterson School of Civil Engineering and Geosciences, Newcastle University, Newcastle-upon-Tyne NE1 7RU, UK E-mail: r.kennedy-walker@ncl.ac.uk Barbara Evans School of Civil Engineering, University of Leeds, Leeds LS2 9JT, UK

outside of planning are considered. Final comments centre on common themes occurring in practice, the future exploration of which offers potential to inform successful sanitation delivery in the future. Key words

| enabling environment, participation, planning, sanitation, urban

INTRODUCTION In the late 1970s John Kalbermatten and colleagues at the World

paper looks to provide an overview of the main urban sanitation

Bank led a shift in the approach to planning and implementation

approaches developed in the last 30 years, to identify how John

of urban sanitation in less-developed countries. They were

Kalbermatten impacted the sector and establish if recent plan-

responding to the repeated failures of conventional sanitation

ning tools are achieving in practice what Kalbermatten ﬁrst set

solutions which were increasingly found to be inappropriate

out to do. It also explores how understanding those initial con-

for the contexts in which they were being implemented. Kalber-

cepts can guide the future of urban sanitation planning.

matten was concerned that this would have disastrous consequences for the planned International Drinking Water and Sanitation Decade running throughout the 1980s. The

THE NEW WORLD BANK PLANNING PARADIGM

new approach ﬁrst formulated in the World Bank publication, ‘A Planning and Design Manual’ addressed not only inadequa-

Before turning to more recent developments it is useful to con-

cies in the technology being recommended but also the

sider how urban sanitation was developing in the late 1970s

planning failures that had caused so many inappropriate sol-

and John Kalbermatten’s inﬂuence upon it. For industrialised

utions to be selected in the ﬁrst place (Kalbermatten et al.

countries, conventional sewerage (waterborne sewerage) had

a, b). Since then a large number of urban sanitation planning

long been the technology of choice for the disposal of human

approaches have been developed, each with unique ideas and

excreta (Kalbermatten et al. b). This preference was also

methodologies but mostly stemming from those original concep-

evident in less-developed countries, with conventional sewer-

tual foundations brought to the sector by Kalbermatten. This

age being considered by engineers and planners as the only

doi: 10.2166/washdev.2013.164

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

sanitation technology option for their cities (Mara ). In

us today – a recent study estimated that even among water uti-

reality, the high cost of installation, operation and mainten-

lities serving sub-Saharan Africa’s largest cities, only 50% offer

ance of conventional sewerage systems and the need for an

sanitation services and of those with sewer networks only 50%

in-house (on-site) water supply meant that conventional sewer-

of their service area has sewer coverage (Morella et al. ).

age proved to be an inappropriate option for many developing country cities that lacked the regular fund ﬂow to pay for

Kalbermatten’s big ideas

proper operations. For these reasons, it proved wholly inappropriate in rapidly growing low-income and unplanned

Kalbermatten and the World Bank proposed an alternative

urban communities which were often excluded from the plan-

model of sanitation planning (Figure 1). The model re-focused

ning and implementation process (Mara ). High

the attention of the engineers who were still largely leading

expectations for sewerage continued despite limited capacity,

planning efforts. The four underlying principles were:

inadequate ﬁnancing and weak institutions in most cities

1. To identify sanitation interventions that would provide

and towns. Given the high costs of the solutions being rec-

maximum health beneﬁts; Kalbermatten asserted that

ommended, investment was concentrated on capital and

conventional sewerage was unsuitable as its aim was to

major cities and often resulted in systems that were only par-

maximise convenience.

tially usable and rapidly fell into disrepair as funds dried up.

2. To consider the whole range of potential sanitation tech-

The result was decades of slow progress within the sanitation

nologies, selecting those that would provide as many

sector (Kalbermatten et al. b). Proof of this remains with

people as possible with the required facilities.

Figure 1

The World Bank model for sanitation programme planning (Kalbermatten et al. 1982b).

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

3. To move away from a top-down technology-centred

Supply and Sanitation along with numerous conferences

approach to planning and encourage the inclusion of

and declarations have resulted in sanitation becoming

additional professional disciplines.

more prominent in the global agenda for development.

4. To include the community in a more iterative planning

Consequently, over the last 30 years a number of sanitation

planning models have been developed which have shaped

interdisciplinary project team would more successfully

this sector. Figure 2 presents a timeline of events, which

interact with the community to identify a wider

have been instrumental in shaping the urban sanitation

range of technically feasible, economically and ﬁnan-

planning sector, and which illustrates the contempora-

cially

neous planning approaches.

process.

The

rationale

affordable,

and

which

was

socio-culturally

that

acceptable

sanitation options (Kalbermatten et al. b; Mara ).

Kalbermatten’s inﬂuence on evolving urban planning approaches

THE EVOLUTION OF APPROACHES

In the following sections urban sanitation planning approaches which have been inﬂuenced by Kalbermatten’s

The global landscape

concepts will be considered. We seek to give an overview of their implementation in practice and their ability in

Since the development of the World Bank model, events

achieving sanitation at scale based on the rationale set out

such as the International Decade for Drinking Water

by Kalbermatten.

Figure 2

Timeline of development of selected urban sanitation planning approaches and signiﬁcant events in the sanitation sector.

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Strategic Sanitation Approach/strategic sanitation planning (1989)

noted that for such an approach to work (as with any plan-

The Strategic Sanitation Approach (SSA) ﬁrst described in

incentive and demand-based focus (Colin et al. ;

1989 by the UNDP-World Bank ‘Water and Sanitation Pro-

Murray ; Peal et al. ).

ning approach) an ‘enabling environment’ needs to be created on the ground to speciﬁcally deal with such an

gram’ (WSP), was strongly inﬂuenced by Kalbermatten, who was responsible for establishing WSP (Black ). WSP developed the approach and used it to guide signiﬁcant World Bank-supported urban sanitation investment, pilot

Household centred environmental sanitation (HCES) (2000)

projects in Kumasi, Ghana and Ouagadougou, Burkina Faso. Since then it has formed the basis for a number of pro-

In 2000 the environmental sanitation working group of

jects in India, Indonesia, Thailand, Brazil and Pakistan (Peal

the Water Supply and Sanitation Collaborative Council

et al. ). Drawing on Kalbermatten’s ideas, the multi-disciplinary

(WSSCC) developed the so-called ‘Bellagio Principles’, a set of principles for good urban environmental sanitation

team codiﬁed a planning approach, which recognised that

(Eawag ). They state that human dignity, quality of

there was a pivotal point of action at the neighbourhood

life and environmental security should be at the centre

level. The key new idea was to respond to demand at the com-

of urban sanitation planning; decision-making should

munity level (an idea which drew strongly from recent

involve participation of all stakeholders; waste should be

developments in the rural water supply sector) where

considered as a resource and should form part of an inte-

demand would be demonstrated both by the participation of

grated water resources and waste management process;

communities in planning and management and by their will-

and that environmental sanitation problems should be

ingness to pay for elements of the system. The approach also

resolved at as low a level as possible (Peal et al. ). Kal-

considered incentives at each level, seeking to understand

bermatten was part of the working group and was key in

what motivated communities, local government and other

the conceptualisation of the Bellagio Principles; the under-

actors along the sanitation value chain. An outcome of that

lying principles of the World Bank model can clearly be

approach was the idea that sanitation services could be ‘unbundled’ – different solutions could be used in different

seen within these (Kalbermatten et al. ; WSSCC ).

parts of the city (horizontal unbundling) and different manage-

HCES was developed to operationalise the Bellagio

ment arrangements could be used along the value chain

Principles. It was conceived by the WSSCC working group

(vertical unbundling) (Tayler et al. ; Peal et al. ).

and further developed by the Swiss Federal Institute of

The SSA also speciﬁcally encouraged a consideration of sani-

Aquatic Science and Technology (EAWAG) and identiﬁed

tation across the entire sanitation value chain (i.e. including

that for any plan to be successfully implemented an

collection, transport and treatment of waste as well as house-

‘enabling environment’ needed to be established within

hold level services). In relation to the four underlying

which the hygiene or sanitation intervention operates

principles of the World Bank model, SSA reiterates the impor-

(Peal et al. ). To achieve an ‘enabling environment’ cer-

tance of household level participation, the need for an

tain requisites were to be met.

inclusion of a multi-disciplinary planning team whilst introducing the idea that different technical solutions and services can be used for different situations/environments within one city. While SSA worked well in Kumasi and Ouagadougou where there was signiﬁcant technical and ﬁnancial support, it presented challenges in cities with less planning capacity (WSP ; Vezina ; Colin et al. ). Reports have

• • •

An adequate level of government support for the project in terms of political support and favourable national policies and strategies. A legal framework, with appropriate standards and codes at national and municipal levels. Institutional arrangements that suit and support the approach of the project.

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

•

Effective training and communication ensuring that all

•

Credit and other ﬁnancial arrangements that facilitate

In 2007 the International Water Association (IWA)

the required level of participation and community

attempted to ‘take stock’ of the state of knowledge around

•

involvement.

urban sanitation, especially planning, and interpret this

Information and knowledge management providing

for the use of professional engineers working in less-devel-

access to relevant information sharing experiences, train-

oped countries. The resultant framework, known as

ing and resource materials, the development of new

Sanitation 21, encourages technical professionals to think

approaches and the dissemination of ﬁndings (Eawag

beyond ‘business as usual’ by reiterating key ideas from

).

models such as those outlined by Kalbermatten, SSA and

Sanitation 21 framework (2007)

participants understand and accept the concepts.

the HCES approach (IWA ). The framework deﬁnes This ‘enabling environment’ framework goes beyond the

‘domains’ within which sanitation exists (from household,

multi-disciplinary approach in Kalbermatten’s original

via neighbourhood and ward, to the wider city and

model and recognised that the entire institutional context

beyond). These domains are deﬁned by different social

inﬂuences whether appropriate planning can be achieved.

and political norms and structures and provide a frame-

This was highlighted in the experiences of WSP with SSA

work, within which the approach can identify aspects

pilots in India. The HCES approach also formalised the

such as stakeholder interests, stakeholder capacities, exter-

value placed on the ecological effects of sanitation by the

nal factors and existing systems and their functionality

Bellagio Principles; particularly the idea of resources from

and success (IWA ). These domains can then map

sanitation being used as close to the point of production

fairly accurately onto the technical elements of the sani-

as possible and the link to integrated water resources plan-

tation

ning – both of which were important ideas in the wider

disposal, reuse etc.). This allows for a more realistic assess-

water sector of the time. This focus on ecological concepts

ment of the feasibility of a range of sanitation solutions by

may also have encouraged a consideration of technologies

considering whether management capacity to operate it

with stronger ‘ecological’ credentials when compared to

exists in the places where it is needed. Solutions to local

conventional sewerage and pit latrines – although this is

problems are thus linked to feasible systems of collection,

not particularly evident from case-study literature. Like the

transport and disposal/reuse of waste (IWA ; Murray

value

chain

(collection,

transport,

treatment,

World Bank approach, HCES aims to respond to the

; Peal et al. ). Relating back to those four concepts

users’ needs and demands by ensuring they are placed at

deﬁned by Kalbermatten, this approach looks to go beyond

the core of the planning and implementation process (Peal

the engineer by ensuring a wide range of stakeholders

et al. ).

(including households) are included within the process.

This approach has been extensively implemented in a

One could argue that the focus on inﬂuencing professional

number of locations. Evaluative literature is limited; how-

engineers may make Sanitation 21 less accessible for non-

ever, reporting from a number of pilot studies highlights

technical stakeholders. Another interpretation is that IWA

that the presence of an ‘enabling environment’ was criti-

considered that the professional engineers were the ones

cal to success. In particular, capacity and access to

who had most to gain from a deeper understanding of the

requisite professional skills was noted as a requirement

non-technical, institutional aspects of effective sanitation

alongside local knowledge of existing sanitation solutions

service delivery. In terms of technology selection Sanitation

(particularly

insti-

21 once again highlights the importance of understanding

tutional arrangements, government/authority skills and

the entire sanitation value chain and opens the door to

support, a suitable legal framework and access to the

technologies

necessary ﬁnancial arrangements (Peal et al. ;

approach seems to focus less on health speciﬁcally but

Rohrer ).

instead on how effective and efﬁcient the chosen

non-conventional

ones),

enabling

which

optimise

ecological

value.

This

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

technology will be within the deﬁned environment. There is

Cities in India programme (RAY) (Eawag-Sandec ;

currently no documented evidence of this approach having

Eawag ).

been tested on the ground so it is difﬁcult to establish its success in implementation.

Other urban sanitation planning approaches

Community-led urban environmental sanitation (CLUES) (2011)

Conceptual links and parallel traditions While it is possible to draw a direct conceptual link between

The implementation of the HCES approach highlighted the

Kalbermatten’s original model and the subsequent develop-

importance and the challenge in achieving community par-

ment of SSA, HCES, Sanitation 21 and CLUES there are

ticipation (including the household level and beyond) in the

other urban sanitation planning approaches which have

planning and decision-making processes and prompted the

emerged from parallel traditions or developments. Despite

development of the new hybrid planning framework,

their alternative provenance many show conceptual consist-

CLUES (Lüthi et al. ). CLUES provides a seven-step

ency with some of Kalbermatten’s principles, and are

approach to planning for environmental sanitation (water

discussed below.

supply, sanitation, solid waste management and storm drainage),

which

emphasises

the

importance

broad

GTZ ecosan approach (2003)

community involvement as well as encouraging a multi sector and multi actor approach (Lüthi et al. ). As with

In 2003, the German Agency for Technical Cooperation

the HCES approach, CLUES calls for an ‘enabling environ-

(Deutsche Gesellschaft für Technische Zusammenarbeit),

ment’ to be established that provides the required

GTZ (now known as GIZ), developed a set of tools to

conditions for sustainable environmental sanitation inter-

encourage the use of ecological sanitation solutions. It is

vention (identical to HCES approach). This approach

linked to the Bellagio principles but it strongly places eco-

addresses some of the pitfalls seen in the earlier HCES

logical considerations at the heart of any sanitation

and identiﬁes the importance of the processes of awareness

intervention, with other objectives being secondary. Conse-

raising and communication, capacity development through-

quently, a ‘toolbox’ was developed to provide planning

out the planning process and also returns to the theme of

guidelines for so-called ‘ecosan’ technologies (Werner

SSA by highlighting the importance of monitoring and

et al. ). The toolbox emphasises that ‘ecological sani-

evaluation, ensuring accountability and tracking success of

tation’ is not synonymous with a particular technology but

the intervention throughout. This approach expands upon

rather an idea that encourages recycling-oriented resource

some of Kalbermatten’s original concepts, especially the

management (UNESCO and GTZ ). However, many

importance of household-level inclusion in the planning

observers conﬂate the use of the term ‘ecosan’ to the speciﬁc

process. It further develops the idea of the ‘enabling

use of urine diverting dry toilets.

environment’ and refers to the need to include expertise

The approach incorporates a ten-step model, adapted

from different sectors and roles. Building on Bellagio it

from the HCES model containing the stringent requirement

highlights the importance of viewing waste as a resource

to recognise human excreta and water as a resource to be

and as integral to a sustainable solution. This is a very

exploited rather than a waste (Werner et al. ). It also

new approach with little evidence of its successful

acknowledges the need for an ‘enabling environment’ to

implementation on the ground. Notwithstanding this the

be in place but also highlights how elements of the environ-

Centre for Urban and Regional Excellence, in partnership

ment may need to be reﬁned to incorporate the ‘ecosan’

with Eawag-Sandec, with the ﬁnancial support of GIZ,

philosophy. This approach encourages a move away from

has utilised this approach for preparing slum upgrading

conventional technology options to consider the use of a

plans (which include Ward Strategy Papers and Detailed

variety of technologies for the whole sanitation value

Project Reports) in Raipur, India, as part of the Slum Free

chain. Although there are a number of schemes which

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

have used this approach there is little evaluative data avail-

sections. Frameworks and in-depth guidelines for CSPs

able (UNESCO and GTZ ). Observations indicate that

have been developed by a number of supporting organis-

elements such as awareness raising and planning for reuse

ation in a variety of cities. In India a number of

are more demanding as ‘ecosan’ is still a fairly unknown

organisations have supported the National Urban Sanitation

concept in many places (Panse et al. ).

Policy for India. These include the WSP and Centre for Environmental

Design for service approach (2009)

Planning

and

Technology

University

(CEPT), GIZ, Bremen Overseas Research and Development Association (BORDA) and CDD (Consortium for DEWATS

Recently the ‘ecological’ view of sanitation has prompted a

Dissemination Society); and the ICLEI – Local Govern-

serious reconsideration of the products of sanitation (speciﬁ-

ments for Sustainability (Government of India ; CEPT

cally nutrients and water). Design for Service (DfS), a ﬁve-

; WSP ; BORDA ; GIZ ). WSP have also

step planning approach developed by Ashley Murray as

supported sanitation planning in Indonesia (WSP )

part of her doctorate (Murray ), presents a radical

and beyond Asia, PS-Eua have supported local authority-

change of approach and highlights the importance of identi-

led planning in various cities in West Africa (Eau )

fying

with

and a number of experiences can be seen from Brazil

stakeholders at the downstream (reuse) elements of the sani-

(Aroeira et al. ; Wartchow & Daronco ). Shortcom-

tation value chain as a starting point. This planning

ings have, however, been identiﬁed with the CSP approach

approach emphasises the importance of health but also

and these, once again, primarily relate to the funding chal-

highlights the importance of the end-use functionality of

lenges ﬁrst identiﬁed by Kalbermatten and the capacity/

the sanitation system to ensure the success of any system

enabling environment gaps ﬁrst identiﬁed in SSA (Govern-

implemented. This may result in a reduction of conventional

ment of India ; WSP ).

sanitation

solutions

and

participating

sanitation solutions as those solutions which provide the best downstream solution (i.e. for reuse) will be prioritised. The model has withstood some initial testing during its

EMERGING CHARACTERISTICS AND CONCEPTS

development in China and Ghana; however, further evaluation of its implementation and usefulness is required.

Linear and parallel developments

City Sanitation Plans

Over the last 30 years a succession of multi-disciplinary teams have produced a series of credible planning frame-

City Sanitation Plans (CSPs) are a recent development in

works

urban planning departments in a number of countries.

governments motivated to prepare serious urban sanitation

Taking a holistic approach to city planning enables CSPs

plans. Within this review, approaches to urban sanitation

to be embedded in city budgets and to relate constructively

planning can be seen to be broadly linear (with a few diver-

to other service provisions thereby addressing many of the

sions along the way) and there is an encouraging

implementation challenges faced by sanitation planners

consistency throughout indicating that the underlying plan-

who would otherwise be working with technical depart-

ning process is well understood and will continue to be

ments alone. In a number of developing countries

relevant into the future. There are other concepts that

production of these plans by local government has been

have emerged from within the sanitation sector and wider

linked to ﬁnancial incentives with the preparation of CSPs

developmental arena, which have had varying levels of

being required by state or central government. These plans

impact on the approaches taken to urban sanitation plan-

take both technical and non-technical aspects associated

ning. The following section brieﬂy explores some of those

with delivering sanitation at citywide level into consider-

concepts, to assess how they relate to the World Bank’s

ation and many draw upon the fundamentals of the

paradigm and identify how they are being implemented in

planning models and approaches identiﬁed in the earlier

practice.

which

could

usefully

deployed

local

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Focus on health

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

institutional challenges of urban sanitation which has to function at both the private household level and the public

The World Bank model proposed that any technology inter-

network level. SSA made this dimension of urban sanitation

vention should be implemented to maximise health beneﬁts.

more explicit through the introduction of institutional and

This analysis has highlighted that some of the approaches

technical unbundling along the value chain.

developed have moved away from focusing on health,

Despite the perceived beneﬁts of viewing waste as a

reverting instead to a focus on technological functionality,

resource, there is little evidence that cities are moving

particularly ecological functionality. There is no evidence

towards viewing sanitation as a resource-generating sector.

currently available to suggest that demoting health improves

There is little evidence that any urban sanitation planning

sanitation service delivery. For approaches such as the GTZ

approaches have successfully stimulated reuse of the pro-

Ecosan Approach the focus on ecological functionality may

ducts

result in prescribed technologies being promoted at the

surprising since it is not holistically incorporated into all

expense of others that may offer greater health beneﬁts. In

stages of any of the planning processes discussed (Murray

practice the enforcement of ecologically based technologies

). It also suggests a genuine challenge for the sector,

in urban areas has been shown to be difﬁcult because of the

namely that those people who currently control sanitation

complexities of the environment itself and the requirements

investments themselves do not value the resources of sani-

needed for such a system to function properly (e.g. enabling

tation. There are numerous technical and cultural reasons

environment). On a more general level, if the connection

for this with lack of knowledge and capacity playing a

between improving sanitation conditions and health in

part. In countries with high capacity and severe resources

urban areas is taken as a given, a shift away from health

constraints a much more progressive approach has been evi-

treated

domestic

wastewater.

This

not

objectives may not be critical, provided that there remains

dent for many years (Kfouri et al. ). For such

a focus on improving access to services which work for as

approaches to become more widespread, knowledge about

many people as possible. Perhaps here, Kalbermatten’s

appropriate treatment and post-treatment interventions are

main contribution was to prompt a consideration of objec-

needed. To achieve this in practice a stronger focus on the

tives in the ﬁrst place, which had rarely been the case up

downstream elements of the value chain (similar to that pre-

to that point.

sented by the DfS tool) would be needed although gaining acceptance of this idea at community and city level remains

Sanitation value chain

challenging.

Since Bellagio, the idea of sanitation as a resource has been

Sanitation ladder

widely acknowledged and has become a key concept in urban sanitation. To be successful, it has to link collection

The ‘sanitation ladder’ is a term widely used to describe a

of wastes (at the household level) via collection, transport

stepwise process by which communities or households

and treatment to ultimate reuse or disposal of by-products.

may progressively experience improved sanitation. The

The early World Bank teams had a solid understanding of

idea recognises that sanitation imparts beneﬁts of varying

the technical ‘sanitation value chain’; however, this under-

magnitude and differing nature depending on both the

standing was so strongly embedded in the conventional

type of facility available to the user and the extent to

approaches to sanitation that Kalbermatten challenged

which waste is subsequently well managed in the value

that it was never explicitly referred to in the World Bank

chain. Often the focus of sanitation ladder analysis is on

approach.

the household experience. Thus, for example, since 2008,

The term ‘sanitation value chain’ has uncertain prove-

the UNICEF/WHO Joint Monitoring Programme on

nance but has been used increasingly in recent years by

Water and Sanitation (JMP) has reported global access to

organisations including the Bill & Melinda Gates Foun-

sanitation using a step scale from open defecation, via unim-

dation.

proved facilities to improved facilities, where ‘improved’ is a

neatly

illustrates

the

real

technical

and

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

technology-based indicator used as a proxy for sanitation

sanitation interventions (Peal et al. ). The concept of

which is more likely to deliver health beneﬁts (WHO/

the ‘enabling environment’ is far broader than the need for

UNICEF JMP ). Many commentators feel that access

inclusion of an interdisciplinary project team ﬁrst noted by

to ‘improved sanitation’ is a poor indicator towards progress

Kalbermatten, but the latter is clearly predicated on the

(Shordt et al. ; Sutton ). Others note that the report-

former. Unfortunately, a common theme throughout the lit-

ing in JMP creates incentives for countries to take a

erature of urban sanitation planning, and in particular the

technology-based approach to regulation and policy which

small canon of case studies and evaluations, is the almost

can hamper innovation (Kvarnström et al. ). This in

universal failure or absence of the required ‘enabling

turn reduces investment in the sanitation value chain as a

environment’. CLUES practitioners try to address this pro-

whole.

blem in part by highlighting that it is not only vital to

To address these concerns Kvarnström et al. () devel-

ensure that the correct stakeholders and sectors are

oped the ‘function approach’ ladder which moves away

included in the planning process but that those individuals

from describing pre-deﬁned technologies and focuses on

and institutional bodies are aware of the importance of sani-

assessing the outcomes or effects of any given sanitation

tation, have the capacity to deal with planning for and

system. This approach assesses how excreta are managed

implementing sanitation interventions, that knowledge and

throughout the whole sanitation value chain rather than

understanding can be transferred between people and they

just at the collection point and a resource-orientated focus

can monitor and be held accountable for failings in provid-

is integral to the ladder. A clear focus of this approach is

ing acceptable outcomes. Nonetheless it is clear that in most

to put the health functions of the sanitation system at the

cases the absence of the appropriate institutions and

earlier rungs of the ladder which, once achieved, then

capacities severely constrains both willingness to prioritise

focuses on the environmental functioning of the system.

sanitation in general and sanitation planning in particular,

This is in line with Kalbermatten’s focus where health

and the ability to handle the necessarily complex process

should be of primary importance. The ladder also highlights

of planning once it begins. Once low-income and informal

that sanitation provision is often a dynamic process where

settlements are included these failures only appear to

incentives may change as progress is made. In this sense it

become more marked.

also brings forward the idea that the enabling environment can develop progressively as the ambition of sanitation inter-

Household participation

ventions grows over time; that is, higher or later rungs on the ladder have higher costs and management and logistical

Household participation has become integral to all urban

requirements associated with them.

sanitation planning approaches. Participation has the potential to overcome lack of effective demand for sanitation on

Enabling environment

the ground and to help develop long-term project sustainability.

Ensuring

upstream

users

(households

Since Kalbermatten, an almost universal theme within

communities) are included in the planning process helps

urban sanitation planning has been the need for a conducive

develop a sense of ‘ownership’ (Mara ). This post-Kal-

‘enabling environment’. This is said to deﬁne aspects of the

bermatten shift in promoting the use of participatory

political, economic, educational, socio-cultural, organis-

approaches has not only been seen in the sanitation sector

ational, technological and legal framework (or sometimes,

but also in water, health and hygiene. However, few studies

captured in the term ‘institutional’ in its broadest sense)

have been completed which show how participation has

within which the sanitation intervention operates (Peal

been undertaken or which explore the relationship between

et al. ). The concept indicates what needs to be in

participation and achieving long-term project success. Over-

place for planning to be successful in practice (Eawag

all, studies which are available conclude that participation is

). Another key factor sometimes noted is how the built

often undertaken with a ‘tick box’ approach and that pre-

environment can impact on the potential and outcome of

deﬁned objectives and expert-led solutions are actually

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

implemented with little understanding of what users really

population growth and coverage rates are actually falling

want (Jones ; Nance & Ortolano ; McConville

(WHO/UNICEF JMP ). In urban areas improved plan-

).

ning is likely to be a part of the solution although not the

Commentators note that for participation to truly work

entire solution. What is perhaps most striking about urban

it must be deeply institutionalised in order for both the pro-

sanitation over the past 30 years is the lack of evaluation

cess to be to be properly facilitated and for the ‘state’ to be

of implementation experiences of approaches discussed in

responsive to the demands of the community. Evidence

this paper. This is not surprising as in reality, sanitation in

shows that those interventions which work best do so

urban areas is said to be delivered in an ad hoc fashion, if

because of their ability to be sensitive and adaptable to vari-

at all, and few cities identify it as an investment priority or

ations in context (Reed ; Mansuri & Rao ). It is

are prepared to invest time and resources in planning for

noted that the institutional structure within urban sector

efﬁcient and effective service delivery (Tayler & Parkinson

institutions could have an impact on how successful partici-

). Even where sanitation planning is undertaken, experi-

pation/demand-driven approaches are as they are typically

ence suggests that capacity and skills gaps persist.

set up with a supply orientated focus and therefore may

Numerous commentators have noted how lack of knowl-

not be adequately staffed or trained to undertake partici-

edge of new developments results in the propagation of

pation in reality (Cotton & Saywell ). Literature also

old-fashioned approaches and solutions which do not meet

suggests that participation should emphasise iterative and

the needs of people (Nance & Ortolano ; McConville

two-way learning between participants and stakeholders

; Lüthi & Kraemer ). In particular ‘participation’

from very different knowledge and perspective backgrounds

does not appear to be yielding the results expected in

(Reed ) but in reality this cyclical process is rarely seen;

terms of improved, better tailored and effective local sol-

there is usually limited honest informative feedback that

utions. Wright () identiﬁes that the challenge for

helps to facilitate learning between the inner and outer cir-

governments and donor agencies is to motivate and build

cles of stakeholder groups (Mansuri & Rao ). This

the capacity of the different stakeholders to participate in

disconnect may be due to the nature of institutions and

appropriate and productive ways. This coincides with

the incentives that drive individual action or it may be

others who note that adequate sanitation knowledge is

more closely related to issues of trust (Wright ). Once

required at the local level to achieve universal sanitation

again it is the ‘enabling environment’ that appears to be criti-

access (Mara ). Large capacity deﬁcits exist at all

cal, since participation needs to be underpinned by ‘a

levels in key water and sanitation agencies in most low

philosophy that emphasises empowerment, equity, trust

income countries caused by adverse institutional structures

and learning for it to be successful’ (Reed ). This takes

and systems of incentives as well as insufﬁcient funds

two forms by ensuring the participants have the power to

(Cavill & Saywell ; DFID et al. ).

inﬂuence the decision and by ensuring participants have the technical capability to engage effectively with the

Learning from the past

decision (Reason & Bradbury ). There is a lack of case-study evidence regarding the implementation of urban sanitation planning approaches

CHALLENGES FOR THE FUTURE

and where there is evidence this is mainly based on shortrun reporting rather than ongoing monitoring or repeat

Planning in practice

evaluations of success. The SSA approach, for example, was identiﬁed as a success based on several case studies

Although sanitation has become more prominent on the

which were published during the planning phase and shortly

global agenda, progress has not been made at the required

after but since then little continuous monitoring and report-

scale and speed. In urban and peripheral urban areas in par-

ing of its ongoing success has taken place. This lack of long-

ticular progress often fails to keep up with the pace of

term monitoring creates gaps in knowledge about the real

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

impact of interventions and reduces potential learning for

commitment to planning and service delivery (Evans ;

the future as most evaluations and reporting take place

Tayler & Parkinson ). However there is an inherent pro-

immediately after the project is implemented (FAO ;

blem in those public institutions who are mandated to

Jones et al. ; Mansuri & Rao ). Few urban sanitation

deliver such services as they generally appear to have low

planning approaches place much emphasis on accountabil-

capacity and to be severely under-resourced (Evans ;

ity – which would require both an explicit deﬁnition of

Cairncross et al. ). They also tend to lack a planning cul-

outcomes and the development of associated monitoring

ture being more commonly focused on addressing crises in

and evaluation processes. Accountability is implicitly

an ad hoc and non-systematic way (Tayler & Parkinson

assumed to arise through processes of participation but

). Their ability to plan for and engage with communities

there is no evidence that this actually happens in practice.

and households in order to understand and inﬂuence house-

Knowledge gain through experience seems to be implicit

hold behaviours and the role of community action as a

in the successive and cumulative development of increas-

means to creating an ‘enabling environment’, thereby

ingly sophisticated planning approaches outlined here, but

achieving increased demand for sanitation, is also usually

there is almost no record of the basis upon which those

weak (Evans ). Finally, these institutions are inherently

developments were made. Conclusive evidence regarding

unable to hold themselves accountable through the collec-

the relative importance of the various planning principles

tion of credible evidence for monitoring purposes and

underpinning these approaches could potentially be gener-

evaluation of their progress (ibid.).

ated if case studies could be revisited; the cohort of welldocumented planning approaches described here provides

Going forward

a potentially fascinating basis for a historical review of the impact of planning on sanitation service delivery. For

It is evident that the challenges of delivering urban sani-

future interventions, greater attention to long-term monitor-

tation go beyond the need for better planning. The

ing would also be highly valuable and enable lessons to be

institutional constraints that hold back planning and invest-

learned and shared more openly.

ment in such an essential service generally constrain all aspects of urban governance; provision of most critical ser-

Inherent problems for urban sanitation planning

vices, from housing to education, remains ad hoc and chaotic in many rapidly growing poor cities. Nonetheless,

Across all the approaches covered in this paper there

sanitation can be seen as a touchstone for urban govern-

appears to be recognition of some common constraints to

ance; a city which can provide its citizens with a

effective sanitation planning and associated sanitation

functioning, articulated urban sanitation system is well

investments. Lack of political will is cited on numerous

placed to deliver much more. But similarly, the delivery of

occasions, evidenced by the low priority given to sanitation

urban sanitation cannot surmount structural failings in the

via government policies and budgets (Tayler & Parkinson

city at large; a rational sanitation plan is no match for politi-

; Cairncross et al. ). Although more market-based

cally motivated land developers intent on withholding basic

and participatory planning models can achieve some trac-

services from unplanned settlements. Perhaps the critical

tion at the local level, the physical nature of the urban

point here is this: just as Kalbermatten called for an iterative

environment and the need to manage some aspects of sani-

planning process based on understanding of what is on the

tation

(and

ground already, sanitation planners need to invest more

successful participation) will always be needed to ensure

time in understanding the nature of the problem to be

that the entire sanitation value chain functions. Local

solved and the capacity of the existing systems to address

demand for improved environmental conditions will rarely

those problems. We may wring our hands at the failure of

be sufﬁcient to support the costs and institutional challenges

the enabling environment, but perhaps we could achieve

of coordinated sanitation in the urban space. Thus urban

more by working with what exists and doing at least part

sanitation

of the job in the right way and in the short term. The

collectively,

always

means

requires

that

public

explicit

support

institutional

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

recent focus on the sanitation value chain and ecological

environment’ is more sophisticated than the call for multi-

objectives tends to push decision makers towards achieving

disciplinarity. As identiﬁed by Kalbermatten, household par-

the perfect complete system in one leap, but the functional

ticipation is still inherent to every planning approach,

sanitation ladder should remind us that even sanitation

despite the lack of evidence about how best to do it, or

system development can be progressive, with progressive

indeed, the relationship between participation and long-

marginal gains keeping step with progressively strengthen-

term success of the approaches on the ground. However,

ing institutional capacity.

beyond this, we would argue that real progress in the sector cannot occur without better evidence of what really works. We need a commitment to better long-term monitor-

LIMITATIONS

ing and evaluation of the effects of urban sanitation planning and its connection to investment and improved

This paper is a partial and biased consideration of progress

service delivery. If, in the process, we can also contribute

in urban sanitation. There are of course other sanitation

to building a stronger enabling environment, greater

planning approaches and perspectives to be seen in the lit-

capacity, more effective participation and more accountabil-

erature but the authors have attempted to bring focus to

ity this will all be to the good; John Kalbermatten would

this analysis by taking as a starting point the four principles

have asked for nothing less.

articulated by the World Bank team in the 1970s. The very limited empirical data mean that such a review must be highly speculative. Furthermore, the very wide range of con-

ACKNOWLEDGEMENTS

texts in which we seek to address the urban sanitation challenge mean that the conclusions drawn here are gener-

This PhD research is funded by the UK Engineering and

alised rather than speciﬁc to any given case. Despite these

Physical Research Council through the project, ‘A global

limitations we feel that it is possible to trace the inﬂuence

solution to protect water by Transforming Waste’ (EP/

and linkages of successive attempts to articulate effective

J00538X/1).

urban sanitation planning tools and to use this as a pointer towards more effective interventions in the future.

REFERENCES CONCLUSION This study has established how John Kalbermatten and the World Bank model impacted upon urban sanitation planning and how subsequent planning approaches ‘evolved’. The paper sought to demonstrate conceptual links and tensions between the differing perspectives of optimising health gains, increasing the repertoire of potential technical solutions, multi-disciplinarity, the sanitation value chain, the functional sanitation ladder, the enabling environment, and participation. The trajectory of change is complex; ﬁrstly, the focus on health has increasingly been challenged by a move towards a focus on achieving ecological outputs within sanitation and the need to holistically achieve access along the whole sanitation value chain. The introduction of ‘enabling

Aroeira, R. d. M., Braga, R. A., Pereira, M. D., Aguiar, I. M. C., Virgínia, C. & Alves, P.  The Municipal Sanitation Plan for the City of Belo Horizonte: Le plan d’assainissement de la ville de Belo Horizonte. [Online]. Available at: http:// documents.irevues.inist.fr/bitstream/handle/2042/35641/ 13106-043DEM.pdf (Accessed: 20/09/2013). Black, M.  1978–1998. Learning What Works: A 20 Year Retrospective View on International Water and Sanitation Cooperation. UNDP-World Bank Water and Sanitation Program, Washington, DC. BORDA  Implementation of decentralised sanitation approaches through pilot interventions in Indian cities of Raipur, Chhattisgarh State and Kochi, Kerala State. Available at: http://www.borda-sa.org/news/newsdetails-south-asia/ article/implementation-of-decentralised-sanitationapproaches-through-pilot-interventions-in-indian-cities-o. html (Accessed: 20/07/2013). Cairncross, S., Bartram, J., Cumming, O. & Brocklehurst, C.  Hygiene, sanitation, and water: what needs to be done? PLoS Med 7 (11), e1000365.

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Cavill, S. & Saywell, D.  The capacity gap in the water and sanitation sector. 34th WEDC International Conference. WEDC, Addis Ababa, Ethiopia. CEPT  Performance Measurement Framework for Urban Water and Sanitation Vols I and II. Ahmedabad, India. Colin, J., Ceetelaar, C., Utomo, T. & Blackett, I. C.  Urban Sanitation in Indonesia: Planning for Progress. Field note. Water and Sanitation Program - East Asia and the Paciﬁc Jakarta, Indonesia. Cotton, A. & Saywell, D.  Strategic Sanitation Approach: A review of literature. Operational Implications. Water, Engineering and Development Centre (WEDC): DFID KAR Project No. R6875. Available at: http://www.lboro.ac.uk/ garnet/ssacover.html (Accessed: 23/09/2013). DFID, IWA and UKAid  World Water Week. Stockholm. Available at: http://www.worldwaterweek.org/documents/ WWW_PDF/2010/monday/K24/sue_mind_gap.pdf. Eau, P. S.  Stratégies municipales concertées. Available at: www.pseau.org/smc (Accessed: 22/07/2013). Eawag  Household Centred Environmental Sanitation: Implementing the Bellagio Principles in Urban Environmental Sanitation. Switzerland. Eawag-Sandec  Validation of the Community-led Urban Environmental Sanitation Planning (CLUES). Switzerland. Eawag  Beratung der indischen Regierung zur Umsetzung der National Urban Sanitation Strategy. Final Report, Switzerland. Evans, B.  Securing Sanitation: The Compelling Case to Address the Crisis. International Water Institute, Stockholm, Sweden. FAO  The use of Monitoring and Evaluation in Agriculture and Rural Development Projects: Findings from a Review of Implementation Completion Reports. Rome, Italy. GIZ  Support to the National Urban Sanitation Policy (NUSP). Germany. [Online]. Available at: http://www.giz.de/en/ downloads/giz2012-national-urban-sanitation-policy-indiaen.pdf. Government of India  National Urban Sanitation Policy. New Delhi, India. IWA  Sanitation 21: Simple Approaches To Complex Sanitation: A Draft Framework For Analysis. International Water Association, London. Jones, P.  Urban regeneration’s poisoned chalice: is there an impasse in (community) participation-based policy? Urban Studies 40 (3), 581. Jones, S., Greene, N., Hueso, A., Sharp, H. & Kennedy-Walker, R.,  Learning from Failure: Lessons for the Sanitation Sector. Discussion paper. UK Sanitation Community of practice, London. Available online at http://www.bpwds.org/web/d/ DOC_360.pdf?statsHandlerDone=1. Kalbermatten, J. M., Julius, D. S. & Gunnerson, C. G. a Appropriate Sanitation Alternatives: A Technical and Economic Appraisal. International Bank for Reconstruction and Development/The World Bank, Johns Hopkins University Press, Baltimore.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Kalbermatten, J. M., Julius, D. S., Gunnerson, C. G. & Mara, D. D. b Appropriate Sanitation Alternatives: A Planning and Design Manual. International Bank for Reconstruction and Development/The World Bank, Johns Hopkins University Press, Baltimore. Kalbermatten, J. M., Middleton, R. & Schertenleib, R.  Household-Centred Environmental Sanitation. Swiss Federal Institute for Environmental Science and Technology, Switzerland. Kfouri, C., Mantovani, P. & Jeuland, M.  Water reuse in the MNA Region: constraints, experiences, and policy recommendations. In: Water in the Arab World: Management Perspectives and Innovations (V. J. Jaganathan, A. S. Mohamed & A. Kremer, eds.). World Bank, Middle East and North Africa Region, Washington, DC, pp. 447–477. Kvarnström, E., McConville, J., Bracken, P., Johansson, M. & Fogde, M.  The sanitation ladder – a need for a revamp? Journal of Water, Sanitation and Hygiene for Development 1 (1), 3. Lüthi, C. & Kraemer, S.  User perceptions of participatory planning in urban environmental sanitation. Journal of Water, Sanitation and Hygiene for Development 2 (3), 157–167. Lüthi, C., Morel, A., Tilley, E. & Ulrich, L.  Community- Led Urban Environmental Sanitation Planning: CLUES. Mansuri, G. & Rao, V.  Localizing Development: Does Participation Work? World Bank, Washington, DC. Mara, D.D.  Low Cost Urban Sanitation. John Wiley, New York. Mara, D. D.  A New Paradigm for Water Supply and Sanitation in Cities, Towns and Large Villages in Developing Countries. UN-HABITAT. Mara, D.  Pits, pipes, ponds - And me. Water Research 47 (7), 2105–2117. McConville, J. R.  Unpacking Sanitation Planning: Comparing Theory and Practice. Chalmers University of Technology, Gothenburg, Sweden. Morella, E., Foster, V. & Banerjee, S. G.  Climbing the Ladder: The State of Sanitation in Sub-Saharan Africa. The International Bank for Reconstruction and Development/ The World Bank, Washington, DC. Murray, A. E.  Don’t Think of ‘Waste’water: Evaluation and Planning Tools for Reuse-Oriented Sanitation Infrastructure. University of California, Berkeley. Nance, E. & Ortolano, L.  Community participation in urban sanitation: experiences in Northeastern Brazil. Journal of Planning Education and Research 26 (3), 284–300. Panse, D., Heeb, J. & Satish, S.  UNESCO-IHP Symposium Paris, 12–14 September 2007. Peal, A., Evans, B. & Voorden, C. v. d.  Hygiene and Sanitation Software: An Overview of Approaches. Water Supply & Sanitation Collaborative Council, Geneva, Switzerland. Reason, P. & Bradbury, H.  The Sage Handbook of Action Research: Participative Inquiry and Practice. SAGE publications, London.

R. Kennedy-Walker et al.

John Kalbermatten’s inﬂuence on urban sanitation planning

Reed, M. S.  Stakeholder participation for environmental management: a literature review. Biological Conservation 141 (10), 2417–2431. Rohrer, T.  106th Aguasan meeting, Bem, 9 September 2010. HELVETAS and Swiss Association for International Cooperation, Béma, Burkina Faso. Shordt, K., Wijk, C. v. & Brikké, F.  Monitoring Millennium Development Goal for Water and Sanitation: A Review of Experiences and Challenges. IRC International Water and Sanitation Centre, The Hague. Sutton, S.  33rd WEDC International Conference, Accra, Ghana. WEDC, Loughborough University, UK. Tayler, K. & Parkinson, J.  Strategic planning for urban sanitation: a 21st century development priority? Water Policy 7 (6), 569–580. Tayler, K., Colin, J. & Parkinson, J.  Strategic Planning for Municpal Sanitation: A guide. GHK Research and Training, London. UNESCO and GTZ  Capacity building for ecological sanitation: concepts for ecologically sustainable sanitation in formal and continuing education. Vezina, M.  The ouagadougou strategic sanitation plan: an holistic approach to a city’s problems. Wartchow, D. & Daronco, G. C.  Santa Rosa leads the way on Brazilian municipal sanitation plans. Water 21, 15.1, 43–44.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Werner, C., Panesar, A., Bracken, P., Mang, H. P., Huba-Mang, E., Gerold, A. M., Demsat, S. & Eiche, I.  An Ecosan Source Book: For the Preparation and Implementation of Ecological Sanitation Projects. Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ) GmbH, Germany. WHO/UNICEF JMP  Progress on drinking water and sanitation: 2012 update MDG assessment report. World Health Organization and UNICEFs Joint Monitoring Programme. Wright, A. M.  Toward a Strategic Sanitation Approach: Improving the Sustainability of Urban Sanitation in Developing Countries. UNDP-World Bank Water and Sanitation Program, Washington, DC. WSP  Urban Environmental Sanitation Planning: Lessons from Bharatpur, Rajasthan, India Water and Sanitation Program. South Asia, New Delhi. [Online]. Available at: http://www.wsp.org/sites/wsp.org/ﬁles/publications/ sa_bharatpur.pdf. WSP  Marching together with a citywide sanitation strategy. In: Water and Sanitation Program; Technical Team for Sanitation Development (Tim Teknis Pembangunan Sanitasi – TTPS) & the Indonesia Sanitation Sector Development Program (ISSDP). WSSCC, S.  Report of the Working Group on Environmental Sanitation to the 5th Global Forum of the WSSCC in Iguaçu, Iguaçu.

First received 22 April 2013; accepted in revised form 8 August 2013. Available online 22 October 2013

Review Paper

04.1

2014

Review Paper The uneven progress of sanitation in India Arabinda Ghosh and Sandy Cairncross

ABSTRACT The 2001 and 2011 Census of India returns are used to document the proportion of households with access to a latrine on their premises, in the different regions, states and districts of India. While some states have already achieved coverage of 90% or more, in others the proportion served is as low as 22%. There are also wide disparities between urban and rural households. Overall, more than 81.4% of urban households, but only 30.7% of rural households have a latrine. The difference varies widely across the country. Coverage has increased by 10.5% over the decade from 2001. This progress also varied widely between states and between districts within each state; 6.3% of districts made negative progress during that period. The variation between states and districts means that some have already achieved the Millennium Development Goal of halving the proportion of their population

Arabinda Ghosh Indian Administrative Service, Additional District Magistrate, Nadia District, P.O – Krishnagar, PIN – 741101, West Bengal, India Sandy Cairncross (corresponding author) Professor of Environmental Health, London School of Hygiene & Tropical Medicine, Keppel St, London WC1E 7HT, UK E-mail: sandy.cairncross@lshtm.ac.uk

that lacks sanitation, while others, if they continue at the present rate, will not do so for many decades. Study of the causes of these differences offers the hope of ﬁnding ways to accelerate progress in the laggard states and districts. For example, we ﬁnd a close association between district sanitation coverage and female literacy, suggesting an important role for education. Key words

| coverage, India, latrine, open defecation, sanitation

INTRODUCTION Poor sanitation is one of the leading risk factors for child mor-

(Prüss-Üstün et al. ). Intestinal helminth infections

tality worldwide. Improved sanitation, the practice of

cause stunting, late entry to school and impaired cognitive

appropriate hygiene and use of improved sources of drinking

function (Bartram & Cairncross ).

water could prevent 2.4 million deaths (4.2% of all deaths)

Regrettably, it is no surprise that much ill-health is attribu-

annually in our world (Bartram & Cairncross ), including

table to a lack of hygiene, sanitation and water. Only 63% of

an estimated 1.2 million children under the age of ﬁve who

the global population use improved sanitation facilities. At

die from diarrhoea (UNICEF ). Diarrhoeal diseases are

the current rate of progress, the world will miss the Millen-

transmitted through human excreta, and it is therefore criti-

nium Development Goals target of halving the proportion

cally important to have effective barriers in place to prevent

of people without access to basic sanitation. In 2010, 44%

this major transmission route. Improved sanitation alone

of the population in the developing regions were without

could reduce diarrhoea-related morbidity by more than a

improved sanitation facilities. The two regions facing the

third (UNICEF ).

greatest challenges are sub-Saharan Africa and Southern

Childhood underweight causes about 35% of all deaths of children under the age of ﬁve years worldwide (WHO

Asia, where 70% and 59% of the population, respectively, lack access to improved sanitation (WHO/UNICEF a).

). An estimated 50% of this underweight or malnutrition

In addition to its impact on health and nutrition,

is associated with repeated diarrhoea or intestinal nematode

improved sanitation generates both social and economic

infections as a result of inadequate sanitation or hygiene

beneﬁts. The main motivations for sanitation adoption and

doi: 10.2166/washdev.2013.185

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

use include the desire for privacy, to avoid embarrassment

The objective of this study is therefore to highlight the

and to be modern, the desire for convenience, to avoid the

extent of access to sanitation facilities in Indian districts, an

discomforts or dangers of the bush, and for social accep-

essential step to ending the practice of open defecation, and

tance or status (Mara et al. ).

to assess its association with socio-economic inequalities.

The economic beneﬁts of improved sanitation include savings in health system costs, fewer days lost at work or at school through illness or through caring for an ill relative, and time

METHODS

savings from increased convenience (Hutton et al. ). Among sanitation practices, the one that poses not only

Population data

the greatest threat to human health but also an affront to human dignity is open defecation. Though the proportion

The current study uses data from the House Listing and

of people practising open defecation is decreasing, the

Housing Census Data from the 2001 and 2011 Censuses of

absolute number has remained at over 1 billion for several

India to depict the scenario of access to latrine facilities in

years, because of population growth. Most of them (59%)

India. Of the 35 states and union territories in India, some

live in India (WHO/UNICEF a). The 2011 census in

are very small, so that data presented by state are for the

India found that 49.8% of all households practise open

20 most populous states of India. The deﬁnition of each

defecation. In the rural parts of India this proportion is as

region, state and district was as used by the Government

high as 67.3% (Govt of India ).

of India.

Figure 1

Proportion of households with a latrine facility available within premises in Indian regions, states and union territories, 2001 and 2011.

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The WHO/UNICEF Joint Monitoring Programme

India, between the states in each region, and between dis-

(JMP) for water supply and sanitation has developed a ‘sani-

tricts. Box plots have been used to illustrate the present

tation ladder’ to provide an understanding of the proportion

scenario of access to latrine facilities and provide a sum-

of the population globally with no sanitation facilities at all,

mary based on the median, quartile and extreme values.

of those reliant on technologies deﬁned by JMP as ‘unimproved’, of those sharing sanitation facilities of otherwise acceptable technology, and those using ‘improved’ sani-

RESULTS

tation facilities (WHO/UNICEF b). In our study, households reporting that they have any type of latrine facil-

Figure 1 highlights the percentage of households in Indian

ity within their premises have been considered to have

states having a latrine facility in their household, along

access to a latrine facility. As far as we know, their responses

with progress in extending coverage from 2001 and 2011.

to the questionnaire were not checked by observation.

It shows that the proportion of households with a latrine facility is highest in the states of North-Eastern Region at

Analysis

69.4% and lowest in Eastern Region at 35.9%. There is a wide disparity in access to a latrine facility, not only between

The present study focuses on sanitation facilities at district

the regions but also within each region. In Eastern India,

level in India. Initially the data are presented descriptively,

coverage is as high as 70.1% in the Andaman and Nicobar

highlighting the differences in coverage between regions in

Islands and as low as 22.0% in Jharkhand and Odisha

Figure 2

Proportion of rural and urban households with a latrine facility available on the premises in Indian states and union territories, 2011.

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

(Orissa). In the North-Eastern states, it is relatively high.

losing 2% in sanitation coverage, while slow progress was

Coverage with a latrine facility in the household ranges

made in Assam, Jharkhand, Mizoram and Bihar at 0.3, 2.4,

from 91.9% in Mizoram to only 62.0% in Arunachal Pra-

2.9 and 3.9%, respectively.

desh. In Northern Region it ranges from 89.5% in Delhi to

Figure 2 highlights coverage with a latrine facility in

35.0% in Rajasthan. There is also a wide disparity in Wes-

2011, separately for rural and urban households. It tran-

tern India, where the proportion of households having a

spires that only 30.7% of the rural households have a

latrine varies from 79.7% in Goa, to 24.6% in Chhattisgarh.

latrine facility against 81.4% of households in the urban

Coverage in Southern Region is also reasonably good. It is as

areas. The rural–urban gap is highest in Northern Region

high as 97.8% in Lakshadweep, but only 48.3% in Tamil

and lowest in North-Eastern Region. In Eastern India,

Nadu.

25.1% of rural households have a latrine within their house-

Figure 1 also shows that availability of a latrine in the

hold against 77.2% in urban areas. Within the Eastern

households in India has increased by 10.5% during the

Region, Jharkhand has the highest rural–urban disparity, fol-

period from 2001 to 2011. However, here too there were

lowed by Bihar and Orissa. Rural–urban disparity is least in

wide disparities between states. The increase in coverage

Sikkim. In North-Eastern India, Meghalaya has the highest

made by Himachal Pradesh during this period was the high-

rural–urban disparity followed by Arunachal Pradesh. In

est in the country at 35.7%, followed by Haryana, Sikkim,

Northern India, the highest rural–urban disparity is in

Punjab and Uttarakhand at 24.1, 23.8, 22.4 and 20.6%,

Rajasthan followed by Uttar Pradesh. Karnataka has the

respectively. On the other hand, performance was worst in

highest rural–urban disparity in Southern Region followed

Jammu and Kashmir, which actually went backwards,

by Andhra Pradesh and Tamil Nadu. Kerala has made

Figure 3

Latrine facility available on premises in Indian districts by state (2011), showing means, medians, quartiles, extremes and outliers.

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

signiﬁcant progress among Indian states toward achieving

the box to those highest and lowest values, excluding out-

the target of sanitation for all. Latrine coverage in the

liers. A line across the box indicates the median, and the

rural and urban households of Kerala is 93.2% and 97.4%,

centre of the circle within the box indicates the mean.

respectively. In Western Region, the rural–urban disparity

In order to formulate strategy at the national level, it is

is highest in Madhya Pradesh followed by Dadra & Nagar

important to know the present situation not only by states

Haveli. Availability of a latrine facility in the rural and

but also by district. The average performance of individual

urban households of Chhattisgarh is extremely low and

states may hide the intra-state disparities among the districts.

needs special attention.

Table 1 depicts the progress in household coverage with

In Figure 3, a box plot has been used to illustrate further

latrines in Indian districts during the period 2001–11.

the wide range of latrine coverage levels in different districts

During this period, some of the districts were reorganised

in each state of India. They provide a summary based on the

and are not comparable between 2001 and 2011. In our

median, quartile and extreme values and represent the inter-

selected 20 major states, there are 558 districts. In this

quartile range that contains the middle 50% of the values.

study, it was possible to compare results for 509 districts

The box plots show the lowest value (the bottom horizontal

(i.e. 91.2% of the total districts) and so assess their progress.

line on each plot) and the highest value (the top horizontal

The districts have been divided into six categories according

line of each plot). The whiskers are lines that extend from

to their rates of progress.

Table 1

Progress in latrine coverage in Indian districts during 2001–11

Progress during 2001–11 No. of

Districts

Very slow

Slow

Moderate

Good

Very good

Excellent

State

districts

studied

Negative

(0–10%)

(10–20%)

(20–30%)

(30–40%)

(40–50%)

(>50%)

Andhra

Assam

Bihar

Chhattisgarh

Gujarat

Haryana

Himachal

Jammu

Jharkhand

Karnataka

Kerala

Madhya

Maharashtra

Orissa

Punjab

Rajasthan

Tamilnadu

Uttar Pradesh

Uttarakhand

West Bengal

Total

558

509

280

122

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Table 1 shows that 6.3% of the districts made negative

within the premises. In 29.1% of districts in India, 20% to

progress, and 55.0% and 24.0% of them have made very

40% of households have such a facility . In 18.4, 18.3 and

slow and slow progress, respectively. Only 9.8% of them

14.7% of Indian districts, respectively, the percentage of

has made moderate progress. On the other hand 3.5, 1.0

households with a latrine available on the premises is 40–

and 0.4% of them made good, very good and excellent pro-

60%, 60–80%, and above 80%. Kerala is the only state in

gress, respectively, during this period. During 2001–11, the

India where in all districts, more than 80% of households

best performing Indian states in terms of their progress

have a latrine on the premises.

were Haryana, Himachal Pradesh, Punjab, Uttarakhand

The pattern visible in the data at national level, in which

and Maharastra, respectively. On the other hand, the

access to sanitation is closely associated with socio-econ-

worst performing Indian states were Assam, Uttar Pradesh,

omic status of the household, is borne out at state level in

Jammu & Kashmir, Jharkhand and Madhya Pradesh.

our data. The percentage of households with a latrine in

Table 2 shows, for each state, the numbers of districts

each state is closely correlated with the state’s wealth con-

where in 2011, the percentage of households with a latrine

centration

index,

indicator

the

degree

was in a given range. It shows that in 19.5% of all districts

concentration of latrines in the hands of wealthier house-

in India, no more than 20% of households have a latrine

holds (r ¼ 0.924; p < 0.001; data not shown). Only Jammu & Kashmir departs markedly from the trend, with

Table 2

Percentage of households with a latrine facility available within premises in

a lower concentration index than its coverage rate would

2011

suggest (IIPS ; Kakwani et al. ). Figure 4 shows the rate of access to sanitation in each

No. of districts according to percentage of households having latrine facility available

district, plotted against the female literacy rate. There is a

within premises

marked inverse relationship; in districts with higher female > 20 –

> 40 –

> 60 –

Total

State

20 40

> 80

districts

Andhra

Assam

Bihar

Chhattisgarh

Gujarat

Haryana

Himachal

Jammu

Jharkhand

Karnataka

Kerala

Madhya

Maharashtra

Orissa

Punjab

Rajasthan

Tamilnadu

Uttar Pradesh 21

Uttarakhand

West Bengal

India

125

186

118

117

640

literacy rates, open defecation is markedly less prevalent. Figure 5 shows, for selected Indian states, how long it will take to achieve the MDG sanitation target at the current rate of progress. The rate of progress was assessed by comparing the 2001 and 2011 census results as before.

Figure 4

Scatter plot of latrine facilities in the household rates in Indian districts against female literacy rates. The trend line is based on ordinary least squares regression, and the narrow strip on either side of it shows the 95% conﬁdence limits for its position. R 2 Linear ¼ 0.476. Source: Data from the 2011 Census of India.

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

equity and good governance, and it is very hard to tell which associations, if any, are due to causal links. Kerala’s high adult literacy rate is probably the ﬁrst of these variables in which Kerala stands out, having been noteworthy since the 19th century when the princely states of Travancore and Cochin devoted more resources than the rest of India to public education, and pulled ahead of the rest of India in all indicators of social development (George ). Target 7C of the Millennium Development Goals is to halve the proportion of the population (baseline 1990) without basic sanitation by 2015. The Millennium Development Goals Report 2012 has highlighted that despite improvement in most of the developing regions, the MDG sanitation target is still out of reach (United Nations ). Given that 76% of households lacked any sanitation facility in 1990, India is required to reduce this proportion to 38% by 2015 (UNDP ). Open defecation is a traditional behaviour in rural India. Figure 5

Predicted delay in reaching the MDG sanitation target: selected Indian states.

This, along with the relative neglect of sanitation in terms of development priorities, was reﬂected in the country’s low sani-

Millennium Development Goals for individual states were set on the basis of the National Family Health Survey (NFHS-1) data (1992–93). Five Indian states had already achieved the MDG target in 2011. The estimated number of years required to achieve the MDG target for sanitation has been expressed as years of delay from the 2015 deadline. For example, Andhra Pradesh will need another 3 years to achieve the target and Jharkhand, at the present rate, requires 156 years. As can be seen from Table 2, the range for individual districts is still wider than that for states.

tation coverage at the close of the 1990s when only one in ﬁve rural households had access to a toilet (Census 2001). This fact, combined with low awareness of hygiene, made the achievement of the sanitation target a pressing challenge, particularly in rural India. In response to this challenge, the Government of India launched the Total Sanitation Campaign (TSC) in 1999 with the goal of achieving universal rural sanitation coverage by 2012 (World Bank Water & Sanitation Program ). In spite of this prioritisation of sanitation, the country is expected to miss the millennium target by about 11 percentage points. If the present trend continues, it will take an additional 20 years to achieve the MDG sanitation target. A minority of states have already reached the target (Figure 5), thus proving that it is achievable and not unrea-

DISCUSSION

listic under Indian conditions. However, the goal can only be reached nationally in the foreseeable future if radical

We have shown very large discrepancies in latrine coverage

measures are taken and substantial effort devoted to the

between different parts of India. If we can improve our

underachieving states and districts.

understanding of the reasons underlying these differences, it would help us to identify policies to increase the rate of progress toward sanitation for all. In this context, the associ-

ACKNOWLEDGEMENT

ation between sanitation coverage and female literacy is a striking ﬁnding. It is likely that both of these variables are

Professor Cairncross’s input to this study was supported by UK

collinear with a number of other factors, such as social

aid from the Department for International Development

A. Ghosh & S. Cairncross

The uneven progress of sanitation in India

(DFID) as part of the SHARE research programme (www. SHAREresearch.org). However, the views expressed do not necessarily reﬂect the Department’s ofﬁcial policies.

REFERENCES Bartram, J. & Cairncross, S.  Hygiene, sanitation, and water: forgotten foundations of health. PLoS Med 7 (11), e1000367. George, K. K.  Kerala’s Economy; Growth, Structure, Strength and Weakness. CSES Working Paper No. 25. Centre for Socio-economic & Environmental Studies, Kochi, India, http://csesindia.org/publications.php?page=2 (accessed 16 October 2013). Government of India  Houselisting and Housing Census Data Tables – District Level. Ofﬁce of the Registrar General & Census Commissioner, India, http://www.censusindia.gov. in/2011census/hlo/District_Tables/HLO_District_Tables. html (accessed 16 October 2013). Hutton, G., Haller, L. & Bartram, J.  Economic and Health Effects of Increasing Coverage of Low Cost Household Drinkingwater Supply and Sanitation Interventions to Countries Offtrack to Meet MDG Target 10. World Health Organization, Geneva, http://www.who.int/water_sanitation_health/ economic/mdg10_offtrack.pdf (accessed 23 November 2012). IIPS  District Level Household and Facility Survey (DLHS-3), 2007–08: India. International Institute for Population Sciences, Mumbai, India. Kakwani, N. C., Wagstaff, A. & Van Doorslaer, E.  Socioeconomic inequalities in health: Measurement, computation and statistical inference. Journal of Econometrics 77 (1), 87–104. Mara, D., Lane, J., Scott, B. & Trouba, D.  Sanitation and health. PLoS Med 7 (11), e1000363.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Prüss-Üstün, A., Bos, R., Gore, F. & Bartram, J.  Safer Water, Better Health: Costs, Benefits and Sustainability of Interventions to Protect and Promote Health. World Health Organization, Geneva. UNDP  Millennium Development Goals India Country Report 2011, http://www.un.org.in/UNDP/MDG/Indian_ MDG_Country_Report_2011.pdf (accessed 16 October 2013). UNICEF  Progress for Children: A Report Card on Water and Sanitation, Number 5, UNICEF, New York. UNICEF  The State of The World’s Children 2012: Children in an Urban World. http://www.unicef.org/sowc2012/pdfs/ SOWC%202012-Executive%20Summary_EN_13Mar2012. pdf (accessed 16 October 2013). United Nations  The Millennium Development Goals Report 2012. United Nations, New York. WHO  Underweight in Children. World Health Organization, Geneva, http://www.who.int/gho/mdg/poverty_hunger/ underweight_text/en/index.html (accessed 16 October 2013). WHO/UNICEF a Progress on Drinking Water and Sanitation: 2012 Update. World Health Organization, Geneva and UNICEF, New York. Available at: http://www. unicef.org/media/files/JMPreport2012.pdf (accessed 29 January 2014). WHO/UNICEF b Definitions and Methods: Introduction. Joint Monitoring Programme for Water Supply and Sanitation, http://www.wssinfo.org/ definitions-methods/introduction/ (accessed 16 October 2013). World Bank Water & Sanitation Program  A Decade of the Total Sanitation Campaign: Rapid Assessment of Processes and Outcomes. WSP India, Delhi, http://www. wsp.org/sites/wsp.org/files/publications/WSP_India_ TSC_Report_Vol_1_Press.pdf (accessed 16 October 2013).

First received 4 July 2013; accepted in revised form 10 October 2013. Available online 12 December 2013

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Economic efﬁciency of sanitation interventions in Southeast Asia Guy Hutton, U-Primo Rodriguez, Asep Winara, Nguyen Viet Anh, Kov Phyrum, Liang Chuan, Isabel Blackett and Almud Weitz

ABSTRACT Economic evidence on sanitation supports decisions on resource allocation and selection of efficient and affordable sanitation interventions. This study presents the economic efficiency (costs versus benefits) of sanitation interventions to better manage human excreta, from 47 field sites, covering six countries of Southeast Asia (Cambodia, China, Indonesia, Lao PDR, the Philippines and Vietnam). Costs were estimated in each location, while benefits (improved health, avoided water pollution, reduced sanitation access time, resource recovery) were estimated using evidence from published studies and field sites. The economic return per currency unit invested, known as the benefit-cost ratio (BCR), and the cost per disability-adjusted life-year (DALY) averted were estimated. Across 25 rural field sites, pit latrines had a BCR of at least 5 in all countries, except Cambodia where the BCR was 2. In 22 urban field sites, septic tanks with wastewater management had a BCR of at least 2. Costs per DALY averted were found to be ‘cost-effective’ for most sanitation interventions in all countries. Economic performance declined significantly when considering non-use of facilities by households or unused infrastructural capacity. However, the economic net returns were positive under all pessimistic scenarios examined in one-way sensitivity analysis. This study demonstrates that sanitation is a highly profitable social and economic investment in six Asian countries. Key words

| Asia, beneﬁt-cost ratio, capital cost, economic return, recurrent cost, sanitation | Cambodia, China, East Asia, Indonesia, Lao PDR, Philippines,

Additional geographic key words

Southeast Asia, Vietnam

Guy Hutton (corresponding author) Water and Sanitation Program, World Bank, 55 Lodi Estate, New Delhi, 110003, India E-mail: ghutton@worldbank.org Kov Phyrum Water and Sanitation Program, World Bank, 113 Norodom Blvd, Phnom Penh, Cambodia Isabel Blackett Water and Sanitation Program, World Bank, Indonesia Stock Exchange Building, Tower 1, 9th Floor. Jl. Jenderal Sudirman Kav 52-53, Jakarta 12190, Indonesia Almud Weitz Water and Sanitation Program, World Bank Group Singapore Ofﬁce, 10 Marina Boulevard, Marina Bay Financial Center, Tower 2, #34-02 Singapore 018983 U-Primo Rodriguez Economics Department, University of the Philippines, Guerrero corner Osmeña Sts. Diliman, Quezon City, 1101, Philippines Asep Winara Mitra Lingkungan Duta Consult (MLD), Ventura Building, 4th ﬂoor, Suite 405, Jl. R.A. Kartini No. 26, Cilandak, Jakarta 12430, Indonesia Nguyen Viet Anh Institute of Environmental Science and Engineering (IESE), Hanoi University of Civil Engineering, 55 Giai phong Road, Hai Ba Trung District, Hanoi, Vietnam Liang Chuan Yunnan Academy of Social Sciences (YASS), 577 Huan Cheng Xi Lu, Kunming City, Yunnan Province, 650034 China

This paper is in the public domain: verbatim copying and redistribution of this paper are permitted in all media for any purpose, provided this notice is preserved along with the paper’s original DOI. Anyone using the paper is requested to properly cite and acknowledge the source as Journal of Water, Sanitation and Hygiene for Development 4(1), 23–36. doi: 10.2166/washdev.2013.158

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

ABBREVIATIONS

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

However, despite its central place in human development, sanitation continues to lose ground to other

BCR

Beneﬁt-Cost Ratio

development targets when it comes to priority setting by gov-

CBA

Cost-Beneﬁt Analysis

ernments, households, the private sector and donors. For

CEA

Cost-Effectiveness Analysis

example, from a global review with a cross-section of 13

CLTS

Community-Led Total Sanitation

countries reporting data, government expenditure on drink-

CSR

Corporate Social Responsibility

ing-water and sanitation comprised a median of 0.7% of

ESI

Economics of Sanitation Initiative (of WSP)

Gross Domestic Product (GDP), compared with 2.3% for

GDP

Gross Domestic Product

health and 5.5% for education (World Health Organization

HCA

Human Capital Approach

). Aid commitments for water and sanitation were at

IRR

Internal Rate of Return

4.7% (US$ 7.8 billion) (billion ¼109) of the total reported

Joint Monitoring Programme for Water Supply

development aid in 2010, at around the same cost as the

and Sanitation

stated administrative costs of donors. However, these aid

MDG

Millennium Development Goal

ﬂows do not target well the lowest income countries, nor

UDDT

Urine-diverting dehydration toilet

the low cost sanitation solutions that reach the poorest and

United Nations

most vulnerable groups (World Health Organization ).

JMP

UNICEF United Nations Children’s Fund

The reasons for low prioritization of water and sani-

US$

United States Dollar

tation are many. Sanitation remains a largely taboo subject

VSL

Value-of-a-statistical life

in most cultures; it is not an attractive subject for media or

WHO

World Health Organization

politicians to promote as a worthy cause; and policy

WSP

Water and Sanitation Program

makers are unmotivated or remain ignorant of evidence of the health and economic beneﬁts of sanitation. One source of valuable advocacy material is evidence of the economic impacts of the lack of good sanitation. Econ-

BACKGROUND

omic evidence has been put to good effect in other sectors to raise awareness of low priority issues. One example is

Good sanitation is a basic human need. At home as well as

the WHO Commission on Macroeconomics and Health

at the workplace, school or a public place, people appreciate

which in 2001 published a report that drew the link between

and value a clean, safe, private and convenient place to uri-

the high rate of avoidable diseases and lack of economic

nate and defecate. The importance of good sanitation was

growth in low income countries (Sachs ). Hence, in

recognized in the Millennium Development Goals (MDG),

2007 the World Bank Water and Sanitation Program

where target 7C is to ‘halve, by 2015, the proportion of the

initiated a series of studies that estimated the economic

population without sustainable access to safe drinking

and social impacts of unimproved sanitation on the popu-

water and basic sanitation’. In 2010, the human right to

lations and economies of ﬁve countries in Southeast Asia

safe drinking-water and sanitation was adopted by the

(Hutton et al. , ). The studies showed that the econ-

United Nations General Assembly, which recognizes that

omic impacts of poor sanitation for the year 2006 were US$

sanitation is essential for the full enjoyment of life and of

23 per capita, or equivalent to 2% of annual GDP across the

all human rights (United Nations ).

ﬁve countries (1.3% in Vietnam, 1.5% in the Philippines,

As well as being important in its own right, good sani-

2.3% in Indonesia, 5.6% in Lao PDR and 7.2% in Cambo-

tation contributes importantly to achieving other MDG

dia). These results have helped galvanize action in and

targets such as child mortality reduction, school enrolment,

beyond these countries.

improved nutrition, gender equality, clean drinking water,

Economic impact ﬁgures highlight a problem society

environmental sustainability and quality of life of slum

has. However, the estimates do not support policy makers

dwellers.

in making rational decisions on how to act. Cost-beneﬁt

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

analysis (CBA) is a well-known technique for assessing the

for-money. In this study, the following technical efﬁciency

comparative performance of different intervention options

measures were calculated: BCR, cost per health impact

(Drummond et al. ), thus supporting the selection of

averted, net present value, IRR and payback period

socially efﬁcient sanitation policies, programs and technol-

(Sugden & Williams ; Hanley & Spash ; Hutton

ogies. CBA speciﬁcally provides efﬁciency metrics such as

et al. ). The ﬁrst two of these are presented in this

the beneﬁt-cost ratio (BCR) and the internal rate of return

paper; full results are available from the Water and Sanitation

(IRR) which help identify which interventions generate

Program. The study aim was therefore to evaluate the techni-

greater economic return (Curry & Weiss ). Previous

cal efﬁciency of different sanitation intervention options, as

global studies have provided an indication that ‘improved’

delivered in real ﬁeld settings, and make recommendations

sanitation delivers high economic returns, in the most

for which interventions should be considered in planning

recent study at US$ 5.5 per US$ invested (Hutton ).

exercises to maximize the returns on the allocation of a

However, improved country studies are needed to provide

speciﬁc budget (‘allocative’ efﬁciency).

greater granularity on which sanitation interventions are more efﬁcient, and in which sub-national contexts.

Alternative scientiﬁc designs for determination of causality were compared. While more robust study designs –

CBA requires all major costs and beneﬁts to be converted

such as randomized controlled trials – are preferred from

into monetary units, which often presents methodological

the scientiﬁc perspective, however, the study budget, time-

challenges. To avoid converting beneﬁts into monetary

line and inherent difﬁculties in conducting trials in

units, cost-effectiveness analysis (CEA) enables comparison

sanitation did not allow for such studies to be conducted.

of physical units of beneﬁts with intervention costs to

Instead, evidence of the health impact of improved sani-

inform where the greatest marginal gain per unit of expendi-

tation was extracted from meta-analysis studies. Field

ture is. Ministries of Health compare the cost per health unit

surveys were implemented in 47 ﬁeld sites across the six

obtained from sanitation interventions, while Ministries

countries (25 rural and 22 urban) to provide key contextual

responsible for environmental or water protection compare

evidence to feed into the economic model. These surveys

the cost per pollution reduction unit from sanitation interven-

included a household questionnaire applied in 8,470 house-

tions. Less common in CBA or CEA is the assessment of

holds (roughly 180 per ﬁeld site, based on a minimum

social impacts. To date, economic efﬁciency studies have

sample of 30 per type of household sanitation); focus

not measured aspects of sanitation such as dignity, comfort,

group discussions; a physical location survey; health facility

prestige, security, gender equality, household cleanliness,

survey; market survey for local prices; and a review of imple-

and aesthetics of the community environment (Van Minh

menting agency and government reports.

& Nguyen-Viet ).

The economic model captures and compares all the costs and beneﬁts, using best available sources for each model variable. The model was run for 20 years, with

METHODS

costs and beneﬁts after the initial year being discounted at

The purpose of this study is to generate improved economic

applied in public sector projects in these countries. The pre-

information to help sanitation decision makers incorporate

sent value of the discounted costs and beneﬁts were

efﬁciency criteria into their selection of sanitation technol-

compared to estimate the technical efﬁciency measures. Pri-

ogies. On the cost side, decision makers need to understand

mary data from surveys in each ﬁeld setting were used where

more about the size of costs (e.g. investment, operation, main-

possible, and these were supplemented with other selected

tenance) and when they occur, in order to make the

data from national surveys or international sources. The

appropriate investment decision that increases intervention

economic model enabled comparison of outcomes of house-

effectiveness and sustainability. On the beneﬁt side, the mon-

holds both with and without improved sanitation, and

etary as well as non-monetary impacts need to be better

comparing different technologies on the sanitation ladder.

understood to select interventions that provide better value-

These include shared sanitation (share with neighbors,

a rate of 8% per year, reﬂecting the typical discount rate

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

public toilet), private dry sanitation (dry pit latrine, urine-

activities due to illness, and the opportunity cost of time.

diverting dehydration toilet (UDDT)), and private wet sani-

The latter was approximated by using 30% of the average

tation (wet pit latrine, biogas digester, septic tank and

income for adults and 15% of the average income for children

sewerage – with and without wastewater management).

(Hutton et al. ). For children under 5, the time of the

The options evaluated varied between ﬁeld sites.

child carer was applied at 15% of average income. Welfare

Capital, operational and maintenance cost information

gains due to averted premature death were valued by apply-

was collected from households, private sector providers and

ing the relative health risk reduction to the risk of death per

government or donor projects, and triangulated to estimate

age group from fecal-oral disease, and multiplying by the esti-

the full costs of on-site and off-site sanitation (where relevant).

mated value of life using the human capital approach (HCA).

For the cost-beneﬁt comparisons, costs were converted to

The HCA approximates economic losses by estimating the

annual equivalent values based on the expected lifespan of

future discounted income stream from a productive person

each sanitation technology. The year of cost data was 2008.

who dies prematurely. The HCA provides a lower bound esti-

Beneﬁts estimated in the ﬁeld sites include improved

mate compared to other techniques such as the value-of-

health, reduced water pollution and hence improved

statistical-life, and hence gives a conservative estimate of

access to clean drinking-water sources, reduced sanitation

the beneﬁts.

access time, and resource reuse (value of fertilizer and

The beneﬁts of avoided water pollution from poor sani-

biogas produced, in the Philippines, Vietnam and Yunnan

tation focused on water use for domestic purposes, in

Province only). Diseases considered in this study include

particular drinking water. When the ﬁnancial means and

all types of diarrheal disease, helminths, hepatitis A and E,

products are available, communities and individual house-

malnutrition and diseases related to malnutrition (malaria,

holds often take mitigative measures to avoid consuming

acute lower respiratory infection and measles). Disease

water drawn from polluted sources. This study modeled the

rates and mortality were collected from Demographic and

potential economic gains from improving sanitation on two

Health Surveys and the World Health Organization

sets of common mitigation measures – switching water

(World Health Organization ), adjusted to each ﬁeld

source and conducting household water treatment. The

site based on socio-economic characteristics of the sampled

costs of these practices were compared with the costs

populations. As not all fecal-oral diseases have a pathway

under a situation of minimal pollution levels, and a partial

from human excreta, 88% (attribution rate) of diarrheal dis-

cost attributed to the contribution of poor sanitation to over-

eases were assumed to be due to poor management of

all water resources pollution. Water quality measurement

human excreta (Prüss et al. ).

studies were conducted to enable detailed analysis of the

For CBA, three health-related economic costs were

likely impacts of improved sanitation on local water quality.

included: health care costs, health-related productivity

Access time savings due to poor sanitation were esti-

costs, and premature mortality costs. For fecal-oral disease,

mated based on practices of those with no household

the relative risk reduction of 36% for improved sanitation

facility, including distance to sites of defecation. These

was based on the latest published meta-analysis (Waddington

were extracted from the household surveys and assessed sep-

et al. ). For improved on-site sanitation with fecal sludge

arately for men and women, and for children, due to their

management, a reduction in fecal-oral disease of 56% was

different practices. The value of time saved was estimated

used as an average of two previous studies (Moraes et al.

based on the same opportunity cost of time as the health-

; Barreto et al. ). Health care savings were calculated

related time savings (see above).

by applying the relative risk reduction to the average disease

Resource reuse was practiced in 13 out of 47 ﬁeld sites:

cost per household (based on treatment-seeking rates for

UDDT in the Philippines, and both UDDT and biogas in

different health care providers, hospital admission and unit

Vietnam and Yunnan province, China. In these sites, the

costs of health care). Gains in health-related productivity

value of excreta reuse was measured based on the uses to

were estimated by applying the relative health risk reduction

which the products were put. More commonly, the house-

to the disease rates, the average time taken off productive

hold used the product, in which case the value of saved

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

expenditure on fertilizer or energy was estimated. When the

to elicit from individuals. Furthermore, perceptions and pre-

product was sold, the selling price (and revenue) was

ferences vary considerably from one individual and social

obtained from the household or the market place. In the

group to another. Therefore, intangible beneﬁts were not

case of biogas digesters, additional excreta volume was com-

converted to monetary values. Instead, the understanding

monly sourced freely from livestock belonging to the

of sanitation, sanitation preferences and the sanitation

household.

decision-making process were elicited from household inter-

Sensitivity analysis was conducted to assess the robustness of results under different input values of key variables.

views. A total of 115 focus group discussions were conducted across the six countries.

Two representative sites and technologies from Indonesia

For CEA, three indicators of disease burden were evalu-

(Malang urban site, community sewerage system; and

ated: numbers of cases (incidence or prevalence), deaths,

Lamongan rural site, private wet pit latrines) were selected

and disability-adjusted life-years (DALYs). DALYs are

to illustrate the impact on the BCR of the alternative input

reported in this paper, calculated using standard methods

values of each of the following seven selected parameters:

by combining morbidity components (disease rate, disability

discount rate; different number of members per household;

weight and illness duration) with mortality components

disease morbidity risk; mortality risk; value of time; the

(mortality rate and life expectancy). Standard weights and

value of averting a premature death using the value-of-statisti-

disease durations are sourced from the Global Burden of

cal-life method, adjusting a value of US$ 2 million from

Disease study (Murray et al. ), and average healthy life

developed countries by the difference in GDP per capita;

expectancy is based on values for each country (Fox-

and the length of life span of the hardware.

Rushby & Hanson ; World Health Organization ).

Intangible beneﬁts of sanitation such as comfort, privacy, convenience, safety, social status, prestige and environmental aesthetics are major factors in personal and

RESULTS

community welfare, and are thus important determinants of sanitation choices. Due to the private and subjective

Figure 1 shows the BCRs for rural areas. The most basic

nature of intangible beneﬁts, reliable information is difﬁcult

sanitation type, the pit latrine, had returns of at least ﬁve

Figure 1

Beneﬁt-cost ratios of sanitation interventions in rural areas, and up-front investment and annual recurrent costs per household (US Dollars, 2008). Note: in Vietnam, ‘pit latrine’ includes both dry and wet pits.

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

times its costs in all but one country, Cambodia. Across both

reuse options and pit latrines in each country. In Yunnan,

wet and dry pit latrines, the returns were highest in Lao PDR

the UDDT had higher returns than a pit latrine. This was

at over eight times. The returns were two or more times in

accomplished through a large-scale government program

Cambodia for both dry and wet pit latrines. The average

that made the unit cost of UDDT facilities similar to ordin-

returns are lower in Cambodia partly due to the fact that

ary pit latrines. In the Philippines, the UDDT had

costs in Cambodia were at least as high as other countries,

signiﬁcantly lower returns than a pit latrine, due to the

while the value of the beneﬁts was diminished by the

high unit costs of delivering UDDTs to populations (both

lower average incomes in Cambodia. A second ﬁnding is

higher hardware and software costs). In Vietnam, the

that dry pit latrines do not consistently have a higher

UDDT had a marginally lower return than a pit latrine,

return, despite having a lower investment cost in all

due to the higher unit cost and limited economic value of

countries. In the Philippines and Cambodia, dry pit latrines

compost. For the biogas digesters, the large on-farm systems

have lower returns than wet pit latrines, while the reverse is

surveyed in Vietnam had high positive economic returns,

true in Indonesia and Lao PDR. This ﬁnding was observed

largely accounted for by the resource reuse value. In

for the Philippines and Cambodia largely due to the fact

Yunnan, the high return on biogas of at least seven times

that the expected life of a dry pit latrine was less than half

was accounted for largely by the health and time savings

that of a wet pit latrine.

rather than the recovered gas. In Yunnan, the value of com-

Returns for private toilets with septic tanks are signiﬁcantly lower compared to pit latrines. In only two

post or biogas alone was enough to cover the cost of the improved sanitation intervention.

countries do the economic returns reach four times (Viet-

Figure 2 shows the BCRs of different technologies found

nam and Lao PDR), while the returns are two times in the

in urban areas. In some urban settings, where density is low

Philippines. However, the economic returns are net positive

and properties have sufﬁcient plot size for constructing

for all technologies and all countries (i.e. the BCR remains

additional buildings, pit latrines remain a feasible, affordable

above one).

and efﬁcient sanitation option. While wet pit latrines in urban

Reuse options have economic returns of at least two

areas had a poorer economic performance compared to rural

times, ranging from 2 in the Philippines to 9 in Yunnan

areas in most countries, they were the most efﬁcient option in

for the UDDT option. Some variation was found between

urban areas with economic returns ranging from 3 times in

Figure 2

Beneﬁt-cost ratios of sanitation interventions in urban areas, and up-front investment and annual recurrent costs per household (US Dollars, 2008).

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indonesia to 8.5 times in Vietnam. This favorable result needs

The results presented above show the ideal performance

to be balanced with the health risk of using pit latrines in den-

of the sanitation technologies, where all households who

sely populated areas, especially the risk due to groundwater

obtain a sanitation facility would use it appropriately over

pollution. The higher cost option of septic tanks (with waste-

the full life of the hardware. However, this was not always

water management) was also economically viable in all

the case, either due to non-compliance of the household in

countries, with economic returns of around two or more

using the facility consistently or due to the non-function of

per unit spending. However, there are signiﬁcant variations

the hardware itself. The household survey obtained infor-

between countries, with returns from under two times in

mation on general usage of the facility by all household

Cambodia and Indonesia, around three times in Yunnan

members; and the survey of the projects found out the

and Vietnam, to at least four times in the Philippines. The

extent to which infrastructure, such as sewerage and waste-

omission of the monetized environmental beneﬁts of sludge

water treatment plant capacity, was underutilized. Decline

wastewater management also leads to lower BCRs for these

in performance from ideal to actual was signiﬁcant for tech-

options than would otherwise be the case.

nologies in all countries. For rural areas, most technologies

The highest annualized cost option – sewerage with

experienced a decline in performance of between 10 and

wastewater treatment – had slightly lower economic returns

30%. In urban areas, the decline in performance was on

than septic tanks in most countries. In Cambodia, the sewer-

average higher than in rural areas, with the BCRs for most

age and treatment system performed poorly with an

technologies reducing by between 20 and 40%. The highest

economic return of 0.14 times. This is the case because

decline was observed in Cambodia (urban site), because at

the urban site is a relatively small town with connections

the time of the study, only 20% of the targeted households

planned for only 3,300 households, with a high land pur-

were connected to the sewerage system.

chase cost and costly infrastructure (wastewater treatment

The sensitivity of these baseline results was assessed by

plant and new sewerage system), made more costly by

examining the impact of changing key assumptions and

hilly terrain. In most sites assessed, wastewater treatment

input values on the BCRs of one urban site (Malang) and

plants release the treated wastewater to the environment.

one rural site (Lamongan) in Indonesia. The results are

In Vietnam, on the other hand, wastewater is reused after

shown in Table 1. While the economic returns did diminish,

treatment in some cities and hence makes an important con-

they remained net positive under all scenarios. In Malang

tribution to the overall economic returns.

urban site, economic returns were most sensitive to the

In all countries and for most sanitation technologies,

value of time, number of household members, and discount

health beneﬁts and time savings accounted for the majority

rate. For Lamongan rural site, where the baseline economic

of the overall beneﬁts. Signiﬁcant variation was found

returns were higher at 6.1, the economic returns did not dip

across countries in the speciﬁc contributors to beneﬁts.

below 2.2 times in any of the scenarios explored. The econ-

This was due to the different levels of impact and the differ-

omic returns for Lamongan increased to above 10 when

ent relative values of those impacts. For example, health

morbidity rates were double the baseline values and when

care savings make up a signiﬁcant share of overall beneﬁts

a high value of time was used. A key variable for both

in Lao PDR and Indonesia, while mortality reductions are

sites is the number of household members – when there

more important in Vietnam. The value of time saving

were three adults but no children in the household, the

makes up at least 50% of the overall beneﬁt in four out of

BCR reduced signiﬁcantly.

six countries. Reduced costs due to less polluted water

The cost per DALY averted differs signiﬁcantly between

sources are an insigniﬁcant beneﬁt in all countries, except

countries and between technologies. According to the

Vietnam where these accounted for close to 10% of

World Health Organization, a health-improving interven-

beneﬁts. However, it is worth noting the full beneﬁts from

tion with a cost per DALY of less than the GDP per

improved water quality and avoided cost of pollution down-

capita reﬂects a very cost-effective use of funds for health

stream was not considered for assessed urban sites in this

interventions, and a cost per DALY of between one and

study.

three times the GDP per capita reﬂects a cost-effective use

Table 1

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Results of one-way sensitivity analysis – beneﬁt-cost ratios Malang (urban)

Lamongan (rural)

Community sewerage system Parameter

Range (high and low values)

Baseline BCR

Private wet pit latrine

Ideale

Actualf

Ideale

Actualf

2.3

1.9

6.1

5.1

Discount rate (baseline 8%)

0% 12% 20%

5.3 2.0 1.6

4.4 1.6 1.3

9.0 5.8 5.5

7.5 4.8 4.5

Morbidity rate (all diseases)

Half baseline value Double baseline value

1.8 3.4

1.5 2.8

3.9 10.6

3.3 8.7

Mortality rate (all diseases)

Half baseline value Double baseline value

2.2 2.5

1.8 2.0

6.0 6.3

5.0 5.2

Value of lifea

Use of VSL

2.8

2.3

7.0

5.8

Value of time

Low assumption High assumption

1.4 5.5

1.1 4.6

4.6 12.0

3.8 10.1

Household membersc

3 adults only 3 adults and 4 children

1.3 3.2

1.1 2.7

2.7 11.3

2.2 9.3

Life span of hardwared

Low assumption High assumption

1.6 2.8

1.3 2.3

4.9 7.2

4.0 6.0

Value-of-statistical life (VSL) is US$ 50,818 (transfer of VSL of US$ 2 million from developed countries based on GDP per capita differential), while using the human capital approach the

value of premature death for adults is US$ 14,367. b In baseline, value of time 30% and 15% of GDP per capita, for adults and children respectively. Low value is 15% and 0%, respectively. High value is 100% and 50%, respectively. c

Baseline is 1.8 children and 3.2 adults in Malang; 1.6 children and 3.4 adults in Lamongan.

Communal sewerage life span assumptions: 20 years in the baseline, 10 years for low scenario and 30 years for high scenario. Wet pit latrine life span assumptions: 5 years in the baseline,

3 years for low scenario and 8 years for high scenario. e The ‘ideal’ scenario reﬂects a situation where all households who obtain a sanitation facility use it appropriately over the full life of the hardware. f

The ‘actual’ scenario reﬂects the situation where households who obtain a sanitation facility do not use it due to disrepair or not changing habit, or in the case of infrastructure, such as

sewerage and wastewater treatment plant, the capacity of the facility is underutilized.

and septic tanks was also lower than the GDP per capita, as was the case for pit latrines in Vietnam. Septic tanks in Lao PDR and Philippines had a cost per DALY averted greater than three times the GDP per country. Hence the health arguments alone cannot justify these latter interventions, based on the international thresholds deﬁned above. In urban areas, the cost per DALY averted was signiﬁcantly higher than in rural areas, due to the higher cost of Figure 3

Cost per DALY averted for wet pit latrines in rural areas, compared with the

the sanitation interventions and also the lower potential

GDP per capita (all prices are for 2008 except Lao PDR 2010 prices).

for health benefits compared to rural areas. Using the same thresholds defined above, pit latrines reflect a cost-

of funds for health interventions (Sachs ; World Health

effective intervention in urban areas of all countries,

Organization ). Figure 3 illustrates the case of wet pit

except Vietnam where they were very cost-effective. Septic

latrines in rural areas. In Cambodia and Lao PDR, a

tanks reﬂect a cost-effective intervention in Yunnan, Indone-

DALY can be averted for less than US$ 500, which reﬂects

sia and Cambodia, and were very cost-effective in Vietnam.

a very cost-effective health intervention. In Indonesia and

Sewerage was a cost-effective intervention only in Indonesia

Yunnan, the cost per DALY averted for both pit latrines

and Vietnam.

G. Hutton et al.

Table 2

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Factors explaining sanitation coverage and lack of coverage

Country

Why those with toilets have them

Why those without toilets do not have them

1. 2. 3. 4.

Saves time and is convenient Safe from danger (women) Improved health Does not smell and spoil environment

1. 2. 3. 4.

High cost of latrine Not used to using latrine No space for building latrine (Unimproved) latrine has bad smell

1. 2. 3. 4. 5. 6.

Source of pride (men) Safe to go any time (women) Offers privacy (women) No need to worry about children (women) No need to queue for public toilets or get up early Clean and comfortable, no ﬂies

1. 2. 3. 4.

High cost of latrine No space in or near house Never been offered latrine Not thought about it

1. 2. 3. 4.

Provided support by project Clean, no smell Convenient Easy to install and use

1. High cost of latrine 2. Never offered a toilet 3. Not enough water for pit

Cambodia

Indonesia

Lao PDR

Philippines 1. Could not tolerate smell 2. Toilet was donated, came with house or respondent grew up with toilets 3. Embarrassment being seen 4. Pollution 5. Health problems

1. 2. 3. 4.

1. 2. 3. 4. 5. 6. 7. 8.

Saves time and is convenient Cleanliness Comfort Safe, especially at night (women) Everyone now has a toilet, and saw the beneﬁt of neighbors’ toilets Gives greater independence Shy to practice open defecation Visitors can be offered the toilet

1. 2. 3. 4. 5.

1. 2. 3. 4. 5. 6. 7. 8.

Privacy, not being disturbed Convenient (especially during rainy periods) and proximity Clean, sanitary Comfortable Water and environmental protection Safety, avoid snakes and pests Civilized Use of toilet room for shower

1. 2. 3. 4. 5. 6. 7. 8.

High cost of latrine Lack of space or do not own land Never been offered toilet Toilet ruined by ﬂood

Vietnam High cost of latrine Never been offered toilet Lack of space in or near house Not enough water for ﬂushing Habit not to have toilet, and do not care about toilet 6. Have use of someone else’s toilet

Yunnan No space High cost of latrine Never considered having a latrine Limited by location and city planning Toilet has not been provided Have public toilet to use Do not own property Toilet not considered necessary

The intangible aspects of sanitation – decisions around

Second, why have those households without sanitation

sanitation choices and the act of toilet going – were assessed

not yet made the decision to invest in their own toilet facil-

with respect to two key questions covered in the focus group

ity? Speciﬁcally, what factors or constraints prevent a

discussions. First, why should a household own a toilet or

household from getting a toilet, whether of a ﬁnancial or cul-

latrine? Speciﬁcally, what do respondents appreciate

tural nature? Table 2 presents the top ranked factors for

about having an improved toilet facility at their home?

households with and without sanitation, in order of

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

importance. Gender differences in sanitation preferences are also indicated.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Examining the non-ﬁnancial barriers to owning a toilet, the lack of land for building a pit latrine or lack

Many of the same reasons why households have toilets

space in dwelling for a toilet room were cited as impor-

recur across the countries, but often in a different order of

tant reasons, especially in urban areas. Other context-

importance. Time saving and convenience were ranked

speciﬁc constraints include lack of water for ﬂushing,

among the top responses in four countries, but as a lower-

and the risk of ﬂooding due to low-lying land. For families

ranked response in Indonesia, and in the Philippines it

renting their dwelling, it was stated that there was little

was not ranked. Convenience also includes not having to

incentive to invest in a toilet in a dwelling owned by

get up early, not having to leave children unattended, and

someone else. Some respondents did not feel the need

not having to leave the household plot in the rain. The

to have their own toilet as they use someone else’s facility

safety aspect is considered important in ﬁve countries,

or they use a public facility. Other cited reasons include

especially for women and at night-time. Linked to this is

lack of interest in having a toilet (‘never thought about

the privacy aspect (including shyness and embarrassment),

it’), the fact that they are not habituated with latrines, or

ranked as a key consideration to have a toilet in four of

they do not want to use a latrine because they are content

the six countries. Health, hygiene, and sanitary conditions

with their traditional practice. Some respondents had had

were mentioned directly, or more indirectly (e.g. reduces

a bad experience with smelly latrines. The ‘recipient’ men-

number of ﬂies), in all six countries. The aesthetic aspects

tality is echoed in all countries – many respondents stated

were also mentioned (pollution, no smell, environmental

that they do not have a toilet because they have never

protection) in most countries. Having a toilet also gives

been offered one.

some households a sense of pride, including being able to offer visitors a place to use the toilet.

Further building on the reasons to invest in an improved toilet facility, information gathered from the household

The most commonly cited reason why households did

survey interviews indicates that a signiﬁcant proportion of

not have a toilet was the high cost of a latrine. This ﬁnding

households felt in danger while going to the toilet in the

is conﬁrmed by the relatively high cost of even pit latrines

open (at least 24%) or have worried about the safety of

as a proportion of the local wages. For example, the full

their children (at least 31%). It is clear that the majority of

investment costs of an improved dry pit latrine, as a pro-

households are concerned about the dangers of open defeca-

portion of the average GDP per capita (which is a lower

tion. A smaller proportion of respondents had heard about

limit on the average wage in a country) was 21% in Cambo-

animal attacks while practicing open defecation (at least

dia, 4% in Indonesia, 2% in Lao PDR and in the

13%).

Philippines, 11% in Vietnam and 7% in China. These percentages approximate the proportion of a household’s income that would need to be spent on a dry pit latrine, if that house-

DISCUSSION

hold earned the same income as the GDP per capita. These percentages are in fact a lower limit for the lowest income

The key ﬁnding from this study covering six countries in

groups, which are most likely to have a total household

Southeast and East Asia is that sanitation is a highly proﬁt-

(cash) income below the GDP per capita, especially in rural

able social investment. The signiﬁcant amount of evidence

areas. On the other hand, these costs reﬂect the full invest-

collected indicates that sanitation spending brings high

ment costs and not what households themselves spent on

rates of economic return when selected beneﬁts are con-

the options. Subsidies in some sites did reduce the cost to

verted to monetary values. In ﬁve out of six countries,

the household. In some settings, especially communities

BCRs were at least 5 in rural areas and at least 3 in

implementing community-led total sanitation (CLTS) in Cam-

urban areas. The lower average ratios in urban areas is lar-

bodia and Indonesia, a considerable cost was non-ﬁnancial in

gely due to higher unit costs of sanitation interventions, but

nature (e.g. own labor input) which makes a latrine more

also slightly lower absolute health gains due to lower start-

affordable.

ing disease or mortality rates. Inclusion and valuation of

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

sanitation beneﬁts was conservative throughout the analy-

their medium-term sanitation development strategic planning,

sis, while other beneﬁts were omitted altogether from the

which serves as a precondition for funding from the central

BCRs, in particular the social and broader environmental

government and sanitation donors.

impacts such as those on water resources. Hence, BCRs

The second key implication of the study is that the results

presented reﬂect the lower bound of likely economic

of economic analysis should be considered when designing

returns.

and implementing sanitation policies and programs. Higher

There are two main implications of the research that

efﬁciency was observed in larger households, especially

decision makers need to consider. The ﬁrst implication

those with more children – thus making these households a

relates to the ﬁnancing of sanitation, while the second

priority for public programs. On the other hand, a decline

relates to the loss in efﬁciency experienced when taking

in efﬁciency was observed in all project sites due to either

into account observed sanitation utilization rates.

non-adoption by the household of the technology on offer,

It becomes apparent from sectoral analyses that current

or else a gradual decline in performance over time due to dis-

spending on sanitation is insufﬁcient to provide universal

repair or unsustained behavior. While these observations do

coverage in the medium-term, for example the next 5

not come as a surprise, the scale of the impact on efﬁciency is

years. Therefore, further funds need to be allocated to sani-

sufﬁcient for policy makers to take note.

tation. Funds from governments, donors and households

Low initial uptake has several interlinked causes. First

can be sourced either by giving sanitation greater priority

and foremost, the unaffordability of the offered sanitation

over other programs – hence increasing overall efﬁciency

options is a key reason why households decide to remain

of spending – or funds can be raised by taking loans to

without improved sanitation. In the urban site of Cambo-

pay for sanitation investments. A further source of funds is

dia, households chose not to pay the connection fee to

from private stakeholders (either ﬁnanciers or providers)

the sewerage because there was no legal obligation and

who need to recognize that sanitation offers signiﬁcant –

they did not see the beneﬁt to themselves of doing so.

but as yet largely unexploited – business opportunities.

This led to very low capacity utilization of the sewerage

In order to attract more resources to sanitation, ﬁnan-

network and the wastewater treatment plant, hence

ciers need to be conﬁdent that the funds will be put to

increasing the cost per household served to ﬁve times

good use. Hence, a ﬁrst step is to draw up sanitation plans,

what it would have been if operating at its design capacity.

detailing the costs of achieving sanitation scale-up and the

Households may not adopt sanitation because they are not

expected realistic ﬁnancing sources. Decisions about how

sufﬁciently aware of the negative impacts of poor sani-

to invest in sanitation need to be made based on evidence

tation, or there is insufﬁcient social mobilization for them

of the comparative returns on different options, such as

to change their behavior.

those presented here. A dialogue needs to be started with a

A decline in efﬁciency over time was observed, which

range of ﬁnanciers including non-traditional ﬁnanciers of

was found to be caused by households not sustaining

sanitation programs, such as ministries of tourism and agri-

improved behaviors either because of force of habit and

culture, private sector investment funds and corporate

lack of conviction, or because the technology itself had

social responsibility (CSR) programs of the private sector.

stopped functioning. In Cambodia, the CLTS approach

The example of Indonesia shows that important advocacy

had a high initial uptake with most households building

can be provided by local government leaders, such as regents

their own latrines, but it was not sustained over time

and mayors. An alliance of districts and cities has been

because the pit latrines collapsed or they quickly became

formed which includes sanitation as one of their development

too ﬁlthy and unhygienic to use. In Indonesia and Lao

mainstreams. Since 2010, the Central Government has been

PDR, some population groups did not sustain behavior

implementing the ‘Acceleration of Residential Sanitation

change for long because they were content with the tra-

Development Program’ (Program Percepatan Pembangunan

ditional practice of open defecation. Furthermore, in rural

Sanitasi Permukiman, known as the PPSP). The program pro-

areas, people who work in the ﬁelds are unlikely to travel

vides technical assistance to local governments to prepare

back to their home to access their sanitation facility.

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

In some sites, lack of consultation and supply of inap-

Decisions are largely cost-based – i.e. what can be afforded

propriate technologies was responsible for non-sustained

with a budget. However, this study has clearly demonstrated

use of sanitation systems. For example, in the Philippines,

that there is signiﬁcant economic value – and to a lesser

7% of households in one site (Bayawan) with a sewerage

extent, ﬁnancial value – in delivering improved sanitation ser-

system answered that they have insufﬁcient household

vices and that efﬁciency varies considerably between the

water supply for ﬂushing their toilet; and over 50% of house-

options. Alternative means of recovering sanitation costs

holds in the same site said that they were not offered any

need to be explored to make the delivery of these demanded

other sanitation options. Across all sites in the Philippines,

services a practical reality. To make this happen, decision

20% of households answered that they experience regular

makers need to be better versed in economic and ﬁnancial

pit ﬂooding, hence making it unusable for certain periods

analysis, the evidence needs to be made more available, and

of the year. For these reasons, many unimproved sanitation

decision makers across different departments or jurisdictions

practices were found to exist, even in sites with sanitation

should work together to solve common problems.

projects. In particular, a high proportion of households

Third, ﬁnancial innovations are needed to promote a

saw children regularly defecating in the open, thus exposing

sustainable market-based supply of sanitation services,

a public perception that children do not need to conform to

including mechanisms to reach the most needy. The study

the sanitation behavior of adults. Also, urination in open

identiﬁed that the upfront costs of sanitation hardware con-

areas and public spaces was still particularly prevalent for

stitute an important share of the annual income of the

men, in both rural and urban areas.

unserved population, who tend to be in the lower income groups. Hence, few households are able to pay upfront the full cost of a facility. On the other hand, households are

RECOMMENDATIONS

usually able and willing to make smaller but more regular payments. These households will be successfully reached

Five recommendations are proposed based on the ﬁndings of

with an appropriate combination of smart subsidies, loans

this multi-country study on the health and economic returns

and lower cost sanitation options. The CLTS movement lar-

of improved sanitation. First, sanitation should be promoted

gely relies on the third of these, while other demand-led

as a central development priority. The evidence generated in

approaches also include a subsidy element to encourage

this study demonstrates the importance of improved sani-

households to invest themselves. However, micro-ﬁnance

tation for a number of development outcomes and provides

and bank or government loans for sanitation are still not

arguments to be made in a variety of forms and to a variety

common.

of audiences. While the responsibility for sanitation policies

Fourth, to capture the full beneﬁts of sanitation pro-

and programs may fall on one or two government ministries,

grams, further attention is needed to improve program

improving sanitation is in the interest of all. Line ministries or

design and implementation. This study showed that there

government departments that need to buy into sanitation, in

is a significant drop in efficiency and benefits actually

particular, are those responsible for education (both schools

received by households and the communities, because the

and higher education establishments), health, workplaces,

wrong technology is chosen, solutions are delivered inap-

water resources management and regulation, municipal ser-

propriately and individual and community practices are

vices, rural development and tourism.

not sustained. Hence, sanitation programs need to be

Second, evidence-based decision-making should be pro-

more people-centered: in other words, they need to be

moted, speciﬁcally drawing on economic analyses as well as

demand driven and involve everyone, not just traditional

other key decision factors (such as budget or resources avail-

social structures. Public decision makers and private suppli-

able, health impacts and cost recovery options). Until now,

ers should make available different sanitation options, so

very few previous decisions on sanitation spending in the six

that individuals and communities can weigh up the costs

countries have been made based on an explicit comparison

and beneﬁts themselves, and they can be supported in this

of the costs and beneﬁts of the alternative technology options.

decision making process.

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Fifth, the evidence base needs to be built to support evi-

Stephanie Cohen, Frederic De San, Vilaythong Chantha-

dence-based decision-making. ‘Evidence’ needs to be

linh, Viengsamay Vongkhamsao and Viengsompasong

understood in its broadest form; it includes not only focused

Inthavong (Lao PDR); Nelissa Jamora, Jeremy Ockelford,

research but also improved (and where possible, standar-

Dieldre Harder, EdKarl Galing, Leila Elvas, Shiela del

dized) monitoring and evaluation of sanitation programs

Torres and Emma Boucart (Philippines); Hoang Thuy

and routine information systems. Where funds are available,

Lan, Phan Huyen Dan, Nguyen Hoang Yen, Le Thu

impact evaluations of projects or programs are highly valu-

Hoa, Le Thu Hang, Bui Nhung, Nguyen Huu Tuyen,

able for understanding the actual impact and determinants

Nguyen Phuong Thao, Tran Kong Khanh, Hoang Van

of impact, in real-life conditions – and are hence key for iden-

Minh, Nguyen Diem Hang and Hoang Anh Nga (Viet-

tifying factors of success and reasons for failure or sub-

nam); Zhou Bo, Yang Liqiong, Zhang Pu, Fang Jinmin,

optimal performance. Further economic research is needed

Dong Di, Dong Lin, Luo Ronghuai, Zhang Tiwei, Zou

on the value of time, the value of life, business opportunities

Yahui, Wang Jing, Zhouzheng Yuxiao and Linda Shi

(e.g. value-chain assessments, willingness to pay surveys),

(Yunnan Province, China); Yosa Yuliarsa and Irvan Tjon-

and the welfare value of hard-to-monetize impacts.

dronegoro (regional inputs). The research was ﬁnanced by the Swedish International Development Agency; the Asian Development Bank; and

COMPETING INTERESTS

the ECO-Asia project of the United States Agency for International Development (USAID).

The authors declare that they have no competing interests.

DISCLAIMER AUTHORS’ CONTRIBUTIONS GH designed the study, recruited the country teams, guided the country studies, compiled and analyzed the data, and drafted and ﬁnalized the manuscript. PR, KP,

The ﬁndings, interpretations, and conclusions expressed herein are entirely those of the authors and should not be attributed to the organizations they represent.

ASW, VAN, and LC lead their respective country teams, ﬁne-tuned the study design, implemented and guided the research, and drafted the country reports. IB and ALW guided the study, recruited the study team and provided overall management. All authors read and approved the ﬁnal manuscript.

ACKNOWLEDGEMENTS The study was led by the World Bank’s Water and Sanitation Program (WSP), with the contribution of WSP teams and consultants in each of the participating countries, as follows: Sam Sok Heng, Pok Kong Chheng and Phalla Yin (Cambodia); Oktarinda, Edi Purnomo, Koderi Hadiwardoyo, Indon Merdykasari, Takdir Nurmadi, Bert Bruinsma, Dedek Gunawan, Dadang Fadilah and

Martin

Albrecht

(Indonesia);

Alan

Boatman,

REFERENCES Barreto, M., Genser, B., Strina, A., Teixeira, M., Assis, A., Rego, R., Teles, C., Prado, M., Matos, S., Alcântara-Neves, N. & Cairncross, S.  Impact of a citywide sanitation program in Northeast Brazil on intestinal parasites infection in young children. Environ. Health Persp. 118 (11), 1637–1642. Curry, S. & Weiss, J.  Project Analysis in Developing Countries. Macmillan, London. Drummond, M. F., Sculpher, M. J., Torrance, G. W., O’Brien, B. & Stoddart, G. L.  Methods for the Economic Evaluation of Health Care Programmes, 3rd edn. Oxford University Press, Oxford and New York. Fox-Rushby, J. & Hanson, K.  Calculating and presenting disability-adjusted life-years (DALYs) in cost-effectiveness analysis. Health Policy Plan 16 (3), 326–331. Hanley, N. & Spash, C.  Cost-Beneﬁt Analysis and the Environment. Edward Elgar, Cheltenham, UK. Hutton, G.  Global Costs and Beneﬁts of Drinking-water Supply and Sanitation Interventions to Reach the MDG

G. Hutton et al.

Economic efﬁciency of sanitation interventions in Southeast Asia

Target and Universal Coverage. Report No. WHO/HSE/ WSH/12.01. World Health Organization, Geneva. Hutton, G., Haller, L. & Bartram, J.  Global cost-beneﬁt analysis of water supply and sanitation interventions. J. Water Health 5 (4), 481–502. Hutton, G., Larsen, B., Leebouapao, L. & Voladet, S.  Economic Impacts of Sanitation in Lao PDR. World Bank, Water and Sanitation Program, Washington DC. Hutton, G., Rodriguez, U., Napitupulu, L., Thang, P. & Kov, P.  Economic Impacts of Sanitation in Southeast Asia. World Bank, Water and Sanitation Program, Washington DC. Moraes, L., Cancio, J., Cairncross, S. & Huttly, S.  Impact of drainage and sewerage on diarrhoea in poor urban areas in Salvador, Brazil. Trans. R. Soc. Trop. Med. Hyg. 97 (2), 153–158. Murray, C., Lopez, A., Mathers, C. & Stein, C.  The global burden of disease 2000 project: Aims, methods and data sources. Research Paper, WHO, 1–22. Prüss, A., Kay, D., Fewtrell, L. & Bartram, J.  Estimating the global burden of disease from water, sanitation, and hygiene at the global level. Environ. Health Perspect. 110 (5), 537–542. Sachs, J.  Macroeconomics and Health: Investing in Health for Economic Development. Report of the Commission on Macroeconomics and Health. World Health Organization, Geneva.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Sugden, R. & Williams, A.  Principles of Practical Cost-Benefit Analysis. Oxford University Press, Oxford. United Nations  Resolution adopted by the General Assembly. 64th Session. A/RES/64/292. Van Minh, H. & Nguyen-Viet, H.  Economic aspects of sanitation in developing countries. Environ. Health Insights 5, 63–70. Waddington, H., Snilstveit, B., White, H. & Fewtrell, L.  Water, sanitation and hygiene interventions to combat childhood diarrhea in developing countries. Int. Initiative for Impact Eval. (3ie). Available from: http://zunia.org/sites/ default/files/media/node-files/3i/179502_3ie%20SR00112 49554991.pdf. World Health Organization  Life tables for 191 countries for 2000: data, methods, results. GPE Discussion Paper 40. World Health Organization  Global Health risks. Mortality and Burden of Disease Attributable to Selected Major Risks. WHO, Geneva. World Health Organization  Global Status Report on Noncommunicable Diseases 2010. WHO, Geneva. ISBN: 97892 41564229. 176 p. World Health Organization  UN-Water Global Level Annual Assessment of Sanitation and Drinking-Water. The Challenge of Extending and Sustaining Services. WHO, Geneva.

First received 20 March 2013; accepted in revised form 2 July 2013. Available online 17 August 2013

04.1

2014

Removal of selected microorganisms using silver-impregnated and coated, low-cost, micro-porous, ceramic water ﬁlters Jean Simonis, Muzi Ndwandwe, Albert Basson and Tlou Selepe

ABSTRACT There is great need to purify the contaminated water which the poor people in Africa have access to, and make it safe for drinking in a way that is affordable and effective. A particular challenge is the removal of pathogenic bacteria and viruses, which traditionally are eliminated by expensive nanofiltration or reverse osmosis. An added requirement is satisfying the recent recommendation of the WHO for household water-treatment systems to eliminate 99.99% of microbial contamination, which is proving exceptionally difficult to achieve in poor countries at a cost they can afford. We report on the successful testing of a low-cost, locally produced ceramic filter that has the potential to meet the WHO criterion at a cost of US$10 per year. In one version the filter consisted of a silver-impregnated, highly porous ceramic; in another modification silver nano particles were incorporated on the ceramic surface. The silver-impregnated filter was tested on water samples contaminated with selected Gram negative bacteria: Escherichia coli, Proteus mirabilis and Pseudomonas aeruginosa, for its oligodynamic effect and for its effective reduction of bacteriophages. The ceramic filters reduced the viral count by 94–99% and we believe that, with further development, our prototype is

Jean Simonis (corresponding author) Department of Hydrology, University of Zululand, Private Bag X1001, KwaDlangezwa, South Africa E-mail: simonisj@unizulu.ac.za Muzi Ndwandwe Dean of Science, Department of Physics, University of Zululand, Private Bag X1001, KwaDlangezwa, South Africa Albert Basson Tlou Selepe Department of Microbiology and Biochemistry, University of Zululand, Private Bag X1001, KwaDlangezwa, South Africa

easily capable of achieving the WHO criterion. Key words

| ceramic water ﬁlter, drinking water, oligodynamic, virus, water treatment

INTRODUCTION Various physical and chemical methods are available for

and spatial variability with respect to protozoa, bacteria and

domestic water treatment (WHO ). In Africa, however,

viruses distribution in countries where the ﬁlters are

ceramic ﬁlters used in households are seen as a superior,

produced.

more cost-effective puriﬁcation method when using raw

The removal of pathogenic viruses from polluted drink-

water and poorly disinfected drinking water when compared

ing water is technologically challenging. Figure 1 shows the

with conventional centralised treatment and water reticula-

various methods of purifying domestic water, and their effec-

tion systems (Lantagne ; Clasen & Boisson ). The

tiveness. The small particle size of viruses (25–200 nm)

small pore size of ceramic ﬁlters normally makes them effec-

makes it possible to remove them using expensive nano-ﬁl-

tive in removing bacteria and protozoa but not viruses,

tration and reverse osmosis, which are not suitable or

which are much smaller and pass through the ﬁlters. Fur-

affordable on a domestic scale.

thermore, the WHO () is applying increasing pressure

Bacteriophages usually do not travel independently in

on the manufacturers of ceramic ﬁlters to comply with

water, however, but ‘hitch a ride’ on larger host bacteria.

their recommended level of microbial reduction of 99.99%

These larger hosts, to which the virus is attached, may

based on representative monitoring data from local, regional

then be removed together with the viruses, either by ﬁl-

or national surveys. Such data should highlight the temporal

tration or by their inactivation through incorporation of

doi: 10.2166/washdev.2013.027

J. Simonis et al.

Figure 1

Improving a low-cost ceramic water ﬁlter for microbial removal

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Effectiveness of different methods to reduce polluted drinking water contaminated with viruses. Effectiveness is deﬁned in terms of log10 reduction (LRV), as follows: LRV1 (reduction by 90%); LRV 2 (reduction by 99%); LRV3 (reduction by 99.9%); etc. The WHO now recommends a disinfection standard of LRV 4 (reduction equal to or greater than 99.99%).

metallic ions into their enzyme system in a chemical reac-

of the coatings have achieved the WHO () disinfection

tion, or by a combination of the two methods.

standard of LRV 4 and for this reason we selected metallic

In the study reported here we combined low-cost, ceramic ﬁltration to trap a host with chemical disinfection

silver and silver oxide, with its electropositive properties, as the primary anti-viral agent.

technology (silver) to inactivate the viruses in infected drinking water.

Physical testing For the physical testing of the samples the pore structure of

EXPERIMENTAL PROCEDURE

the ceramic ﬁlter, which had been produced by sintering, was determined in three complementary ways: scanning elec-

Ceramic ﬁlter preparation and characterisation

tron microscopy (SEM), hydraulic conductivity and water

A locally developed, low-cost, micro-porous ceramic water filter (OUTBAC; out bacteria), manufactured using a traditional slip casting process, was employed to trap and inactivate the viral host. This filter is more effective and efficient than that of other low-cost producers as a result of using less raw material, lower energy consumption and expenditure (Simonis & Basson , ). The ceramic filter production process uses a slip casting process which requires a minimum amount of capital investment for manufacture and is therefore ideally suited for manufacture in rural Africa (Simonis & Basson ). The ceramic filter material used in this

immersion (Lambe ; ASTM C20 - 00 ). SEM demonstrated the bi-modal nature of the pore former, with larger pores resulting from the pore former after ﬁring and the smaller pores resulting from the necks (contact areas) between the larger pores. The ‘necking’ traps the bacteria and viruses and therefore becomes the determining factor for viral protection. Water immersion demonstrated the high apparent porosity and water absorption values of the ceramic. A standard testing protocol for porous ceramics was followed (found in the Supplemental Material available online at http://www. iwaponline.com/washdev/004/027.pdf).

study was obtained from this production process. The ﬁlters were cut into 25 mm × 25 mm squares for physical

characterisation

and

coating.

According

Silver coating of ceramic: sputter coating

Wistreich (), Naegeli in 1893 observed that metallic

The sputtering system chamber consisted of platens, which

silver kills green algae in highly diluted aqueous suspension.

both rotate and are water cooled, and magnetrons that are

Various substrate coatings on ceramic ﬁlters such as iron

planar (three off-axis). The ﬁlter samples were silver-coated

and aluminium hydroxide have been used by other authors

at room temperature in a vacuum enclosure pumped to

which also increased the inactivation of viruses in aqueous

1.3 × 10–4 Pa. The distance between the object and magne-

media from less than 30% without coating, to more than

tron is adjustable for obtaining a homogeneous coating

80% effectiveness with a coating (Farrah et al. ). None

and rate of deposition.

J. Simonis et al.

Improving a low-cost ceramic water ﬁlter for microbial removal

The ceramic surfaces were sputter coated at room temp-

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

the preparation over the silver-coated ceramic was as

erature using a silver metal target. The procedure used for

follows:

coating was as follows: The chamber was cleaned with

1. The silver coated 25 mm × 25 mm square ceramic

acetone to reduce contamination. A silver target was set into the DC magnetron and two ceramic samples together with one silicone substrate were loaded. Vacuum was applied for a period in excess of 2 hours (pump down). Argon gas was applied to the chamber. The magnetron was powered at 80 W for 60 s at a deposition pressure of 3 × 103 Pa. The chamber was then de-pressurised with air and the magnetron powered off.

samples were placed aseptically into sterile Petri dishes. 2. 10 ml sterile melted nutrient agar (Merck Cat. No. W

1.05450.0500) was prepared and cooled to 50 C. 3. Inoculate the cooled, sterile nutrient agar with 200 μl of E. coli culture. 4. The silver-coated ceramic was aseptically placed in a sterile Petri dish and the inoculated nutrient agar was gently poured into the Petri dish. W

5. The agar was left to solidify, and then incubated at 37 C

Silver and silver oxide impregnation of ceramic: wet chemical impregnation

for 24 h. 6. Steps 1 to 4 were repeated for Proteus mirabilis and

A second method of introducing silver into the porous ﬁlter material involved wet chemistry. The untreated ceramic ﬁlter units were immersed in a 1 mg l–1 silver nitrate solution for 30 min (Van Halen et al. ). They were then removed W

from the solution, air dried and heated to 600 C in an oxidising or a reducing environment, to produce a superﬁcial deposit of Ag2O or Agþ on the ceramic material, respectively (Wegmann et al. ). Testing treated ﬁlters against selected microorganisms Subsequent to the coating and impregnation, prototypes of

Pseudomonas aeruginosa. 7. After incubation the zones of inhibition were measured in mm and recorded. 8. The silver coated ceramic’s layer thickness and particle size were inspected using an atomic force microscope (AFM). Test 2: The plaque assay method for bacteriophages Phage suspensions were prepared for the ﬁltration tests with the uncoated, silver-coated and silver impregnated ﬁlters. 1. Somatic phages (host Salmonella typhimurium) and F

the ceramic ﬁlters were ﬁrst tested for their oligodynamic

speciﬁc phages (host Escherichia coli) were used for

action in susceptibility tests with selected microorganisms,

pre- and post-ﬁltration with the uncoated, silver-coated

Escherichia coli as fecal indicator and selected Gram-negative

bacteria

(Muller

),

Proteus

mirabilis

and

Pseudomonas aeruginosa (Test 1). This was then followed with Test 2 filtration test using pre- (assay challenge water) and post-filtrate (filtered through the silver coated and impregnated ceramic filters) testing using bacteriophages from hosts Escherichia coli and Salmonella typhimurium

and silver impregnated ﬁlters. 2. After growth, the phages were centrifuged and stored in W

McCartney bottles at 5 C. 3. The phage suspensions were filtered through the sputtercoated ceramic filter and the filtrate collected. 4. The plaque concentration before and after filtration was determined using the double-agar method.

(Clokie & Kropinski ). Test 1: Susceptibility of bacteria to silver-impregnated

RESULTS

and coated ﬁlters Physical testing The pour plate technique was used to distribute three bacterial cultures throughout the agar trapped in position as

From a previous study (Simonis & Basson ) on the same

the medium hardened. The procedure followed for pouring

ﬁlter it was demonstrated that the ceramic ﬁlters had a

J. Simonis et al.

Improving a low-cost ceramic water ﬁlter for microbial removal

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

bi-modal pore size. The smaller pores had a pore size of 0.3

The results complement those for the hydraulic conductivity

micron and the larger pores a size of approximately 5

and can be understood in relation to the SEM images.

micron. The bi-modal pore size distribution suggests why these ﬁlters act as such an effective micro-sieve in ﬁltering

Microbial testing

out microbes and other suspended material from polluted drinking water. The two pore sizes are clearly visible

Oligodynamic action by treated ﬁlters

in the SEM images in Figure 2 where the larger pores predominate but with the smaller pores still visible within

An image of the homogeneous silver coating using the AFM

the matrix.

method is shown in Figure 3. Table 3 shows the results of the inhibitory actions of the silver nano coating using the pour

Hydraulic conductivity

plate technique. Both the Agþ and Ag2O are shown to be effective in relation to the magnitude of the zones of inhi-

Table 1 summarises the ﬂow properties (K values) for the

bition for the microbes tested.

porous ﬁlter previously tested (Lambe ; Simonis & Basson ) using different pore former concentrations for

Elimination of test phages by treated ﬁlters

increased porosity. The nano coated ﬁlters gave slightly lower K values. The medium K was selected for the ﬁltration

Both coated and uncoated ﬁlters reduced somatic phage in

testing as it produced a ﬁltration rate of 2–4 litres per hour,

the test ﬁltrate by 94–99% (Table 4). F phage was relatively

double the volume of water compared with the PfP ﬁlter

poorly eliminated by silver-impregnated ﬁlters (22–46%),

(Potters for Peace water ﬁlter: Van Halen et al. ). This

whereas the sputter-coated ﬁlter reduced the F phage test

makes the ﬁlter suitable for supplying sufﬁcient quantity of

sample by a satisfactory 97%.

purified water for the daily needs of a family of five. The high apparent porosity and water absorption capacity of the OUTBAC filter is indicated in Table 2.

Figure 2

Water passing through the ultra-porous, ultra-ﬁne ceramic

pores

slowed

the

bacteriophages

down

for

inactivation by incorporation of metallic ions into their

SEM images of the OUTBAC ﬁlter surface showing the homogeneous pore size distributions of the larger and smaller pores with d50 (average pore size) of 5 and 0.3 micron, respectively.

J. Simonis et al.

Table 1

Improving a low-cost ceramic water ﬁlter for microbial removal

Hydraulic conductivity (K) values for the porous uncoated and coated ﬁlters (Simonis & Basson 2011, 2012)

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

enzyme system. The uncoated ceramic was not effective in eliminating viruses and therefore needs the combined effec-

Relative conductivity

K (cm s 1)

High

> 10 1

Medium

10 1 to 10 3 3

tiveness of the silver ions.

Coating

Uncoated

Low

Very low

10 5 to 10 7

Practically impermeable

< 10 7

to 10

CONCLUSIONS

Coated

The viral load of contaminated water was greatly reduced using silver-coated (97–98%) and slightly to signiﬁcantly reduced for silver-impregnated ceramic ﬁlters (22–95%).

Table 2

Porosity and water absorption of the OUTBAC ﬁlter (Simonis & Basson 2011, 2012)

The oligodynamic effects of silver-coated ceramic ﬁlters have shown signiﬁcant zones of inhibition for Escherichia

Coating

AP (%)*

WA (%)

ASG (g cm

Uncoated

2.42 (± 0.01)

Coated

109

2.44 (± 0.05)

)

coli, Proteus mirabilis and Pseudomonas aeruginosa. The efficiency of the Agþ sputter-coated filters was dramatically increased in the F and somatic phage filtration tests, showing 97% and 99% improvement, respectively.

* Sample size (n = 10).

The efficiency of incorporating Agþ as active surfaces in the porous ceramic using a wet chemical impregnated process proved to be less effective for Agþ, with the somatic phage, showing 94% reduction in viral load and the F phage demonstrating only a 46% reduction. The incorporation of Ag2O as active surfaces in the porous ceramic resulted in the lowest reduction of the F phage but much improved somatic phage reduction of 95%. Both the sputter coated and impregnated silver (Agþ), with the additional requirement of being fired under a reducing environment, proved to be more effective than silver oxide as an active surface coating agent in the porous ceramic. The tested phage filtration results compared well with other low-cost filter products but in comparison are closer to meeting the WHO () recommendations.

Figure 3

AFM image of the silver-coated ceramic surface where the diameters of the nano-particles vary between 40 and 50 nm.

RECOMMENDATIONS Table 3

Oligodynamic action of silver-coated ceramic

We believe that changing the nano-particle diameter and increasing the thickness of the silver coating could further

Diameter of zones developed (mm)

increase the effectiveness, bringing it in line with the

Wet chemical

WHO () recommendations for viral protection.

impregnated Organism

E. coli P. mirabilis P. aeruginosa

Sputter coated Agþ

Agþ

Ag20

Due to the high surface area of the ceramic ﬁlter (5 m g 1) it is believed that increasing the silver impregnation 2

concentration from 1 to 2 mg l–1 could further improve the effectiveness, bringing it in line with the WHO () recommendations for viral protection (Mpenyana-Monyatsi

J. Simonis et al.

Table 4

Improving a low-cost ceramic water ﬁlter for microbial removal

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Reduction in bacteriophages for the coated and impregnated ceramic ﬁlter Plaque concentration (per ml of ﬁltrate) and reduction (%) Sputter coated Uncoated

Agþ

Ag2O

2.5 × 105

2.3 × 105

6.8 × 104

1.4 × 105

2.0 × 105

100

130

Host

Phage

Initial

E. coli

S. typhimurium

Somatic

2.1 × 10

2.0 × 10

et al. ). Wet impregnation is a much easier and less þ

expensive way of incorporating Ag

as active surfaces in

the porous ceramic compared with sputter coating the outside

surface

only,

which

Wet chemical impregnated

requires

expensive

and

sophisticated equipment. The results suggest that nano-silver coated ceramic ﬁlters could potentially be used as a low-cost, householdtype water ﬁlter, for microbial removal. They also have the advantage of avoiding harmful chemical by-products resulting from the use of chemicals to disinfect water.

REFERENCES ASTM C20 - 00  Standard Test Methods for Apparent Porosity, Water Absorption, Apparent Specific Gravity, and Bulk Density of Burned Refractory Brick and Shapes by Boiling Water. Annual Book of ASTM Standards. ASTM, Philadelphia. Clasen, T. & Boisson, S.  Household-based ceramic water filters for the treatment of drinking water in disaster response: and assessment of a pilot programme in the Dominican Republic. Water Practice Technol. 1 (2) doi:10.2166/WPT.2006031. Clokie, M. R. & Kropinski, A. M.  Bacteriophages: Methods and Protocols, Vol. 1: Isolation, characterisation and interaction, vol. 501. Farrah, S. R., Preston, D. R., Toranzos, G. A. & Preston, D. R.  Use of modified diatomaceous earth for removal and recovery of viruses in water. Appl. Environ. Microbiol. 57 (9), 2502–2506. Lambe, T. W.  Soil Testing for Engineers. John Wiley & Sons, New York, pp. 58–62. Lantagne, D. S.  Investigation of Potters for Peace colloidal silver impregnated ceramic filter Report 1: Intrinsic effectiveness. USAID P.O. 524-0-00-01-00014-5362.

Muller, H. E.  Oligodynamic action of 17 different metals on Bacillus subtilis, Enterobacteriaceae, Legionellaceae, Micrococcaceae and Pseudomonas aeruginosa. Zentralbl. Bakteriol. Mikrobiol. Hyg. B 182, 95–101. Mpenyana-Monyatsi, L., Mthombeni, N. H., Onyango, M. S. & Momba, M. N. B.  Cost-effective filter materials coated with silver nanoparticles for the removal of pathogenic bacteria in groundwater. Int. J. Environ. Res. Public Health 9 (1), 244–271. Simonis, J. J. & Basson, A. K.  Evaluation of a low-cost ceramic micro-porous filter for elimination of common disease microorganisms. Phys. Chem. Earth 36, 1129–1134. Simonis, J. J. & Basson, A. K.  Manufacturing a low-cost ceramic micro porous filter and filter system for elimination of common pathogenic bacteria. Phys. Chem. Earth 50–52, 269–276. Simonis, J. J. & Basson, A. K.  Manufacture of a low-cost ceramic microporous filter for the elimination of microorganisms causing common diseases. J. Water, Sanitation Hyg. Dev. 03.1, 42–50. Van Halen, D., van der Laan, H., Heijman, S. G. J., van Dijk, J. C. & Amy, G. L.  Assessing the sustainability of the silver-impregnated ceramic pot filter for low-cost household drinking water treatment. Phys. Chem. Earth 34, 36–42. Wegmann, M., Michen, B. & Graule, T.  Nanostructured surface modification of microporous ceramics for efficient virus filtration. J. Eur. Ceram. Soc. 28, 1603–1612. WHO  Evaluating Household Water Treatment Options: Health-Based Targets and Microbiological Performance Specifications, 1st edn. World Health Organization, Geneva (November 2011). WHO  Managing water in home: accelerated health gains from improved water supply. Available from: http://www. who.int/water_sanitation_health/dwq/wsh0207/en/. Accessed 25 September 2013. Wistreich, G. A.  Microbiology Laboratory: Fundamentals and Applications. Prentice Hall, New Jersey.

First received 31 January 2013; accepted in revised form 21 August 2013. Available online 21 October 2013

04.1

2014

When water is scarce: the perception of water quality and effects on the vulnerable Hamdiyah Alhassan and Paul Adjei Kwakwa

ABSTRACT The world has over 1 billion people without access to safe drinking water and it is expected that the number of people living in water-stressed or water-scarce places will increase to 3.4 billion by 2025 and one cannot ignore the effect such a situation will have on the vulnerable groups. The current study among other things examines household water insecurity and assesses its impact on vulnerable groups especially women and children in Dungu, a rural community in the northern region of Ghana. Relying on 125 respondents sampled, the findings show that the main source of drinking water in the community, a dam, fails to fulfil the household water security situation with respect to quantity, quality, reliability and accessibility. Because of the unwholesome nature of the water, the majority of the respondents treat water by using cloth filtering, boiling and the use of alum.

Hamdiyah Alhassan Department of Agricultural and Resource Economics, University for Development Studies, Tamale, Ghana Paul Adjei Kwakwa (corresponding author) Faculty of Development Studies, Presbyterian University College Ghana, Akuapem Campus, Akropong-Akuapem, Eastern Region, Ghana E-mail: pauladkwa@yahoo.com

Contentiously, the insecurity of water in the community has numerous negative effects ranging from health to economic effects on women and children who have to walk for 2 km in search of water and this could be reduced with the availability of a good quality water source close to home. Key words

| Ghana, household, vulnerable, water insecurity, water management

INTRODUCTION Water is a very important resource needed to sustain life,

According to the UN, the number of people living in

and safe drinking water is a fundamental requirement for

countries deﬁned as water-stressed or water-scarce (a situ-

human life (UNCHR et al. ). According to Ariyabandu

ation where available water is unable to meet the demand

(), easy accessibility, reliability and timely availability

for speciﬁc purposes) is expected to increase from the

of adequate safe water to satisfy basic human needs

2005 ﬁgure of over a billion to 3.4 billion people by 2025

ensures water security (Asare ). Safe drinking water

(UN ). In Africa it is estimated that 300 million out of

for human consumption, according to the World Health

the 800 million people on the continent live in a water-

Organization (), should be free from pathogenic bac-

scarce environment (Donkor ). This situation can be

teria, viruses and parasites, chemical and radiological

attributed to a number of reasons such as infrastructural pro-

hazards, and it must also be acceptable in appearance,

blems (Newman ), high population and urbanization

taste and odour.

(Roudi-Fahimi et al. ), ‘land grabbing’ – a situation

As of the end of 2011, 89% of the world population used

where countries in sub-Saharan Africa and foreign compa-

an improved drinking water source (WHO/UNICEF JMP

nies enter into a deal involving the sale of rights to inland

) as against the 1.1 billion people without access to

water resources (Freitas ) – and the rising global

safe drinking water in 2002 (WHO ). Despite this

temperature that accompanies climate change which

achievement, about 768 million people, including 185

ultimately has affected Africa’s water situation through

million people who relied on surface water to meet their

changes in the frequency and distribution of rainfall,

daily drinking water needs, did not use an improved

drying-up of rivers and receding of water bodies, among

source (WHO/UNICEF JMP ).

others (Urama & Ozor ).

doi: 10.2166/washdev.2013.140

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The quantity of water collected and used by households

dry harmattan winds from November to February and

has an important inﬂuence on health; WHO has noted that

high sunshine from March to May. With a population of

a person needs between 50 and 100 litres of water per day to

1,769, consisting of 771 males and 998 females, Dungu is

meet basic needs (UNCHR et al. ). The inadequacy of

one of the water-scarce areas in the region with a dam as

rural water supply and sanitation coverage has been recog-

the major source of water. The only standpipe in the area

nized as a major contributor to human disease and

has not been functional for some time. The common

malnutrition in many developing countries including

reason offered to explain this by the appropriate authorities

Ghana. For example, diseases like cholera, typhoid, river

is that the Dungu community is situated on high land, so

blindness, hepatitis, shigellosis and malaria are either

when the water is pumped from the treatment plant, it

water-borne or water-related (Asare ) and the Inter-

does not get there. The residents therefore refer to the

national Fact-Finding Mission on Water Sector Reform in

non-functional standpipe as a waste and for that matter a

Ghana (IFFM) (2002) has estimated that about 70% of dis-

‘stick’. The women here are engaged in petty trading

eases in Ghana are the result of inadequate rural water

while the men are mostly farmers. The commonest crops

supply and sanitation coverage (Asare ).

produced by these peasant farmers during the rainy

Thus, living in an area where potable water is lacking comes with a lot of consequences including the reliance

season are yam, maize, millet and groundnut, which they consume, selling the surplus to meet other basic needs.

on other water sources that may even be (perceived as)

Studies on the state of water supply in Ghana have been

harmful to the users. In some cases, communities receive

carried out with varying foci. For instance Tayeh et al. ()

the services of authorized bodies that test the water quality

looked at drinking water sources and water-related diseases

of their alternative water sources and proffer advice on

such as Guinea worm. Asante () inspected the socioeco-

how to render the water suitable for human consumption.

nomic aspects of access to safe drinking water (Nketiah-

Other communities that may not be so lucky in having

Amponsah et al. ). Again, Akumiah & Rydhagen ()

such tests done for them would, based on their own percep-

examined the relationship between water management and

tions, devise strategies for dealing with the harmful nature

health, and how effective water management through full

of the water. As a result studies need to be carried out to

community participation could help provide adequate safe

reveal how households who perceive their water to be

drinking water, while Arku & Arku () discussed how

unwholesome deal with that. Again the effects of scarce

drought affects the livelihoods of agro-based rural commu-

water are likely to be severe on women and children who

nities in the Volta Region of Ghana, and the survival

attend to water issues in households due to their vulner-

strategies of rural women and men during drought periods.

ability, especially in developing countries. Studies that

Again, a few studies such as Addo () illustrated the effects

would reveal such effects are therefore imperative. How-

of inadequate water supply on the female child; Arku & Arku

ever, studies that have looked at these issues compared

() shed light on the effect of drought on women; and Doe

with general studies on water scarcity can be said to be lim-

() revealed how water crises affect both women and chil-

ited. As a result, the present study is carried out in a rural

dren. A series of studies by WASHCost Ghana () looked

community (Dungu) in Ghana to investigate the impli-

into the cost of water and sanitation services.

cations of household water insecurity on women and

Household water security in this study is measured by

children and to identify the perceptions of water quality

four key dimensions, namely: accessibility, reliability, sufﬁ-

among households and how they deal with such water.

cient quantity and acceptable quality. A water supply

The study area, Dungu, is a rural community in the

facility is said to be accessible if it is cited within a reason-

Tamale Metropolitan of the Northern Region of Ghana

able distance and is without exclusion on grounds of

and it faces the two mean climatic seasons in Ghana,

gender, religion, disability or other cause (WaterAid ).

namely the rainy and dry seasons. The rainy season is

The distance travelled to fetch water was used as a proxy

associated with high humidity, slight sunshine with heavy

for measuring water accessibility, and the recommended

thunder storms, while the dry season is characterized by

reasonable distance according to the WHO is 200 metres

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

(0.2 km). A sufﬁcient quantity of water is necessary to meet

aimed at examining the impacts of household water security

the basic needs of humans and to measure this the WHO’s

on women and children.

recommended average quantity of 20 litres per person per

A focus group discussion consisting of 12 volunteers

day was used as a measure for the acceptable quantity of

made up of ﬁve males and seven females was conducted

water for domestic use. The quality of water is an important

to get a better understanding of how people discuss the

indicator of household water security because households

water issue as members of a group rather than as individ-

cannot be secured from water-borne and water-related dis-

uals. The researchers, with the help of an interpreter and

ease when the quality is bad. Good quality water among

three research assistants, administered the questionnaires

other things must be acceptable to consumers in appear-

from June 2012 to July 2012. Because the questionnaires

ance, taste and odour (WHO ). Hence, consumers’

were written in English, there was a need to interpret in

perceptions about the appearance, taste and odour of their

the local language, Dagbani, for those who could not read

water source were used as a proxy for water quality. Water

and write to solicit the needed information from them.

supply must be available at all times and of sufﬁcient quan-

Their responses were translated into English appropriately.

tity and acceptable quality all year round in order to provide

Data from the close-ended questionnaire were trans-

reliable access. Availability of safe water all year around of

formed for analysis by coding them using the Statistical

adequate quantity and quality was used as a proxy for

Package for Social Sciences (SPSS) version 16.0. Respon-

reliability.

dents’ backgrounds, pattern of water usage and perception of water quality were analysed through frequencies and percentage distributions. Their perceptions on satisfaction with

METHODOLOGY

water quality were probed by using ﬁve questions: one on taste, one on odour, two on appearance and one on poten-

Sampling technique and data collection

tial ill-effects of drinking the water. The response to the ﬁve questions were coded as 1 ¼ yes and 0 ¼ no. A univari-

The study area, Dungu, was purposively selected for this

ate statistical analysis was also done by using the Pearson

study based on the fact that it is one of the communities in

correlation to test the strength of relationships between

the northern region with the least access to an improved

water use and socioeconomic variables (sex coded as 1 ¼

water supply. This is so because the Dungu dam is their

male, 0 ¼ female; age measured in years; household size

only source of drinking water; it is unreliable and women

measured by number of persons per household; and

and children have to travel about 2 km to access the dam.

monthly household income measured in Ghana cedis).

The unplanned nature of settlements and the absence of

The effect of water shortage on women and children was

household sample frames for the selected community pre-

ascertained through their response to open-ended questions

cluded the use of any random means of selecting the

which were entered into SPSS as string variables.

participating households. Faced with this situation, the convenience sampling approach was used to select the 125 households. Guided by the objectives of the study, a semi-

RESULTS AND DISCUSSION

structured questionnaire consisting of both closed and open-ended questions was used in the collection of data

Demographics

from the 125 sampled respondents. The questionnaire was divided into three sections. The ﬁrst section collected data

Out of the 125 respondents interviewed, 88.8% and 11.2%

on the respondent’s background, while the second section

were females and males, respectively. The majority of the

addressed the water security (thus accessibility, reliability,

respondents (56.0%) were aged 26–35 years, followed by

quantity and quality) situation at the household level as pro-

those aged 16–25 years at 28.0%. Those aged 36–45 years

vided by the respondents, which was compared with

followed at 12.8% with the 46–55 age group at 4.0%. The

international recommended standards. The third section

level of educational attainment in the Dungu community

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

seems to be low as 80% of respondents were without formal

irrigation practices to balance crop-water availability. It is

education, followed by those who have had primary edu-

associated with a lot of environmental to market challenges

cation

secondary/technical/

and such situations make farming unattractive to a number

vocational and tertiary education at 4% each. The data

of people, especially the youth, which may be the reason

further show a slight bias against female education

why low numbers of people move into farming.

(12%)

and

those

with

(Table 1). Though not shown in the table, a high number of the respondents (86%) without any form of formal edu-

Water security at the household level

cation were females. The commonest occupation was trading, providing a

Water sources, quantities consumed and accessibility

source of livelihood to 36% of the respondents, followed by shea butter production (28.8%) and farming (15.2%).

From the focus group discussion, it was revealed that the

Four per cent were students and masons and 11% were

Dungu dam was the main source of water available to the

unemployed (Table 1). Many people are engaged in trading

people. They do, however, have alternative sources from a

instead of farming as is the general case in rural Ghana due

dam and pipe-borne water at the nearby communities of

to the land tenure problem and lack of farmland resulting

Datoyili and Sawaba, respectively, although they have to

from the keen competition from industry, housing and

walk for 2.5 km to access them. During the rainy season

road construction for land use in the area. Farming in

they also take advantage of the rainwater for domestic activi-

rural Ghana is mainly rain-fed with virtually no serious

ties. The WHO/UNICEF Joint Monitoring Programme for Water Supply & Sanitation (JMP) () classiﬁes sources of water supplies as either safe or unsafe. Standpipes, bore-

Table 1

holes, protected dug wells, rainwater collection and

Demography of respondents

Variable

Percentage (%)

Number of respondents

125

Sex Male

88.8

Female

11.2

Age (years) 16–25

28.0

26–35

56.0

36–45

12.0

46–55

4.0

Educational level No formal education

80.0

Primary

12.0

Junior high school

4.0

SHS/Technical/Vocational

4.0

Occupation Shea butter making

29.6

Petty trader

36.0

Student

4.0

Unemployed

11.2

Farmer

15.2

Mason

4.0

protected springs by the use of a pipe system are examples of safe water sources while unprotected wells, unprotected springs, rivers, dams and water from tanker trucks are examples of unsafe water sources. This implies that the main source of water for the Dungu community is from an unsafe water source and hence poses a health risk to the consumers. More so, based on what Howard & Bartram () have pointed to in the literature, the health of the entire Dungu community may be at risk since they journey for more than 1,000 m or for more than 30 minutes to fetch water. Average daily household water consumption in the study area was 297.6 litres per day and an average of 9.7 litres per person per day (l/p/d) was used. This means that the average per capita water use in the study area is far below the Ghana water service norms of 20–40 litres/ person/day for basic service and the WHO’s recommended level of household use of 20 litres/person/day. One can conclude that per capita water use in Dungu is too low to maintain the level of personal and domestic hygiene needed for good health. Thus, as shown by some studies (see Howard & Bartram ), the quantity of water collected in such cases is often less than 5 litres per person

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

per day, an amount which is inadequate for good hygiene

reliable as this is subject to the vagaries of the weather. A

practices (UNICEF & WHO JMP ).

source of water needs to work for 95% of the time to be

Through the focus group discussions, it was discovered

regarded as reliable according to Ghana water service

that the distance from their homes to the dam of approxi-

norms; therefore the dam can also be deemed as unreliable.

mately 2 km was not only too far but also when it rains Water usage patterns and storage

the path leading to the dam becomes slippery which is also dangerous to their health. Apart from the long distance covered by the inhabitants to get potable drinking water

The respondents made little or no separation of water

from Sawaba and Datoyili, they also pay GH¢0.10 (US

source for usage purpose. Meaning that water in Dungu,

$0.05) and GH¢0.20 (US$0.10) per jerry can and locally

without due regard to the source, is used for drinking, cook-

made bucket, called Gariwa, respectively. WHO deﬁnes

ing, washing/cleaning, bathing/personal hygiene and other

the term reasonable access to a water source as the ‘avail-

economic activities, particularly shea butter preparation.

ability of at least 20 litres per person per day (L/capita-

Households on a weekly basis used most water for wash-

day) from a source within one kilometer of the user’s dwell-

ing/cleaning (35.4%), 27% for bathing/personal hygiene

ing’ (WHO ) and Ghana water service norms requires

and 22.6% for cooking (Table 2). Fifteen per cent of the

the distance to be 500 m or less. The low per capita daily

total water collected was used for drinking weekly and this

water used and the longer distance travelled in the study

relatively high percentage can be attributed to the arid

area is an indication of low level accessibility to safe water.

nature of the region in which Dungu is located. The strength of the relationship between water use and the age, sex and monthly household income was tested using Pearson corre-

Reliability of water supply

lation analysis and the result showed strong positive coefﬁcients of correlation between daily per capita water

Having identiﬁed the sources of water and the quantities

use and household size and household income of 0.886

consumed the study sought to ﬁnd out the reliability of the

and 0.847, respectively, at 1% level of signiﬁcance but was

water sources available to the community. From the group

insigniﬁcant for the other variables (Table 3).

discussion, it was revealed that the dam may not be a

Due the problem of water supply in the community the

reliable source of water to the community because it dries

people are forced to store some of their water to enhance con-

up two months into the dry season in November and

tinuous water availability. Four main storage facilities of

remains in that state until the rains start in June. Again,

varying sizes were identiﬁed and these were clay pots,

the water they source from the rain water obviously is not

gariwa, jerry cans and barrels. The clay pot is generally the most important water storage facility used by 81% of houseTable 2

holds in the study area. This conﬁrms ﬁndings by Sobsey

Pattern of water usage

Purpose

Percentage

Cooking

22.6

Drinking

15.0

Washing/Cleaning

35.4

Bathing/Personal hygiene

27.0

Table 3

clay pots to store their water. A safe household water storage container for drinking purposes must have some desirable properties as recommended by the Centers for Disease Control and Prevention (CDC) and the Pan American Health Organization (PAHO). Mintz et al. () mentioned the

Correlation between daily per capita water use and socioeconomic variables

Socioeconomic variables

Daily per capita water use a

(), that communities in Africa use traditional African

Statistically signiﬁcant at 1%.

Pearson correlation

Sex

Household size

Age

Income

0.091

0.886a

0.016

0.847a

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

CDC and the PAHO recommended qualities that a water sto-

Because of the unwholesome nature of the water, the

rage vessel should have. The recommendations include that

majority of the respondents (98%) treated water by using

the vessel should: (a) be made of translucent high-density

cloth ﬁltering (84.7%), boiling (14.3%) and the use of alum

polythene plastic or similar material that inter alia is pro-

(1%) (Table 4) and this treatment activity is normally done

duced locally; (b) hold an appropriate standard volume,

by the women. The widely used method of removing particles

have a stable base and sturdy and comfortable handle for

and some microorganisms from water samples is ﬁltration

easy carriage; (c) have a single opening 5 to 8 cm in diameter;

(Sobsey ) and the types of ﬁlters used for household

(d) have a non-rusting, durable, cleanable spigot for extracting

water treatment in developing countries reduce turbidity by

water; and (e) allow air to enter as water is extracted. How-

90% and bacteria by 60% according to Clasen & Bastable

ever the storage containers used in the community when

(). Alum is a ﬂocculating product or inorganic coagulant

shown to the researchers by the respondents were observed

used for removing turbidity of water supplies (Divakaran &

to fall short of the criteria for water storage facilities as rec-

Pillai ; Jain ) and this was used by only 1% of respon-

ommended by the CDC and the PAHO.

dents who treated their water. The possible assigned reason why the use of a chemical (alum) was the least preferred treat-

Perceptions of water quality and treatment

ment was that the community does not have the requisite skill and knowledge to use it. Also, none of the households

The perception by households of water quality is important in

reported using solar disinfection and ceramic ﬁlters because

determining households’ health, production and water treat-

they said they do not know how the method works and

ment decisions. Water intended for human consumption

there is a problem of accessibility.

should be both safe and wholesome. For this study, no analysis or measurement of water was performed; however, the

Impacts of household water insecurity on women and

perception of the respondents was used to determine the

children

quality of water. Respondents were asked their opinion on the quality of their main source of drinking water. The

Lack of clean water is likely to have dire consequences on

majority of the respondents (96.5%) expressed dissatisfaction

one’s education, employment, physical strength and health

with the quality of their main source of water (the dam). To

as well as on the development of society. When asked to

examine their perceptions about the factors that are likely

give the consequential effects of lack of safe water, 96% of

to inﬂuence households’ dissatisfaction with the quality of

the respondents said that the situation has negative impacts

water, the respondents were asked questions on their satisfac-

on women and children in the community.

tion with the taste, odour and appearance of their main source of water. In response, households expressed much water insecurity with the presence of particles (95%), taste (98%), odour (89.5%), colour of water (97%) and fear of infection (99%). The result from the study suggests that the major source of water in the study area, the dam, is perceived as unsafe and unwholesome by the respondents. According to

On the impact of children they mentioned that:

• • •

they are either late or absent from school, the effect of which may be that they drop out they risk their lives when crossing the roads to fetch water they also suffer from water-borne diseases such as dysentery, diarrhoea and typhoid.

the respondents the dam water is very turbid, has an unpleasant taste and does not have the clear crystalline colour associated with clean water. The result is similar to that of Obi et al. (), who conducted a study in the Venda communities in the Vhembe district of South Africa and revealed that the water from selected rivers in Venda was contaminated, thus rendering them poor, unsafe and unacceptable for human consumption.

Table 4

Household water treatment methods

Treatment methods

Percentage

Cloth ﬁltering

84.7

Boiling

14.3

Use of alum

1.0

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

On women, they mentioned the following as some of the effects of water shortage:

• • • • •

as caregivers of the household, when children get typhoid, dysentery or diarrhoea a heavy burden is put on them loss of time in search of water and treatment of water affects their economic activities increase in women’s poverty in the area for the above reasons there are cases of women slipping and hurting themselves along the path leading to the dam during the rainy season stress on their health.

CONCLUSION The paper has shown that Dungu, a rural community in the Tamale metropolis of northern Ghana, is facing a lot of challenges in accessing safe water. The source of water in the community compared with both national norms and WHO guidelines can be considered as inadequate, unreliable and inaccessible. The absence of water increases the burden on women and children as they have to walk long distances to access water which is normally unsafe in their own estimation. These women and children are deprived of the right to live a happy life as they spend most of the time fetching water. They either wait in long queues for safe water or walk long distances to access unsafe water which causes water-borne diseases. The availability of a good quality water source close to home will offer numerous beneﬁts: improved female and child health, increased dignity, and reduced exposure to both gender-related hazards and water-borne diseases. It will also enhance school attendance and academic performance of children.

REFERENCES Addo, L. Y.  Institutional Analysis of Urban Water Supply in Ghana: The Case of Accra Metropolitan Assembly. MSc Thesis, University of Aalborg, Aalborg, Denmark. Available at: http://www.projekter.aau.dk/projekter/ﬁles/42618610/ Thesis_Final_version_.pdf (accessed 12 September 2012).

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Akumiah, O. P. & Rydhagen, B.  Water Management and Health in Ghana; Case study – Kumasi. Master’s Thesis, Linköping Universitet, Norrköping, Sweden. Ariyabandu, R. De. S.  Household Water Security Using Rainwater Harvesting, RWH conference, IITD, New Delhi, April 2001. Arku, F. S. & Arku, C.  I cannot drink water on an empty stomach: a gender perspective on living with drought. Gender & Development 18 (1), 115–124. Asante, F. A.  Economic Analysis of Decentralisation in Rural Ghana. Development Economics and Policy. vol. 32. Peter Lang, Frankfurt am Main, Berlin, Bern, Bruxelles, New York, Oxford, Wein. Asare, Y. B. O.  Household Water Security and Water Demand in the Volta Basin of Ghana. Available at: http:// www.glowa-volta.de/ﬁleadmin/template/Glowa/ Downloads/thesis_osei.pdf (accessed 11 November 2013). Clasen, T. F. & Bastable, A.  Faecal contamination of drinking water during collection and household storage: the need to extend protection to the point of use. Journal of Water Health 1 (3), 109–115. Divakaran, R. & Pillai, V. N. S.  Flocculation of kaolinite suspensions in water by chitosan. Water Research 35 (16), 3904–3908. Doe, H. W.  Assessing the Challenges of Water Supply in Urban Ghana: the case of North Teshie. Available at: http:// www2.lwr.kth.se/Publikationer/PDF_Files/ LWR_EX_07_06.PDF (accessed 8 November 2013). Donkor, S. M.  Water in Africa: Management Options to Enhance Survival and Growth. New Partnership for Africa’s Development, Economic Commission for Africa. Available at: http://www.unwater.org/downloads/nepadwater.pdf (accessed 7 November 2013). Freitas, A.  Water as a Stress Factor in sub-Saharan Africa. Brief Issue 12, European Union Institute for Security Studies. Available at: http://www.iss.europa.eu/uploads/media/ Brief_12.pdf (accessed 5 January 2013). Howard, G. & Bartram, J.  Domestic water quantity: Service level and health. World Health Organization, Geneva. Jain, M.  Status of Household Water Treatment and Safe Storage in 45 Countries and a Case Study in Northern India. Master’s Thesis, Department of Civil and Environmental Engineering. Massachusetts Institute of Technology. Mintz, E., Reiff, F. & Tauxe, R.  Safe water treatment and storage in the home: A practical new strategy to prevent waterborne disease. JAMA 273, 948–953, Available at: http:// www.uvm.edu/~bwilcke/reiff.pdf (accessed 7 November 2013). Newman, R. L.  The Real Cause of Water Scarcity in Africa: Infrastructure. PRLog – Global Press Release Distribution. Available at: http://www.prlog.org/10160752-the-real-causeof-water-scarcity-in-africa-infrastructure.html (accessed 7 August 2013). Nketiah-Amponsah, E., Aidam, W. P. & Senadza, B.  Socioeconomic determinants of sources of drinking water:

H. Alhassan & P. A. Kwakwa

When water is scarce: the effects on the vulnerable

some insight from Ghana. In: Conference on International Research on Food Security, Natural Resource Management and Rural Development. University of Hamburg, 6–8 October. Available at: http://www.tropentag.de/2009/ abstracts/full/185.pdf (accessed on 12 February 2013). Obi, C. L., Potgiter, N., Bessong, P. O. & Matsaung, G.  Assessment of the microbial quality of river water sources in rural Venda Communities in South Africa. Water SA 28 (3), 287–292. Roudi-Fahimi, F., Creel, L. & De Souza, R.-M.  Finding the Balance: Population and Water Scarcity in the Middle East and North Africa. Population Reference Bureau, MENA Policy Brief 2002. Available at: http//www.prb.org/pdf/ findingthebalance_eng.pdf; (accessed 7 August 2013). Sobsey, M. D.  Managing Water in the Home: Accelerated Health Gains from Improved Water Supply, WHO/SDE/ WSH/02.07. World Health Organization, Geneva. Tayeh, A., Cairncross, S. & Maude, G. H.  Water Sources and other Determinants of Dracunculiasis in the Northern Region of Ghana. Journal of Helminthology 67, 213–225. UNICEF and WHO Joint Monitoring Programme for Water Supply and Sanitation  Drinking Water Equity, Safety and Sustainability. Available at: http://www.wssinfo.org/ fileadmin/user_upload/resources/report_wash_low.pdf (accessed 20 September 2012). United Nations  Water for Life Decade [2005–2015]. United Nations Department of Public Information. Available at: http://www.un.org/arabic/waterforlifedecade/ waterforlifebklt-e.pdf (accessed 8 November 2013). United Nations Human Rights, United Nations Human Settlement Programme & World Health Organization  The Right to Water. Office of the United Nations High Commissioner for Human Rights (UNHCR). Available at:

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

http://www.ohchr.org/Documents/Publications/ FactSheet35en.pdf (accessed 20 September 2012). Urama, K. C. & Ozor, N.  Impact of Climate Change on Water Resources in Africa: the Role of Adaptation. African Technology Policy Studies. Available at: http://www. ourplanet.com/climate-adaptation/Urama_Ozorv.pdf (accessed 20 September 2012). WASHCost Ghana  Mapping and Community Reports. Available at: http://www.washcost.info/page/2162 (accessed 8 November 2013). WaterAid  Water Security Framework. WaterAid, London. World Health Organization (WHO)  Health Systems: Improving Performance. WHO, Geneva. Available at: http:// www.who.int/whr/2000/en/whr00_en.pdf (accessed 10 September 2012). WHO  Water, Sanitation and Hygiene Links to Health: Facts and Figures. World Health Organization, Geneva. Available at: http://www.who.int/water_sanitation_health/ factsﬁgures2005.pdf (accessed 9 October 2012). WHO  Guidelines for Drinking-water Quality. First Addendum to Third Edition Volume 1 Recommendations. WHO, Geneva. Available at: http://www.who.int/water_sanitation_health/ dwq/gdwq3rev/en/ (accessed 4 August 2013). WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation  Progress on Sanitation and Drinking-Water: 2010 Update. World Health Organization, Geneva. WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation  Progress on Sanitation and DrinkingWater: 2013 Update. World Health Organization, Geneva. Available at: http://www.wssinfo.org/ﬁleadmin/ user_upload/resources/JMPreport2013.pdf (accessed 4 August 2013).

First received 12 December 2012; accepted in revised form 10 October 2013. Available online 13 December 2013

04.1

2014

Exploring the methodology of participatory design to create appropriate sanitation technologies in rural Malawi Ben Cole, John Pinfold, Goen Ho and Martin Anda

ABSTRACT The methodologies of demand-led sanitation programmes (including community-led total sanitation [CLTS] and sanitation marketing) encourage participation of users in the design of appropriate sanitation facilities. There has been limited examination of the application of established methodologies in participatory design in the sanitation sector. This paper describes and reﬂects upon three case studies that applied established participatory design methodologies to create sanitation technologies in rural Malawi. Participants of the design sessions represented two groups: (i) researcher–designers (government staff); and (ii) users (local builders and householders). The methodology created a space to develop a common language between the two groups and allowed an exploration of tensions about the use of sanitation hardware subsidies. The design sessions

Ben Cole (corresponding author) Goen Ho Martin Anda Murdoch University, School of Environmental Science, South Street, Murdoch, Western Australia 6150 E-mail: ben_cole_h2o@mac.com John Pinfold Chief of Water Supply, Sanitation and Hygiene, UNICEF Malawi, Mantino Complex Area 40/31, Area 3, Lilongwe, Malawi

created a number of innovations including corbelling structures, trapezium shaped bricks and reinforcement of wooden frame structures with sandbags. The paper critically reﬂects on the processes of participatory design in relation to power, ownership and continued participation. Key words

| design, participation, rural, sanitation

INTRODUCTION Top-down sanitation programmes that promote a speciﬁc

design approaches (Blackett ; Baker et al. ).

technology based on the presumptions of ‘outside experts’

Human-centred design attempts to create sanitation pro-

have been criticised for endorsing unsustainable, expensive

ducts and services that match the needs, practices and

and inappropriate technologies (Cairncross ; Jenkins

preferences of users and suppliers of sanitation technol-

& Sugden ). In response to these failings, a new era

ogies. In Vietnam, Cambodia and Kenya sanitation

of demand-led sanitation programmes encourage greater

marketing

participation of users to create, identify and select appropri-

researchers–designers

programmes to

have

engaged

conduct

international

qualitative

market

ate sanitation technologies (Cairncross ; Kar &

research with users and suppliers (Sijbesma et al. ;

Chambers ). Although comprehensive compilations of

Baker et al. ). The market research informed the

sanitation systems and technologies exist (e.g. Tilley et al.

researchers-designers to create iterative prototypes that

), few studies critically examine the methodologies

were subjected to extensive user testing.

used to engage local users and suppliers in the design of

The CLTS methodology does not recommend inputs

appropriate sanitation technologies. This paper engages

from external researchers–designers but suggests that local

with that space by presenting and reﬂecting upon a partici-

facilitators should ‘help in establishing linkages with local

patory design methodology applied in rural Malawi.

markets’ (Kar & Chambers ). The role of local facilita-

Two common demand-led sanitation programmes are

tors also includes: encouraging local innovation and

sanitation marketing and community-led total sanitation

production; identifying locally available products and

(CLTS) (Mara et al. ). Sanitation marketing programmes

materials; and training locals to manufacture sanitation

in Lesotho, Cambodia and Kenya applied human-centred

technologies (Kar & Chambers ). Recently, a founder

doi: 10.2166/washdev.2013.166

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

of the CLTS approach emphasised that the selection of tech-

in existing sanitation technologies in rural Malawi (Phiri

nology is crucial to overcome second- and third-generation

; Cole et al. ). Research conducted in three rural dis-

problems such as groundwater contamination and environ-

tricts found more than half of all sanitation facilities

mental pollution (Kar ).

collapsed within 12 months of construction, resulting in users returning to open defecation or sharing a facility

Participatory design practices

(Cole et al. ). Cole et al. () also reported that sanitation suppliers were unengaged in the sector because of

This research applied a participatory design methodology

low demand for existing sanitation technologies.

described in Spinuzzi () and IDEO () to engage

The overall objective of this paper is to record, examine

users and researchers-designers in the creation of appropri-

and critically reﬂect on the use of Spinuzzi’s () and

ate sanitation technologies. Participatory design evolved

IDEO’s () participatory design methodology to identify

through the US labour movement and Scandinavian code-

appropriate sanitation technologies. To achieve this objec-

termination laws during the 1970s. The approach was used

tive the following speciﬁc aims are addressed:

to democratically engage workers in the design and inte-

•

To describe the methodology used during the participa-

•

To evaluate the outcomes of the participatory design

gration of new technologies within their work processes (Nieusma ; Spinuzzi ). Participatory design is now an established methodology used across a wide breadth of development programmes (Nieusma ; WinschiersTheophilus et al. ). Steen () argues that participatory design is a speciﬁc approach of human-centred design that ‘attempts to give future users of a system a role in its

•

tory design sessions. methodology between the three case studies. To explore the challenges of power, ownership and ongoing engagement in the application of participatory design methodology.

design, implementation and evaluation’. Participatory design provides a space for users to express traditional, tacit and often invisible knowledge and skills used in their

METHODS

daily lives (Spinuzzi ). Participatory design’s methodological and philosophical

Country context

position aligns to participatory action research (PAR) (Winschiers-Theophilus et al. ). Both approaches

Up to 49% of households in rural Malawi have an

attempt to create a democratic space between researchers-

inadequate or absent sanitation facility (WHO & UNICEF

designers and users and link research objectives with action-

). The Malawian government has identiﬁed CLTS and

able goals (David ; Bozalek ). Criticisms of PAR

sanitation marketing as the two leading mechanisms to

methodology include the application of Western-democratic

achieve its objective of total sanitation coverage by 2015

practices in non-Western cultures (Campbell ) and the failure to include participation from marginalised groups (Bozalek ; Winschiers-Theophilus et al. ). These criticisms have encouraged commentators of participatory

(MoAIWD ). Since 2009, CLTS programmes have been applied extensively across rural communities (Maulit & Kang ). In contrast, few rural sanitation marketing programmes have been implemented (DeGabriele ).

design to advocate that researchers have a duty to critically reﬂect on their research processes and outcomes (Steen

Programme context

). In 2011, UNICEF Malawi initiated a rural sanitation marExisting sanitation technologies in rural Malawi

keting

programme

with

three

district

government

partners. In partnership with staff members of the District Evaluations of CLTS programmes and formative market

Environmental Health Ofﬁce (DEHO), the ﬁrst author

research identiﬁed an urgent need for design improvements

conducted formative market research (Cole et al. ).

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The research informed the development of an integrated marketing strategy that addressed the four Ps of ‘Price’, ‘Product’, ‘Promotion’ and ‘Place’. In 2012, the participatory design sessions were applied to address the ‘Product’ and ‘Price’ components of the integrated marketing strategy. The development of the ‘Product’ and Price’ components was based on a ground-up philosophy such that the primary source of information was derived from the collective knowledge and skills of local builders and villagers. It was intended that in each district, the designs would be collated and provided to the DEHO. The DEHO would

Photo 1

identify designs that would be provided to the Nationallevel Open Defecation Free (ODF) Taskforce. The ODF Taskforce is a co-ordinating body for organisations operat-

User describes the components of existing brick-lined pit latrines.

enough to make the topic manageable’ (IDEO ). The design challenge used during the sessions was:

ing in the sanitation sector and includes representatives from the Ministry of Health, Ministry of Irrigation and

‘Can we create a toilet that matches what the majority of

Water Development, UNICEF and leading non-government

villagers want, need and can afford using local materials?’

organisations. Teams visualised their designs through drawing and text Participatory design approach

(Photo 2). After two hours of brainstorming, each team was asked to identify three designs that they would like to

The participatory design sessions were derived from the

prototype. The teams identiﬁed the materials required to

methodologies presented in Spinuzzi () and IDEO

create the prototypes and these were submitted to the

().

DEHO staff for collection from local suppliers.

The three-day design sessions consisted of four stages: Stage 3: Prototyping Stage 1: Initial exploration of work Day 2 was dedicated to the process of creating small and Teams of ﬁve builders/masons, two village health workers/

medium-sized prototypes. Local building materials were

householders and one staff member of the DEHO were formed. Each team was asked to draw and label the existing sanitation technologies in their villages. The teams were then instructed to identify the advantages and disadvantages of each technology. Each team then presented their ﬁndings to the group (Photo 1). Stage 2: Discovery processes The discovery process asked each team to identify numerous potential design options. The design options were framed by a design challenge. A design challenge presents a challenge in human terms, in a broad manner that offers opportunities for discovery in areas of unexpected value but is ‘narrow

Photo 2

Design team visualise their designs.

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

provided to allow each team to explore and create their design

soil proﬁles. The sites were located in Salima, Mangochi

directions (Photo 3). Users were encouraged to share and dis-

and Nkhata Bay districts. The design sessions were con-

cuss their ideas with people from outside their nominated team.

ducted at the group village level. In Malawi, a group village commonly consists of three to four villages.

Stage 4: Feedback

Group villages were purposefully selected using the following criteria: (i) consisting of more than three villages or

The ﬁrst half of day 3 allowed the design teams to estimate

greater than 300 households; (ii) not more than 90 minutes

the material and labour costs of their prototypes.

drive from a central market; (iii) representative of a typical

During the ﬁnal afternoon on day 3, teams presented

soil type within the district; (iv) a suitable location to con-

their designs to 12–15 local villagers including both men

duct building and construction of prototypes; and (v) a

and women. The villagers were invited to review the proto-

group village leader or senior health worker with a proven

types and provide feedback (Photo 4). The feedback sessions

record in supporting innovative social programmes.

were intended to allow the design teams to hear critiques from potential users.

Participant selection

Study sites

The participatory design engaged with two groups, ‘researcher-designers’

and

‘users’.

Researcher–designers

Three study sites were chosen to represent areas located

included environmental health ofﬁcers (EHO) and their

along the lakeshore of Malawi with both clay and sandy

support staff. The EHO and their support staff were engaged in formative market research prior to the design sessions (Cole et al. ). Users included builders, masons and householders. Users were purposefully selected from the group villages. Five builders/masons were invited to attend from each village (total of 20 representatives). The selection criteria were: (i) proven reputation as a builder, carpenter or mason; (ii) constructed a latrine in the last 12 months; and (iii) a permanent resident of their village. Two representatives of households were invited from each village (up to eight representatives). Householder representatives were individuals that had acted as natural leaders during previous

Photo 3

Design team creates a small-scale prototype of corbelled pit lining.

CLTS events, village health workers or people with a proven interest in sanitation. Data collection and analysis The study applied a qualitative multiple-case study methodology (Baxter & Jack ). The first author conducted overt participant-observation at one design session held in each of the three study sites. Qualitative data was collected using triangulated sources that included the first author’s descriptive and reflective field notes, participants’ drawings and text, and recorded interviews with participants. The interviews were conducted during the design sessions with builders/

Photo 4

Villagers inspect a prototype in Salima district.

masons,

householders

and

government

staff.

The

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

participant observer. These criticisms were addressed in

interpreter speaking in Chichewa. The interpreter had ﬁve

this study by recording notes during and immediately after

unstructured

interviews

were

conducted

through

years experience and a university diploma in teaching. The

each design session, through reﬂective bracketing (Ahern

local interpreter was trained for two days prior to conduct-

) and through re-listening to interviews to elaborate

ing the design sessions. The training developed a rapport

the ﬁrst author’s ﬁeld notes.

between the ﬁrst author and interpreter and was used to

The selection of participants is a central determinant of

clarify the methodology, interview process and terminology

the outcomes of PAR (David ). This study identiﬁed par-

(Pitchforth & Teijlingen ; Squires ). The lead

ticipants through identifying selection criteria with district-

author and interpreter identiﬁed potential biases and

level government staff. Local health workers and village lea-

addressed these through the process of ‘bracketing’

ders applied the criteria to select the participants. It is

described by Ahern (). The interviews were recorded

possible that participants were selected based on their pos-

and replayed to inform the ﬁrst author’s descriptive ﬁeld

ition

notes. Key statements from the transcripts were transcribed.

connections with village health workers or village leaders.

Additional data was collated through photographs of

This may have biased the selection against the inclusion of

prototypes and a report prepared by a structural engineer.

within

the

community

through

personal

users from marginalised and lower income groups.

The ﬁeld notes, transcripts, photographs and reports were analysed using qualitative content analysis to identify and prioritise key themes and outcomes across the three case studies (Sandelowski ). The themes reported in this paper were selected for inclusion based on the interpretation of the ﬁrst author.

RESULTS AND DISCUSSION Conﬁrming language and knowledge between users and researcher–designers

Limitations

An important goal of participatory design is the creation of a

The design sessions were conducted in three of the 28 dis-

(Spinuzzi ; Steen ). The initial exploration of work

common language between users and researcher–designers tricts of Malawi. In each district, three traditional

(Stage 1) provided a strong platform for users and

authorities were identiﬁed of which three group villages par-

researcher-designers to develop a common vocabulary for

ticipated in the design sessions. Up to 35 participants

sanitation-related terminology.

attended each workshop, with a total of 104 participants attending the three sessions. The ﬁndings from the three

‘We can learn from the language of the builders. It helps

case studies must be interpreted in the context of their geo-

us to speak with their words. Researcher–designer,

graphical scope. The designs identiﬁed are speciﬁc to their

Nkhata Bay’

place of origin. It is likely that design sessions held in different geographical locations and with different participants

The three main types of sanitation technologies identiﬁed

would result in design ideas reﬂecting the different needs,

during the design sessions were categorised as unlined,

resources and challenges of different locations and social

nkhokwe (woven wooden-, reed- or bamboo-frame that is

contexts.

cylindrical in shape) and brick-lined with cement mortar

Participant observation brings both advantages and

pit latrines. The advantages and disadvantages of existing

challenges to the rigour of a study. Advantages include the

sanitation technologies aligned with the ﬁndings of market

ability of the researcher ‘to be open, discovery oriented

research previously conducted by the researcher–designers

and inductive and less reliant on preconceptions’ (Boyd

(described in Cole et al. ). The Stage 1 process offered

). Criticisms of participant observation include the

a signiﬁcant beneﬁt of allowing the two parties to speak con-

risk of selective memory, encoding and attention, which

ﬁdently and openly about the challenges presented by

bias the focus and recording of information by the

existing sanitation technologies.

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Divergence of attitudes towards hardware subsidy

subsidy to come back. I’ve seen this too many times.

programmes

Researcher–Designer, Mangochi’

A signiﬁcant challenge arose during the discovery process

The dichotomy of views toward the inclusion of cement as a

where some users (builders and householders) insisted on

building material created a tension between users and

the inclusion of cement as a building material for latrine

researcher-designers. Participatory design scholars recognise

construction.

staff)

that overcoming disagreement in design objectives is an

argued that including cement in the latrine design would

important process that must be mediated (Nieusma ).

Researcher–designers

(government

make it unaffordable for the majority of households. The

To overcome this tension, the ﬁrst author recommended

users argued that Government should provide the cement

that the group recognise the important characteristics of

through a subsidy programme. This created debate amongst

cement while creating designs that reduce or eliminate its use.

the group on the role of Government in the provision of hardware subsidies in sanitation programmes. Two main

‘I have heard that you like the cement slab because it is

themes emerged from the users’ perception of cement. The

strong and durable, they like it because the children can

ﬁrst was that cement is an essential component of a strong

use it and are not scared and because termites cannot

and modern latrine, and the second was that subsidy pro-

eat it and it is easy to clean … but now I want to ask you,

grammes for cement can be successful if managed

can we keep these ﬁve things, these ﬁve characteristics,

appropriately.

but push the price down to allow households the capacity to buy a latrine on their own? First author, Nkhata Bay’

‘Cement is progress, why would we want to use wood? Wood is a primitive way for latrines. Cement is a part of progress. User, Nkhata Bay’

The instruction not to include cement in the building materials encouraged the design teams to explore latrine designs that did not rely on a cement slab. The instruction

‘The other (cement) subsidy programmes were not well managed, we can manage them better. The fertiliser subsidy is working, why can’t we do the same for cement? User, Nkhata Bay’

resulted in design teams exploring numerous design options that replicated the attributes of a cement slab (that is, strength, water- and termite-resistance) while keeping material costs affordable. The resulting design options included metal frames, wrapping of logs with plastic, widen-

Malawi has a long history in the provision of subsidised cement through Government and non-government sanitation programmes (DeGabriele ). It is therefore unsurprising that users assumed a hardware subsidy would be provided in future sanitation programmes. To counter this assumption, the researcher–designers stated that subsi-

ing of the roof structure accompanied with surface water diversion channels and self-supporting brick-domes. Users and researcher–designers identiﬁed the self-supporting brick-dome as a potentially viable design that offered the same attributes of a cement slab at a signiﬁcantly reduced cost (as discussed further below).

dies were no longer promoted by National Government policy (as described in the Open Defecation Free by 2015

Consistent design themes identiﬁed for clay soils

Strategy [MoAIWD ]). Researcher–designers also presented

their

personal

experiences

subsidised

The design sessions in Mangochi and Nkhata Bay identiﬁed

programmes that had failed under various forms of

a consistent design direction for environments with clay

management.

soils. The design overlaid burnt bricks to form a self-supporting brick-dome to create the ﬂoor and slab of the pit latrine

‘You build 10 toilets with a subsidy then go away. You

(Figure 1). Corbelling is an established building practice for

come back in ﬁve years and still only 10 toilets have

pit latrines. It overlaps one brick over another to reduce the

been built in that village. People are waiting for the

diameter of the pit (Government of Zimbabwe n.d.).

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The identiﬁcation of the corbelled design provided evidence of the participatory methodology drawing out the tacit knowledge of users. Users reported the corbelled design was inspired from building technologies observed both within and beyond their local villages. ‘I saw this approach in Nhkota Kota. It is used there but I hadn’t used it before to build a latrine. I just thought we could try it and then other people in my group said ‘yes, that is a good idea’. User, Nkhata Bay’ ‘This approach is very common in the areas around the lakeshore in Mangochi. We have been using this approach for many years. User, Mangochi’ Figure 1

Diagram of corbelled pit lining, ﬂooring and slab.

Lead users created innovative designs

Existing designs (such as the Blair VIP Latrine) apply corbel-

The Mangochi and Salima design sessions were conducted in

ling to reduce the size of the cement slab. In contrast, the

areas that had previously been exposed to sanitation pro-

design that emanated from the design sessions in both Man-

grammes. In Mangochi, sanitation suppliers from one group

gochi and Nkhata Bay used corbelled bricks to form the slab

village had been exposed to innovative sanitation technologies

(Photo 5). Users stated the beneﬁts of brick corbelling was

through a Swiss-funded subsidy-based programme (Photo 6).

that it eliminated the need for: (a) wood, which is commonly

In Salima, a CLTS programme had been conducted two

attacked by termites resulting in an unstable ﬂoor; and (b)

years prior to the design session. The CLTS programme had

cement, which is unavailable in local markets and is prohi-

encouraged sanitation suppliers to develop innovative sani-

bitively expensive for the majority of households. All

tation designs. Using the terminology of human-centred

participants identiﬁed burnt bricks as widely available in

design, these suppliers were recognised at lead-users. The

clay soil environments. The construction of burnt bricks

lead-users identiﬁed the challenges of existing sanitation

requires the collection of local clay, a wooden frame to

designs and offered immediate solutions. One solution com-

shape the brick and wood and other plant-based fuels to

bined the widely used nhkokwe but wrapped the frame in

ﬁre the bricks.

plastic and then reinforced it with sandbags (Photo 7). The sand-bags were constructed by cutting and re-sewing locally

Photo 5

Bricks and mud mortar are used to form the pit opening.

Photo 6

Masons use trapezium shaped bricks to form a circular pit lining.

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

element of participatory design (Steen ). During the design sessions the ﬁrst author instructed the design teams to minimise the amount of cement used in the designs. This decision was taken to increase the affordability and accessibility of the designs. This decision challenged a number of users who were unwilling to move away from the concept of the Government providing a hardware subsidy for cement. It is clear that the inherent power of the ﬁrst author (as a foreign researcher) played a part in directing the boundaries of the participatory design process. Photo 7

Nkhokwe (wooden frame) wrapped in plastic with additional sand-bag reinforcement.

While it could be argued the use of this inherent power may contradict the principles of a genuine participatory process, it is important to recognise the necessity for establishing

available maize bags. This innovative design shows how enga-

speciﬁc design criteria and boundaries when applying parti-

ging villagers who have been exposed to CLTS triggering

cipatory approaches.

events brings lead-users to the design team who are able to

The question of ownership of the designs is a vexed

use their previous experiences to enhance the participatory

question that was not discussed during the design sessions.

design outcomes.

It could be argued the intellectual property of the designs should retain with the original designer or design team.

Responding to criticism The feedback sessions were intended to allow design teams to ‘present solutions with a neutral tone, highlighting both pros and cons of a solution’ (IDEO : 80). During the feedback sessions users found it difﬁcult to engage in an open and honest exchange of ideas, criticisms and feedback. The ability to ‘design on the ﬂy’ is a challenging task and the majority of users found the feedback process both challenging and confronting. The researcher–designers recorded the villagers’ feedback so that it could be fed into future iterations of the prototypes. The feedback sessions did create tension between the users and researcher-designers. Users were reluctant to respond to criticisms while researcher–designers embraced the feedback. Future design sessions could overcome this tension by providing additional time (up to one week) for villagers to provide feedback to the design team. This would allow the design team to hear the feedback in an informal setting without being pressured by time.

Indeed, in Western cultures new ideas are regarded as the property of individuals or corporations. However this sits in contrast to numerous African cultures where ideas and knowledge ‘are ‘owned’ by ancestors or the land’ (Winschiers-Theophilus et al. ). The issue of ownership is further complicated as some designs, including the brick corbelled design, were created at multiple design sessions. It could also be argued the potential value of a new design to improve public health outweighed the rights of the individual’s intellectual property. The issue of ownership deserves further attention in follow-up sessions with the users. Participatory design is an iterative process that requires continual participation, actual use of prototypes and sustained reﬂection (Spinuzzi ; Brereton & Burr ). The methodology described in this paper offers only a preliminary glimpse into the potential tacit knowledge of villagers and builders to design appropriate sanitation technologies. The programme has been designed to continue engagement with the users through returning to discuss, redesign and test the revisions and iterations suggested by

Reﬂections on power, ownership and participation

the National ODF Taskforce and the structural engineer’s report. It is hoped that continuing the prototyping process

Recognising the inherent power of the researcher–designer

will allow users to ‘critically reﬂect on the implication of

and the ability to exert that power is an important reﬂective

the research results for their own work’ (Spinuzzi ).

B. Cole et al.

Identifying rural sanitation technologies through participatory design

CONCLUSIONS

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

level, the use of participatory approaches in designing sanitation technologies naturally aligns to the historical

Be prepared to discuss and debate the issue of

development of CLTS from participatory rural appraisal and

hardware subsidies with users

PAR. At the methodological level, participatory design sessions allow the community to innovate and produce local

The research found that encouraging an open discussion on the role of hardware subsidies provided a forum for dialogue

sanitation products and services (Kar & Chambers ). The sessions also align with the development of an inte-

between the users and researcher–designers. Although not

grated

sanitation

marketing

programme.

Sanitation

all members of the two groups reached consensus, the pro-

marketing programmes have evolved from the discipline of

cess created a space for dialogue. The dialogue should aim

social marketing. Key characteristics of social marketing are

‘not to convince each other of the rightness of one’s opinion

its consumer orientation and the exchange process (Donovan

or to merge individual pre-factored ideas, but rather aims to

). By engaging users in the design process, the participatory

jointly create something new by talking’ (Winschiers-Theo-

sessions increase the likelihood the ﬁnal product will align with

philus et al. :168).

both the needs and preferences of the end-user/consumer. The participatory sessions also establish relationships with existing

Develop specific design criteria in a small team that are familiar with the findings of the formative market research Clear and specific design criteria, framed in human terms, is an essential ingredient of successful and productive design processes (IDEO ). This research developed the design criteria in a small team that was engaged in the collation and analysis of the formative market research. Developing the criteria in small, well-informed teams ensured the design sessions were suited to the specific environmental, socio-economic and market conditions of the three study sites.

sanitation suppliers and may encourage them to exchange a broader range of sanitation products and services. The case studies do not suggest that participatory design is a ‘silver bullet’ in identifying sanitation products and services

for

sanitation

marketing

programmes.

Indeed

sanitation products and services designed without participatory methodologies have demonstrated rapid uptake in Vietnam (Sijbesma et al. ). For these case studies, the use of participatory design was chosen after extensive market research and supply chain analysis (described in Cole et al. ). The formative research identiﬁed the majority of rural consumers had limited access to cash and water, combined with poor transport infrastructure and limited distribution channels for prefabricated sani-

Identify and engage with lead users and innovators in

tation products (such as plastic- and cement-slabs and

the local sanitation sector

ceramic pans). In the future, however, market conditions may change and the introduction of a range of prefabricated

Engagement with leading thinkers and innovators is an established approach in the ﬁeld of human-centred design (Steen ). Members of the design team that had been previously

sanitation products may become favourable. This may require engaging local shop owners to provide distribution and sales points for the prefabricated products. Widening

exposed to CLTS programmes were found to drive the inno-

the availability, price-range and attributes of sanitation pro-

vation process within their team and the wider group.

ducts may increase the size of the target audience and hence the coverage of improved sanitation across rural Malawi.

Participatory design methodology can be integrated into CLTS and sanitation marketing programmes

Technical review is vital prior to the commercial release of new sanitation technology

The participatory design methodology aligns with the philosophical and methodological approach applied in both CLTS

The structural engineer report provided vital input into the

and sanitation marketing programmes. At the philosophical

iterative development of the original prototypes. The

B. Cole et al.

Identifying rural sanitation technologies through participatory design

report provided recommendations on the safety and durability of the designs and offered suggestions for locally managed trials to test improvements and certain aspects of the designs. Failure to include this stage could result in unsafe designs being released into the market. The engineer’s report will be used to test and reﬁne the designs at Mzuzu University’s SMART Centre. Testing will include mortar strength rating, load testing and standardisation of designs. Throughout the second-half of 2013 the new round of designs will be presented to the design teams in two group villages in each district. The new designs will be presented with drawings and images that identify any changes or additions to the original design. This will allow users and researcher–designers to understand the strengths and limitations of their original designs and offer an opportunity for them to provide ongoing feedback and input into the design process. Need to provide focus on hand washing facilities The design sessions missed the opportunity to engage participants on the design of hand washing facilities. Future sessions should include hand washing facilities as an essential component of the latrine design. Overall, the participatory design methodology was engaging and highly productive. Users and researcher–designers displayed high levels of engagement throughout the threeday design sessions. The participatory design methodology, supported with ongoing technical feedback, offers a potential mechanism to create safe, affordable and desirable sanitation technologies that match a range of local conditions.

REFERENCES Ahern, K. J.  Ten tips for reﬂexive bracketing. Qualitative Health Research 9, 407–410. Baker, T., Roberts, M., Cole, B. & Jacks, C.  Sanitation Marketing Pilot Project: End of Project Report. International Development Enterprises, Cambodia. Baxter, P. & Jack, S.  Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report 13 (4), 544–559. Blackett, I.  Low-cost Urban Sanitation in Lesotho. UNDP and World Bank Water and Sanitation Program, Washington, DC.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Boyd, D.  A framework for capacity development: Closing the gap between theory and practice. PhD Thesis, Murdoch University, Western Australia. Bozalek, V.  Acknowledging privilege through encounters with difference: Participatory learning and action techniques for decolonising methodologies in Southern contexts. International Journal of Social Research Methodology 14 (6), 469–484. Brereton, M. & Burr, J.  New challenges for design participation in the era of ubiquitous computing. CoDesign 4 (2), 101–113. Cairncross, S.  The Case of Sanitation Marketing. Water and Sanitation Program: Field Note, Washington, DC. Campbell, J.  A critical appraisal of participatory methods in development research. International Journal of Social Research Methodology 5 (1), 19–29. Cole, B., Pinfold, J., Ho, G. & Anda, M.  Investigating the dynamic interactions between supply and demand for rural sanitation, Malawi. Journal of Water, Sanitation and Hygiene 2 (4), 266–278. David, M.  Problems of participation: the limits of action research. International Journal of Social Research Methodology 5 (1), 11–17. DeGabriele, J.  Sanitation Sector Status and Gap Analysis: Malawi. Global Sanitation Fund & Water Supply and Sanitation Collaborative Council, Geneva. Donovan, R.  Social marketing’s mythunderstandings. Journal of Social Marketing 1 (1), 8–16. Government of Zimbabwe n.d. The Blair VIP Latrine: A Builder’s Manual for the Upgradeable BVIP Model and a Handwashing Device Draft. National Action Committee for Rural Water Supply and Sanitation, Zimbabwe. IDEO  Human Centered Design Toolkit: An Introduction, 2nd edn. Available from: www.ideo.com/work/humancentered-design-toolkit/ (accessed 15 February 2013). Jenkins, M.W. & Sugden, S.  Rethinking Sanitation: Lessons and Innovation for Sustainability and Success in the New Millennium. Human Development Occasional Papers (1992– 2007) HDOCPA-2006–27. Human Development Report Ofﬁce (HDRO), United Nations Development Programme (UNDP), New York. Kar, K.  Why not basics for all? Scopes and challenges of community-led total sanitation. IDS Bulletin 43 (2), 93–96. Kar, K. & Chambers, R.  Practical Handbook to Implementing Community-Led Total Sanitation. Plan International, London. Mara, D., Lane, J., Scott, B. & Trouba, D.  Sanitation and health. PLoS Medicine 7 (11), e1000363. Maulit, J. A. & Kang, M.  Effects of program and institutional design on district-level CLTS management in Malawi. Presentation to 35th WEDC International Conference, 6–8 July, 2011, Loughborough, UK. MoAIWD (Ministry of Agriculture, Irrigation and Water Development)  Open Defecation Free Malawi by 2015: Strategy Document. Government of Malawi, Lilongwe. Nieusma, D.  Alternative design scholarship: working toward appropriate design. Design Issues 20 (3), 13–24.

B. Cole et al.

Identifying rural sanitation technologies through participatory design

Phiri, S.  TA Mkanda CLTS Research Summary. Engineers without Borders Canada. Pitchforth, E. & Van Teijlingen, E.  International public health research involving interpreters: a case study from Bangladesh. BMC Public Health 5 (71), 1–7. Sandelowski, M.  What’s in a name? qualitative description revisited. Research in Nursing & Health 33, 77–84. Sijbesma, C., Truong, T. X. & Devine, J.  Case study on sustainability of rural sanitation marketing in Vietnam. Technical Paper. Water and Sanitation Program, Washington, DC. Spinuzzi, C.  The methodology of participatory design. Technical Communication 52 (2), 163–174. Squires, A.  Methodological challenges in cross-language qualitative research: a research review. International Journal of Nursing Studies 46, 277–287.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Steen, M.  Tensions in human-centred design. CoDesign 7 (1), 45–60. Tilley, E., Luthi, C., Morel, A., Zurbrugg, C. & Schentenleib, R.  Compendium of Sanitation Systems and Technologies. Swiss Federal Institute of Aquatic Science and Technology (Eawag), Dübendorf, Switzerland. WHO & UNICEF  Progress on Sanitation and Drinkingwater 2010 Update (online). Joint Monitoring Programme for Water Supply and Sanitation. Available from www. who.int/water_sanitation_health/publications/ 9789241563956/en/index.html (accessed 10 September 2010). Winschiers-Theophilus, H., Bidwell, N. & Blake, E.  Altering participation through interactions and reﬂections in design. CoDesign 8 (2–3), 163–182.

First received 20 April 2013; accepted in revised form 29 August 2013. Available online 21 October 2013

04.1

2014

Viscous heating analysis of simulant feces by computational ﬂuid dynamics and experimentation Jagdeep T. Podichetty, Md. Waliul Islam, David Van, Gary L. Foutch and A. H. Johannes

ABSTRACT Highly viscous substances, such as feces, produce significant heat when layer deformation occurs. We describe the use of viscous heating sufficient to destroy disease-causing microorganisms and whipworms in feces. Computational fluid dynamics (CFD) was used to evaluate preliminary design and provide initial geometric specifications for a laboratory-scale unit. The laboratory device has a rotating core separated from a fixed shell wall by a defined space. Data were obtained over a range of operating conditions with simulant materials. The CFD model was validated with the experimental W

results. The temperature observed with the smallest spacing was 190 C. Alternative geometries are

Jagdeep T. Podichetty Md. Waliul Islam David Van Gary L. Foutch (corresponding author) A. H. Johannes 423 Engineering North, School of Chemical Engineering, Oklahoma State University, Stillwater, OK 74078, USA E-mail: foutch@okstate.edu

considered for high-volume sludge processing. Potential design modiﬁcations include enhancing efﬁcient water evaporation and recovery. Key words

| Comsol Multiphysics, extrusion, feces, parasite, viscosity, viscous heating

LIST OF SYMBOLS Cp

speciﬁc heat capacity, J/kg K 3

volume force vector, N/m

volumetric force component radial direction, N/m3

Fφ

volumetric force component rotational direction, N/m3

volumetric force component axial direction, N/m3

thermal conductivity, W/m K

pressure, Pa

heat, J or kJ

Qvh viscous heat, J r

radius, m

time, s

absolute temperature, K

velocity, m/s

radial velocity, m/s

rotational velocity, m/s

axial velocity, m/s

ω angular velocity, m/s β efﬁciency

Superscript T transpose

INTRODUCTION Worldwide, numerous technologies are being considered for the treatment of human wastes to minimize the spread of disease. Diarrheal diseases kill approximately 1.3 million people annually; most are children aged 0–4 years (World Health Organization ). Contaminated water, as a result of poor or inadequate sanitation, accounts for substantial mortality

wW velocity component in angular direction, m/s

and is the leading cause of diarrhea in developing countries,

Wp pressure work, J

especially urban slums. Fecal matter carrying pathogens can

Greek

enter the water supply in multiple ways: storm water runoff, 3

ρ density, kg/m

septic tank leakage, sanitary sewer system breakage and over-

η dynamic viscosity, Pa.s

ﬂow, and improper disposal of human and animal waste

doi: 10.2166/washdev.2013.070

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

(Santo Domingo et al. ). Parasites that spread throughout

or transport and can subsequently be used in energy conver-

the environment with human feces include helminths and

sion or agricultural processes.

protozoa, with particular interest focused on Ascaris lumbricoides,

Giardia

intestinalis,

Trichuris

trichiura,

Similar devices for non-fecal applications have been reported; for example, Yesilata () worked on the rotation

Cryptosporidium spp. and Taenia spp. (Trönnberg et al.

of mass between two parallel disks. Hooman & Ejlali ()

). In areas where sanitation is non-existent or ineffective

included viscous heating in a correlation to improve ﬂow.

(Trönnberg et al. ), these parasites and protozoa cause

They explained liquid viscosity decreases with incremental

diseases such as cholera, diarrhea and typhoid. The helminth

temperature which resulted in lower pressure drop. Although

(parasitic worm) infections in the human stomach caused by

they provided a theoretical solution for both no-slip and slip

the microbes listed above are ascariasis, trichuriasis (whip-

ﬂow in cases with forced convection of liquid in a micro chan-

worm) and hookworm. These diseases cause both physical

nel ﬂow, no experimentation was included. However, an

harm and reduce scholarly and cognitive development (Beth-

analytical solution for nonisothermal ﬂow with wall slip pro-

ony et al. ).

vided a better understanding of temperature rise in die ﬂow

Viscous heating is a well-known phenomenon; technol-

for viscoelastic ﬂuids (Lawal & Kalyon ). Their mathemat-

ogies include polymer melts and sludge dewatering. Rock

ical solution indicates application over a varied range of

formation is also associated with viscous heating in nature.

viscoplastic ﬂuids. Depending on the resultant viscosity, a por-

A discussion of the contribution of viscous heating in the

tion of the processed product might be recycled to sustain

ﬁeld of geology, along with modeling of the effects is pre-

constant viscosity. Temperature distribution due to viscous

sented by Burg & Gerya (). During viscous heating,

heating of a shear stressed ﬂuid is discussed by Collins ().

temperature rise is caused by internal friction at high ﬂow

In this study, preliminary designs were evaluated using

velocities (Sunden ). Sunden showed viscous heating

computational ﬂuid dynamics (CFD). These simulation pro-

effects in forced convective ﬂow across a circular cylinder

grams solve and analyze energy and mass balances within

at low Reynolds numbers. To evaluate viscous heating for

any geometry. For a shear reactor, these equations describe

this application fecal simulants were used. Structural, ther-

the mass ﬂow and temperature proﬁles. CFD is superior to

mal and viscoelastic properties for potatoes (Singh et al.

simpliﬁed approaches, such as one-dimension ﬂow analysis

) indicate their potential as a simulant. Yavuz et al.

methods using hydraulic radii and lubrication approxi-

() investigated laminar duct ﬂow heat transfer with

mations (Gifford ). Results indicated initial geometric

viscous dissipation for a Newtonian ﬂuid to deﬁne tempera-

speciﬁcations prior to construction and reduced mechanical

ture distribution within annular pipes. Finite difference

trial and error. The CFD model was validated with exper-

analysis of the heat transfer mechanisms for non-Newtonian

imental results using ﬂuid properties and rheology data for

ﬂuids in circular tubes showed that viscous dissipation leads

red potatoes. Variable analysis was performed within the

to high temperature in processing ﬂuids. Heat generated

CFD simulator to evaluate the effects of shear reactor geo-

decreases the viscosity of the ﬂuid and results in less temp-

metry on temperature gradient.

erature rise than for a material with constant viscosity (Costa & Macedonio ). We are evaluating the application of viscous heating as a

MATERIALS AND METHODS

component of sanitation. Heat is generated within fecal simulants by applying shear stress with an extruder. Heat

CFD modeling

generation is sufﬁcient to sanitize the mass. For watery solids, paper, sawdust or grass clippings could potentially

Governing equations

be added to increase viscosity. Alternatively, allowing water in the heated mass to vaporize and recycling the

Typically, an extruder-die design depends on three elements,

solids may maintain high viscosity and achieve sufﬁcient

an accurate viscosity model, ability to solve three-

temperature. Once treated, the feces will be safe to handle

dimensional (3D) ﬂow equations, and an objective function

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

to distinguish among designs. An accurate ﬂow prediction

direction are different. The ﬂow Equations (6) through (8)

requires a viscosity model that includes the effects of shear

for a stationary axi-symmetric geometry are described by

rate. When the viscosity correlations are unknown, the

Gresho & Sani ():

designer must deﬁne appropriate equations prior to model tonian ﬂuid at low shear rates; however, they become shear-

@u v0 2 @u @p 1@ @u0 u0 @ 2 u0 ρ u0 þ w0 þ ¼η r 2 þ 2 þ Fr @r r @z @r r @r @r r @z

thinning ﬂuids under high shear force. A ﬂow analysis

(6)

development. For example, most polymers behave as a New-

requires a continuous viscosity correlation extending over the full shear-rate range observed in the analyzed ﬂuid. The applicability of a model depends on the representation

0 0 u0 v0 1@ @v0 v0 @ 2 v0 0 @v 0 @v r þ þw ¼η 2 þ 2 þ Fφ ρ u r @r @r r @z @r r @z

of the experimental data. When a power-law expression is

(7)

not adequate, a more suitable model, such as those by Carreau-Yasuda, Bird-Carreau and Cross, may be required (Gifford ; Vlachopoulos & Strutt ). Comsol Multiphysics v4.2a (C-M) – a commercial CFD

@w0 @w0 @p 1@ @w0 @ 2 w0 þ Fz (8) ¼η r þ w0 þ þ ρ u0 @r r @r @r @z @r @z2

package – was used to analyze the temperature proﬁle

In this study, the effect of body force components of the

within the reactor geometries (COMSOL MULTIPHYSICS

momentum equation are assumed negligible. Hence, for this

). C-M can handle steady state, transient flow with heat equations using finite element algorithm in both two-dimensional (2D)and 3D geometries. The flow is governed by the time-dependent Navier–Stokes equations. Equation (1) rep-

model the volumetric force components Fr , Fφ and Fz are set to zero. The velocity components in the plane are zero. The velocity components in the angular direction are computed by Equation (9).

resents the vector form of the conservation of momentum equation and Equation (2) is the continuity equation for the conservation of mass. The heat transfer in the ﬂuid

wW ¼ rω

(9)

domain is described by Equations (3) through (5).

@u þ ρ(u:∇)u ¼ ∇ [ pI þ η(∇u þ (∇u)T ] þ F @t

ρ∇:u ¼ 0 ρCp

Qvh

Model geometry (1) We describe how CFD is used to devise a shear reactor that (2)

(3)

2 (∇:u)I :∇u ¼ μ ∇u þ (∇u)T 3

(4)

T @ρ @pA þ u:∇pA ρ @t p @t

With experimental data the CFD model can be ﬁne-tuned for additional analysis. The reactor geometry is rotationally sym-

@T þ ρCp u:∇T ¼ ∇ (k∇T ) þ Q þ Qvh þ Wp @t

Wp ¼

would disinfect human feces by performing a variable analysis.

metric (Figure 1) and can be setup with a 2D cross-section and extended to the original 3D geometry. At the outer cylinder wall a no-slip condition applies. Initial conditions of 0 Pa and

(5)

The terms Qvh and Wp are added to the general heattransfer equation to include viscous heating and pressure work effects. All three velocity components must be included in the model since the velocities in the angular

Figure 1

Reactor geometry used for CFD studies.

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

298 K are used. The thermal properties of high-strength alloy steel were used for the reactor geometry. The outside surface of the reactor was set to free convection with a heat transfer coefﬁcient of 5 W/m2 K. The angle of the extruder allows multiple spacing tests with a single experimental device. By ﬁxing the shell and moving the core axially the spacing can be adjusted. The geometry was meshed using an interactive technique.

Journal of Water, Sanitation and Hygiene for Development

Table 2

04.1

2014

Composition of simulant stool

Ingredient

Percentage

Salt

12.2

Rice

18.5

Soy bean paste

33.8

Water

35.5

A total of 454,902 domain nodes and 3,510 boundary nodes were created within a mesh area of 0.0032 m2 for the 2D cross-section. The 2D cross-section was rotated around the

Instruments Model CVOR 200, East Brunswick, NJ. A

axis of rotation to obtain the full 3D proﬁle. A grid dependency

sample was placed in a plastic cup within a ball mill of a

test was performed to determine the optimum number of grid

THINKY grinder from Phoenix Equipment, Rochester, NY

nodes. Increasing the number of nodes altered the tempera-

to create a paste. A 2.5 ml sample by syringe was passed

ture predictions less than 0.4%. Hence, the number of nodes

inside the serrated cylinder within the rheometer. The exper-

was considered optimum. Table 1 shows the properties used

iment used a 150 μgap and 40 C. The composition of W

for simulations. The viscosity model for red potato was devel-

simulant stool is given in Table 2 (Susana.org ). Based

oped from experimental rheology data and was used to

on these data, we selected red potato as our fecal simulant.

validate the CFD model. The simulations were performed

To prepare feed to operate the extruder, red potatoes were

with stationary ‘Fully Coupled’ solver conﬁguration. A Parallel

boiled for at least one hour. After 10–15 min, the skin was

Sparse Direct Solver was used to obtain simulation results. To

removed and potatoes were mashed in a ricer.

validate the CFD model, experiments were performed, as described in the next section, at various operating conditions.

Shear reactor setup and operation The viscous heating unit is shown in Figure 2. The machine

Experimental methods

operates with a variable pressure (0–690 kPa) for feed rate control, spacing (0.75–1.25 mm) between the rotating

Rheology studies

inner core and the ﬁxed outer shell and rotations per

In order to measure the viscosity of different types of pota-

temperature is assumed constant and feed occupies the spa-

toes and wheat ﬂour, we used a rheometer from Bohlin

cing completely. The simulant mass is placed inside the feed

minute (rpm) (0–1,800). For typical operation, the room

chamber, the spacing is adjusted, pressure and rpm are set Table 1

and the extruder is activated. An electric switch opens the

Properties used for CFD modeling

air valve and moves the plunger that pushes simulant or feces into the gap. Once some initial mass was observed to

Red potato Density, ρ

1,030 kg/m3

exit the shell, a rubber stop cork was used to close the

Viscosity model, μ

0.4 Pa s; t ¼ 0 s 206.4*(Shear rate) 0.92 Pa · s; t > 0 s

outlet for a time (holdup time) to allow the equipment and

Thermal conductivity, k

0.03 W/(m K)

Heat capacity, Cp

4,186 J/(kg K)

High-strength alloy steel Head capacity, Cp

475 J/(kg K)

Thermal conductivity, k

44.5 W/(m K)

Density, ρ

7,850 kg/m3

the mass to heat due to friction. After the desired temperature is achieved the stopcock is removed and mass ﬂow is established. The mass in the annular space has a deﬁned holdup time. During experimentation temperature is recorded by a thermocouple – an Omega HHM 31 Digital Multimeter – and rpm by a digital photo laser tachometer (non-contact). Air pressure, spacing, rpm and torque data are collected.

J. T. Podichetty et al.

Figure 2

Modeling viscous heating of simulant feces

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Instrumented reactor to process fecal and simulant solids.

The plunger on the left in Figure 2 moves inside a cylindrical feed chamber to press the feed inside the shell. A hole on the cylindrical chamber allows the operator to charge the feed. The plunger is air driven and can push the feed with a gauge pressure from 0 to 100 psi. In the ďŹ gure, the pressure gauge is hanging on the metallic vertical wall. Below the gauge, a regulator controls the pressure. On the right side of Figure 2, a Hitachi WJ200 Series 200 V three-phase inverter is attached on the vertical wall to set the required rpm of the AC motor. Not shown in Figure 2, but mounted on the wall power supply, is a wattmeter to measure the energy consumption. By recording the instantaneous power a plot with time for both no-load and load operating conditions allows for calculation of the energy input into the simulant. The cone and the shell are shown in the lower center of Figure 3. The ďŹ xed shell has a spacing calibration ranging from 0 to 1.25 mm. The cone moves inside the shell. In contact with the inner cone surface, mass starts rotating, deforms, and creates friction. The moveable bearing is attached around the stem of the cone with a black circular plastic rubber seal. The combination aligns the cone in a

Figure 3

The cone (left) rotates inside the housing (right).

concentric position inside the cylindrical shell. The black hole shown on the shell in Figure 3 serves as an outlet.

moved to set the desired spacing based on the calibration.

The shell was calibrated for three positions: 0.75, 1.00

The spacing between the housing and rotating shaft is

and 1.25 mm. Spacing between the rotating cone and the

increased if the metal handle is pushed forward to the ver-

stationary shell is ďŹ xed by the operator. A handle is

tical metal wall (anticlockwise direction as you look from

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

the side of the air cylinder inlet). Vegetable dye tracer

). Plots of the same type are found for wheat and

studies conﬁrmed plug ﬂow behavior through the reactor.

maize starch paste in Ellis et al. (). While wheat ﬂour had the closest match with human feces over a range of shear rate, red potatoes were selected as the test simulant

RESULTS AND DISCUSSION

when matched with moisture content and ease of preparation.

Shear rate dependency on viscosity Viscous heating of simulant (red potato) and CFD model validation

Finding a reasonable and safe simulant for experimentation was required. A search of shear thinning ﬂuids (where viscosity decreases with increasing shear rate) indicated

As the core rotates inside the housing, the potatoes pass

several alternatives with similar properties to feces. Figure 4

between the two metallic surfaces. The outlet was closed

presents viscosity data for pig caecal, chicken caecal and

for a certain holdup time and the potatoes gained heat.

human stool compared with several starchy materials and

When allowed to exit the reactor, the elevated temperature

simulant stool (Doucleff ). The viscosity decreased

observed initially reduces as the mass cools due to water

with increase in shear rate. The graphical representation

evaporation and release to atmospheric pressure. Table 3

for fecal matter follows a similar trend line (Woolley et al.

shows data for experiments at 1,800 rpm with a spacing of

Figure 4

Table 3

Shear rate decrease with viscosity for various feces and simulants.

CFD and experimental results: Temperature with time at constant feed pressure, rpm and spacing

Feed

Pressure kPa

rpm

Spacing mm

Holdup time seconds

Temperature outlet C

CFD average ﬂuid temperature C

Red potato

690

1,800

0.75

0 60 120 180 240 260

34 64 74 86 162 190

35 76 103 131 159 168

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

0.75 mm and feed pressure of 690 kPa at various holdup

deviation between CFD predictions and experimental

times. As the holdup time increased, the temperature

measurements can be attributed to the thermal mass of the

increased.

system, heat loss and temperature measurement technique.

The temperature increased with holdup time. A tempW

An average ﬂuid temperature was obtained from CFD

erature of 190 C was obtained at 260 s holdup time.

while outlet temperature was measured during experiments.

Experiments were stopped at this temperature to prevent

A maximum temperature of 218 C was predicted by CFD

excessive wear and tear on the equipment. This temperature

simulations. Figure 6 shows the ﬂuid temperature proﬁle at

exceeded that required to kill all microorganisms in human

260 s. The CFD model can be used effectively to make

feces by thermal energy alone. However, additional

improvements to the current design. Experiments were

microbial destruction is anticipated from the shear stress

also performed by varying reactor spacing at constant

within the extruder. Temperatures for steam sterilization

pressure, rpm and holdup time.

required 121 C for 30 min or 132 C for 4 minutes (Rutala & Weber ).

Table 4 indicates that temperature decreases with increased spacing. These results suggests the likelihood

Figure 5 presents results for experiments with the reac-

that lower spacing, less than 0.75 mm, will generate even

tor feed pressure at 690 kPa, 1,800 rpm and 0.75 mm

higher temperature. However, the temperature requirements

spacing. We performed additional experiments – not pre-

to kill microorganisms have been achieved; hence, further

sented – at constant rpm and spacing and found a similar

reduction was unnecessary. The experiments were done at

trend. The maximum temperature in Figure 5 shows the

constant feed pressure of 690 kPa, 1,209 rpm and 180 s

limitations of the equipment. We believe that increasing

holdup time. After 180 s, the rubber stop cork was removed

rpm and decreasing spacing would increase temperature

and the mass was allowed to leave to the environment. The

further. Additional experimentation on the destruction of microbes was performed and is presented in a companion paper. The data for various holdup times was used to validate the CFD model. Table 3 compares CFD results with data at the same conditions. The CFD results in Figure 5 indicate a linear increase in temperature with holdup, suggesting that increasing hold time would increase ﬂuid temperature. The

Figure 6

Table 4

CFD ﬂuid temperature proﬁle: red potato at 260 s holdup time.

Temperature data with the change of spacing at constant pressure, rpm and holdup time

Pressure

Figure 5

Average ﬂuid temperature from CFD and outlet temperature from experimental measurements at 1,800 rpm and 0.75 mm reactor spacing.

Holdup time

Spacing

Temperature outlet

Feed

kPa

rpm

seconds

Red potato

690

1,209

180

0.75 1 1.25

72 70 50

J. T. Podichetty et al.

Table 5

Modeling viscous heating of simulant feces

Temperature with rpm change at constant pressure, holdup time and spacing

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

larger section of the reactor. An average ﬂuid temperature W

Pressure

Spacing

Holdup time

Outlet temperature

Feed

kPa

seconds

rpm

Red potato

690

0.75

180

913 1,209 1,504

55 72 104

of 114, 168 and 225 C was obtained for reactor lengths 0.05, 0.1 and 0.15 m, respectively. Increase in average ﬂuid temperature was caused by increase in surface area for the same amount of feed. The change in angle of the cone follows the same trend as changes in length. Increasing the angle of the cone increases surface area and viscous heating. A smaller clearance results in higher maximum temperature,

mass was observed to be hot, moist, and dried quickly when

reducing the clearance from 0.75 to 0.5 mm increased the

exposed to air.

average ﬂuid temperature by 68 C. W

In the next attempt, the effect of rpm on temperature

An increase in angular velocity increases viscous heat-

rise is evaluated (Table 5). We see at 690 kPa feed pressure,

ing signiﬁcantly. An average temperature of 168, 127 and

0.75 mm spacing, and 180 s holdup time, that the tempera-

76 C was obtained at 1,800, 1,000 and 500 rpm, respect-

ture rises linearly.

ively. The velocity component in the angular direction

While 120 C is sufﬁcient to sanitize the waste, oper-

(Equation (9)) decreases with decrease in length for the reac-

ation at higher temperature supplies additional energy

tor, resulting in relatively larger surface area for the ﬂuid in

with the primary beneﬁt being increased water evaporation

the larger section of the reactor with corresponding higher

rate. The ‘cost’ for the additional temperature depends on

temperature.

equipment and utility costs. The operation conditions must be based on optimization of these operational (energy

Shear reactor energy balance

cost) and capital (equipment wear) factors. In summary, higher holdup time, lower spacing and higher rpm enhance

The wattmeter installed in the power supply line to the reac-

outlet temperature.

tor motor was used to measure the energy consumption. Potatoes were charged to the reactor and the experiment

Effect of dimension

operated at 690 kPa feed pressure, 0.75 mm gap spacing and 1,800 rpm setting with a holdup time of 260 s (time to

To evaluate the reactor geometry, the length, angle of the

achieve 190 C per Table 3) for two different conditions – W

cone and clearance were varied to deﬁne the effect of dimen-

with and without feed into the reactor. For both cases,

sions on temperature gradient. The temperature proﬁles in

power consumption was recorded every 10 s. A plot of

Figure 7 show an increase in average ﬂuid temperature with

instantaneous power versus time was obtained for both no

greater length. Higher temperatures were observed at the

load and load conditions (Figure 8). Energy input to the feed was calculated by subtraction. The area under the power versus time curve (Figure 8) gives the total energy consumed. For a 260 s run, the energy with load was 243 kJ, and without load was 172 kJ; for a difference of 71 kJ. Most of the energy input to the simulant occurs in the ﬁrst 90 s and then drops to a near steady state value as viscosity is reduced. Correspondingly, temperature increases rapidly initially and then climbs gradually, nearly at steady state. Some energy is retained by the mass, but most is conducted through the metal shell and core. If the experiments are stopped at 90 s, 48% of the energy input would go to viscous energy of the mass

Figure 7

CFD average ﬂuid temperature proﬁle: the effect of length (L).

and most of that energy would transfer out of the reactor

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

temperature change. Temperature was observed to increase with decreasing spacing, increasing rpm and increasing conW

tact time. The maximum temperature achieved was 190 C within 3–4 min with red potatoes. An insulated system would ensure the desired temperature in the shortest time and, in turn, higher mass throughput for a specific temperature and geometry. A significant factor is the requirement of high viscosity to generate heat. For cases where people have diarrhea, where urine is not separated or when water is added, this technology will require feed mass modification to increase viscosity. Possibilities include the addition of paper or biomass, or design modifications that includes recycle of some of the dried solid waste. By balancing the ratio of recycle to fresh feed the required viscosity can be achieved.

ACKNOWLEDGEMENT Figure 8

Power versus time for no load and load conditions at 690 kPa, 0.75 mm spacing and 1,800 rpm.

The authors would like to thank Mr Ronny E. Markum of the Advanced Technology Research Center, Oklahoma

by conduction. The longer the system runs at steady state,

State University for assisting with the design, building the

the lower the percentage of total energy goes into the simu-

equipment and troubleshooting the operation. This work

lant. By the end of the full 260 s run the total fraction of

was supported by the Bill & Melinda Gates Foundation

energy input into the simulant was 29%. Clearly, if viscous

through a Grand Challenges Explorations grant.

heating is the goal then the device should be well insulated and stopped as soon as the required temperature is achieved. In design, for a given rpm and geometry the simulant ﬂow

REFERENCES

rate can be established for a fixed percentage power input between 29 and 48%. However, if a specific temperature is required then a singular efficiency will result for a specific material and system geometry.

CONCLUSIONS CFD studies showed the temperature gradient in the reactor depends significantly on the angular velocity, inlet velocity, spacing and extruder length. The CFD model was validated with experimental data. Experimentation confirmed that viscous heating by extrusion may be effective in decontaminating fecal wastes. For each experiment, fixing two variables from among rpm, holdup time and gap spacing allowed one variable to be compared with the resulting

Bethony, J. S., Brooker, M., Albonico, S. M., Geiger, A., Loukas, D., Diemert, R. J. & Hotez, P. J.  Soil-transmitted helminth infections: ascariasis, trichuriasis, and hookworm. The Lancet 367 (9521), 1521–1532. Burg, J. P. & Gerya, T.V.  The role of viscous heating in Barrovian metamorphism of collisional orogens: thermomechanical models and application to the Lepontine Dome in the Central Alps. Journal of Metamorphic Geology 23 (2), 75–95. Collins, M. W.  A finite difference analysis for laminar heat transferr of non-newtonian fluids in circular tubes. Numerical Methods in Thermal Problems, Volume 3 – Proceedings of the Third International Conference, Seattle, WA, August 2–5, 1983, pp. 540–550. COMSOL MULTIPHYSICS  COMSOL Multiphysics User’s Guide. Stockhlom, Sweden: Version 4.2a. Costa, A. & Macedonio, G.  Viscous heating effects in fluids with temperature-dependent viscosity: triggering of secondary flows. Journal of Fluid Mechanics 540, 21–38.

J. T. Podichetty et al.

Modeling viscous heating of simulant feces

Doucleff, M.  Why is the world’s largest foundation buying fake poop? www.npr.org/blogs/health/2012/08/08/ 158447235/why-is-the-worlds-largest-foundation-buying-fakepoop (Accessed: 25 August 2013). Ellis, H. M., Ring, S. G. & Whittam, M. A.  A comparison of the viscous behaviour of wheat and maize starch pastes. Journal of Cereal Science 10, 33–44. Gifford, W. A.  The use of three dimensional computational fluid dynamics in design of extrusion dies. Journal of Reinforced Plastics and Composites 16, 661–674. Gresho, P. M. & Sani, R. L.  Incompressible Flow and the Finite Element Method, Volume 2, Isothermal Laminar Flow. Wiley, New York. Hooman, K. & Ejlali, A.  Effects of viscous heating, fluid property variation, velocity slip, and temperature jump on convection through parallel plate and circular microchannels. International Communications in Heat and Mass Transfer 37 (1), 34–38. Lawal, A. & Kalyon, D. M.  Viscous heating in nonisothermal die flows of viscoplastic fluids with wall slip. Chemical Engineering Science 52 (8), 1323–1337. Rutala, W. A. & Weber, D. J.  Guideline for Disinfection and Sterilization in Healthcare Facilities. CDC, USA. Santo Domingo, J. W., Lu, J., Shanks, O. C., Lamendella, R., Kelty, C. A. & Oerther, D. B.  Development of host-specific metagenomic markers for microbial source tracking using a novel metagenomic approach. Disinfection 16, 646–661. Singh, N., Isono, N., Srichuwong, S., Noda, T. & Nishinari, K.  Structural, thermal and viscoelastic properties of potato starches. Food Hydrocolloids 22 (6), 979–988. Sunden, B.  Vicous heating in forced convective heat-transfer across a circular-cylinder at low reynolds-number.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

International Journal for Numerical Methods in Engineering 35 (4), 729–736. SuSanA – Sustainable Sanitation Alliance  MISO (Soybean Paste) 14 March 2008. http://forum.susana.org/media/ kunena/attachments/52/certificate.pdf (Accessed: 03 June 2012). Trönnberg, L., Hawksworth, D., Hanses, A., Archer, C. & Stenstrom, T. A.  Household-based prevalence of helminths and parasitic protozoa in rural KwaZulu-Natal, South Africa, assessed from faecal vault sampling. Transactions of the Royal Society of Tropical Medicine and Hygiene 104 (10), 646–652. Vlachopoulos, J. & Strutt, D.  The Role of Rheology in Polymer Extrusion. New Technologies for Extrusion Conference, Milan, Italy. Woolley, S., Buckley, C., Pocock, J., Cottingham, R. & Foutch, G.  Some rheological properties of fresh human feces with a variation in moisture content & Rheological modeling of fresh human feces. 3rd SA YWP Conference, Stellenbosch, Western Cape, South Africa. World Health Organization  Water, sanitation and hygiene links to health. www.who.int/water_sanitation_health/ publications/facts2004/en/ (Accessed: 12 Sep 2012). Yavuz, T., Erol, O. & Kaya, M.  Heat transfer characteristics of laminar annular duct flow with viscous dissipation. Proceedings of the Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Science 225 (C7), 1681–1692. Yesilata, B.  Viscous heating effects in viscoelastic flow between rotating parallel-disks. Turkish Journal of Engineering & Environmental Sciences 26 (6), 503–511.

First received 6 May 2013; accepted in revised form 5 September 2013. Available online 21 October 2013

04.1

2014

Power earth auger modiﬁcation for waste extraction from pit latrines Tate Weston Rogers, Francis L. de los Reyes III, Walter J. Beckwith and Robert C. Borden

ABSTRACT The extraction auger was developed to meet the need for a low cost, effective method to empty pit latrines in difficult to access locations. The basic design consists of a motor that rotates an auger inside a pipe, lifting waste from a pit and depositing it into containers through a wye fitting at the top of the device. Laboratory testing of the auger showed increases in flow rates with increasing auger rotational speed and waste viscosity. An auger with an external hydraulic drive was capable of lifting dairy waste over 2.5 m, at flow rates of over 125 liters per minute. Field-testing showed the equipment was effective at lifting medium viscosity wastes containing a mixture of liquid and

Tate Weston Rogers Francis L. de los Reyes III Walter J. Beckwith Robert C. Borden (corresponding author) Department of Civil, Construction and Environmental Engineering, North Carolina State University, Campus Box 7908, Raleigh, NC 27695, USA E-mail: rcborden@ncsu.edu

solid material. However, the auger was not effective in removing low viscosity, liquid waste that would flow backward down the auger reducing lifting efficiency. The auger was capable of drilling into dense solid waste, forming a ‘posthole’ in the waste. However, since the dense solid waste would not flow towards the auger intake, actual waste removal from the pit was limited. Improved methods are needed to mix liquid and solid waste in pits prior to removal with the extraction auger or other technologies. Key words

| auger, pit emptying, portable

INTRODUCTION Pit latrines can provide improved sanitation at a relatively

of up to 2 m (Mara ; Still ). ‘Wet’ pits containing a

low cost, but require periodic emptying to remain effective.

substantial amount of liquid material are reported to separ-

However, physically accessing the pits and removing the

ate into three layers: ﬂoating scum, a liquid layer, and a

accumulated waste is a major challenge in many developing

sludge layer (Hawkins ) where the consolidated sludge

areas. Mechanical emptying equipment used in developed

layer is reported to have densities of up to 1,750 kg/m3 (Rad-

countries such as vacuum tankers cannot access pit latrines

ford ). However, Seal () found that the contents of

located far from paved roads and often only removes the

pits in Kibera, Nairobi were much more variable. While

liquid fraction, leaving dense solids in the pit which are dif-

the contents of some pits were consistently soft and others

ﬁcult to remove. In the absence of an effective and

consistently hard, some pits had multiple layers with hard

economical alternative, many pits are emptied by manual

layers sometimes overlying watery layers. ‘Dry’ or well-

laborers (manual extraction) exposing workers and the sur-

drained pits usually only contain the sludge layer and

rounding community to a variety of pathogens (Still &

show a trend of decreasing moisture content from the sur-

Foxon ).

face of the pit to a 1 m depth and little to no change past

Waste pit characteristics and their contents vary greatly,

that depth (Still & Foxon ). Waste accumulation rates

further complicating the emptying process. Pit volumes are

can vary tremendously, with pit ﬁll times ranging from 3

reported to vary between 1 and 4 m with typical depths doi: 10.2166/washdev.2013.183

to 20 years (Still ). Fill rates vary depending on the

T. W. Rogers et al.

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

number of users, soil type and depth to water table, material

& O’Riordan () described the development of an electric

used for anal cleansing, and the amount of rubbish (trash)

motor powered screw auger for waste removal from pits, but

added to the pit (Pickford ; Still ). In some

reported problems include: (a) difﬁculty in transporting,

countries, up to 20% of the pit contents may be solid

maneuvering, and cleaning the auger due to the large size

waste (trash), which can cause major operational problems

and weight of the equipment; (b) clogging with rubbish in

with many mechanical pit-emptying devices (Still ).

the pit; and (c) difﬁculty with use in shallow or deep pits

Some pits are only accessible through the toilet hole inside

or inside a toilet superstructure because of the ﬁxed size of

the latrine superstructure or by drilling holes in the side of

the equipment.

the latrine, further complicating the emptying process (Still ). A variety of devices have been developed to aid in pit

Based on the work of Still & O’Riordan () and others, the following performance criteria were developed for the proposed extraction auger.

emptying. Manual extraction tools include buckets, shovels,

(a) Hydraulic performance: The auger should be capable of

and specialized scoops that allow laborers to remove the

lifting 40 liters per minute (Lpm) of semi-solid sludge 2–

waste without having to physically enter the pit. Pumping

3 m allowing a typical 2–4 m3 pit to be emptied in 1–2 h.

systems include: (a) traditional diaphragm pumps; and (b)

(b) Portable and maneuverable: The auger should weigh

the Gulper, a hand powered pump that draws waste through

less than 50 kg allowing hand transport by two laborers

a foot-valve submerged in waste, forcing it up a solid pipe

to remote pit locations. Once inserted into the pit, the

through a wye ﬁtting and into a collection device (Ideas at Work ). Mechanical and manual pumps can effectively remove liquid wastes, but are prone to clogging, have increased wear due to the direct contact between the waste and mechanical components, and cannot pump dense or dry sludge (Thye et al. ). Vacuum systems include: (a) traditional vacuum trucks; (b) smaller, more maneuverable mechanized vacuum systems including the Microvac ( Jere et al. ), Dung Beetle (Tucker ), and

auger should be easily maneuverable by one operator while a second person directs the waste into containers for off-site transport. (c) Simple and durable: The auger should be simple to operate and maintain using locally available materials and labor. The auger should be sufﬁciently durable for daily use with little maintenance, and should be capable of lifting waste containing some rubbish without damage to the device.

Vacutug (Wegelin-Schuringa & Coffey ); (c) the

(d) Sanitary: The auger should allow safe and effective emp-

MAPET system, a hand powered piston pump and tank

tying of pit latrines while providing for hygienic

mounted on a small pushcart (Muller & Rijnsburger );

separation of waste. This consideration includes equip-

and (d) the eVac, a lightweight vacuum system powered by

ment transport, assembly, waste transfer from the pit

gasoline or electric motors that deposit the waste directly

into the transport containers, equipment cleanup and

into 47-liter plastic containers for off-site transport (Still &

removal from the site. If rubbish does get caught inside

O’Riordan ). Important advantages of vacuum systems include: (a) reduced wear and clogging of mechanical equipment by direct contact with the waste; and (b) reduced contact with waste, odors and ﬂy nuisance (as long as the

the auger, the device should allow simple and hygienic removal of the trash. (e) Manufacture: Ideally, it should be possible to manufacture the auger using locally available materials.

pipes remain unblocked). However, many vacuum systems cannot extract waste from more than 0.8 m below the ground surface and have difﬁculty removing dense and/or dry sludge (O’Riordan ; Thye et al. ; Still & Foxon

EXTRACTION AUGER DESIGN

; Still & O’Riordan ). In this work, we report on the development and testing

The extraction auger described here was developed through

of a screw auger inside a pipe to extract waste from pits and

a process of design, fabrication and testing, followed by rede-

deposit the waste into containers for off-site transport. Still

sign and additional testing. The design described below

T. W. Rogers et al.

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

meets many of the criteria presented above, and provides the

The main body of the extraction auger consists of a

basis for further development of effective pit emptying

straight section of high-density polyethylene (HDPE) auger

technologies.

ﬂights (9.5-cm outer dia.) mounted on a 2.5 cm dia. stainless

The basic design consists of a continuous ﬂight auger

steel hex center shaft (Lundell Plastics Corp., Odebolt, IA).

that rotates inside a pipe, transporting waste upward,

The HDPE ﬂights are enclosed in 10 cm nominal dia. sche-

before discharging through a 45 wye ﬁtting into containers

dule 40 polyvinyl chloride (PVC) pipe. The top end of the

for off-site transport (Figure 1). In our ﬁnal design, the auger

PVC pipe is glued to a 10 cm × 10 cm × 10 cm PVC wye ﬁt-

is rotated by a hydraulic motor that provides high power and

ting. Waste moves upward through the auger to the wye,

is easily reversible to discharge debris that enters the auger.

turns 270 , and discharges out the wye through a collapsible

The hydraulic motor is powered by a remotely located gaso-

10-cm PVC hose. A handle attached to the discharge end of

line driven hydraulic pump.

the collapsible hose assists in directing the waste into a suit-

able collection container. At the bottom (intake end) of the auger, two flights of steel auger extend past the end of the PVC pipe to convey waste into the solid pipe portion of the auger. Steel replaces the HDPE in this section to improve the durability of the exposed flights. A 2.5 cm dia. steel ball can be welded to the tip of the steel auger to prevent the auger from damaging the bottom of the pit, if desired. The length of the extraction auger can be increased to empty deeper pits by attaching 30, 60 or 90 cm lengths of HDPE auger and PVC pipe. Inside the wye at the top of the screw, there is a 20 cm long section of double helix reverse flights to aid in discharging the solids through the wye into a collection container. Field testing showed the reverse flights and wye are needed to prevent accumulation of compacted material inside the upper portion of the auger. In the current version, the reverse flights are manufactured from welded steel plate. However, weight could be reduced by replacing the reversed steel flights with HDPE or other lightweight material. The auger and wye fitting are connected to the hydraulic motor using a fabricated coupling machined from 15-cm diameter nylon round stock. The top 5 cm of the PVC wye fitting slides into the coupling to provide a secure fit and is held in place by four cap screws. A fabricated steel shaft passes through the fabricated coupling, connecting the output shaft of the hydraulic motor to the auger. The hydraulic motor used in our design is an Eaton Char-Lynn model 103–1,034 (75 cm3 per rotation displacement) and is rated for continuous operation at 493 Figure 1

Final extraction auger design with pipe cut-away view showing metal cutting head, plastic auger ﬂights to transport waste inside plastic pipe, wye discharge connection, reverse ﬂights at top of auger to aid in waste discharge, hydraulic motor, handle bars and controls. External gasoline-driven hydraulic pump not shown.

revolutions per minute (rpm), providing a torque of 148 Nm at a ﬂow of 38 Lpm and maximum pressure of 138 bar. Hydraulic power to drive the motor is supplied by a 10.7 HP Honda GX340 gasoline engine (Little Beaver

T. W. Rogers et al.

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Inc., Livingston, TX). Hydraulic ﬂuid is transferred from the

oriented vertically, when auger performance was expected

hydraulic pump to the hydraulic motor through 7.6 m

to be lowest (Roberts ; Dixon & Humphries ). In

lengths of hydraulic hoses with quick-connect ﬁttings. The

all of the laboratory tests, a 1.5 m long auger was used

hydraulic motor, forward and reverse controls, pressure

with a 10 cm × 10 cm × 5 cm tee ﬁtting outlet. Clear man-

gage and extraction auger are mounted to a handle bar

ometer tubes were attached to the sides of the auger at

frame constructed from 1.9-cm steel pipe. Total assembled

four locations to examine the change in head produced

weight of a 2.7 m long unit is 48 kg excluding hydraulic

over the length of the auger. Auger speed was monitored

pump and hoses.

with a laser tachometer. The auger discharge rate was measured using a stopwatch and a 19 L pail. For the initial testing phase, a simulated waste was pre-

LABORATORY TESTING

pared from high yield bentonite clay and water. Mixtures of 5, 6, and 7% by weight bentonite were used to generate

The hydraulic performance of the auger was ﬁrst evaluated

shear thinning materials with effective viscosity varying

in the laboratory using a bentonite slurry to simulate fecal

from 0.3 to 700 Pascal-seconds (Pa-s) resulting in shear

waste by monitoring pumping rate and discharge pressure

rates of 0.2 and 21 L/second (L/s), with higher bentonite

while varying auger rotational speed (100, 200, 300, 400

concentrations yielding higher viscosities (Rogers ).

and 500 rpm), auger submergence, and slurry viscosity. For

Advantages of the bentonite mixtures include: (a) thixotro-

these initial tests, the auger was driven by a 1 HP electric

pic properties that are similar to human waste; (b) absence

motor with a rheostat to control the auger rotational

of pathogens; and (c) biological stability, allowing testing

speed. The motor was attached to a metal stand that held

over extended periods in the laboratory.

the prototype vertical and allowed for different height set-

Figure 3 shows the variation in ﬂow rate for three differ-

tings (Figure 2). All tests were conducted with the auger

ent solids contents and two different submergences (intake

Figure 2

Experimental setup diagram for extraction auger.

T. W. Rogers et al.

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

limiting for more viscous materials, especially as the pits are emptied, reducing auger submergence. Field-testing of a similar screw conveyor in South Africa found the flow rate to decrease by 80% as the pit neared empty (Still & O’Riordan ). In a series of papers, Roberts (, , ) developed relationships to describe transport of non-cohesive, granular material in screw conveyors. Screw conveyor efficiency is a function of the fill efficiency and vortex efficiency. For materials with a high friction coefficient, conveyance is limited by the rate that material enters the screw Figure 3

Flow rate vs. auger rotational speed for 1.5 m long × 10 cm diameter auger for

conveyor intake (ﬁll efﬁciency). However, for materials with

shallow (30 cm submergence) and deep (45 cm submergence) settings with 5 cm of exposed auger and 5, 6 and 7% bentonite mixtures.

a low friction coefﬁcient, conveyance is limited by backwards ﬂow through the screw, especially when the conveyor is oriented vertically. Roberts’ work indicates

30 and 45 cm below sludge level) with 5 cm of auger extend-

that conveyance will be zero at low rotational speeds

ing beyond the pipe end. Auger rotational speed and waste

when the vortex efﬁciency is negative. As the friction coefﬁ-

viscosity had a major impact on the ﬂow produced with a

cient increases, less rotational speed is required to achieve

speed of over 200 rpm required to pump the less viscous 5

ﬂow.

and 6% material. In contrast, the more viscous 7% mixture

While the relationships developed by Roberts (,

could easily be pumped at 200 rpm. A pumping rate of at

) for predicting conveyance of non-cohesive granular

least 40 Lpm is required to empty a typical pit of 2–4 m3

materials are not directly applicable to the cohesive, thixo-

in 1–2 h, which would require a rotational speed of at least

tropic bentonite mixtures, the experimental results we

300 rpm. Typical direct drive earth augers have maximum

obtained were generally consistent with his approach. For

speeds of 180–250 rpm, indicating standard ‘off-the-shelf’

low viscosity mixtures (analogous to materials with a low

augers may not be appropriate for pit emptying applications.

friction angle), ﬂow was zero for low rotational speeds. As

The discharge pressure generated by the auger increased

the bentonite viscosity increased (analogous to materials

with rotational speed, but was low for all conditions tested

with a higher friction angle), lower rotational speeds were

(maximum pressure head produced ¼14 cm) indicating: (a)

sufﬁcient to generate ﬂow. In our testing, the auger achieved

the auger will not generate sufﬁcient pressure to pump

a maximum volumetric efﬁciency of 26%, which is also gen-

material through any signiﬁcant length of hose; and (b) the

erally consistent with Roberts’ work. This relatively low

auger will need to extend above the collection container

efﬁciency is due to the vertical orientation of the auger,

and discharge by gravity into the container.

which increases the slip of material backwards through the

Changes in intake submergence had minimal impact on

auger (Roberts ). Pumping efﬁciency increases signiﬁ-

ﬂow rates for the 5 and 6% bentonite mixtures, suggesting

cantly if the auger is placed at an angle, reducing

that ﬂow of low viscosity materials into the auger intake

backwards ﬂow through the auger.

was not rate limiting. However, for the more viscous 7% bentonite mixture, the smaller intake submergence did result in a reduction in pumping rate. An increase in

ANIMAL WASTE TESTING

length of exposed auger beyond the pipe intake (choke length) had minimal impact on ﬂow rate for the lower vis-

Following the laboratory testing and a series of initial ﬁeld

cosity materials, but resulted in a small increase in ﬂow

tests to reﬁne the equipment, the hydraulic motor-powered

for the highest viscosity material (data not shown). These

auger design described above was tested on waste from the

results suggest that ﬂow into the auger intake can be rate

dairy cattle barns at the North Carolina State University

T. W. Rogers et al.

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Lake Wheeler Farm. The test material consisted of a mixture

To resolve this problem, the outside diameter of the HDPE

of cow manure and a small amount of sand or wood shav-

auger ﬂights were reduced slightly, increasing the gap

ings that was washed out of the dairy barns and collected

between the ﬂights and the pipe to 3.2 mm from 0.8 mm.

into a settling basin. This material had a moisture content

This change greatly reduced jamming, increased rotational

of 80.9%, organic content of 9.7%, inorganic content of

speed from 380 to 410 rpm, and reduced the hydraulic

9.5% and contained solid particles (wood shavings) up to

pressure required to drive the auger from 12.4 to 8.3 MPa,

1 cm in length. The waste moisture content of the waste is

but did result in a small reduction in ﬂow. Roberts ()

within the range of moisture contents reported for fresh

suggests a clearance of between 1.5 and 3 times larger

human feces (Still & Foxon ). Figure 4 shows ﬁeld test-

than the maximum particle size to prevent jamming and

ing of a 2.7 m long unit with a 15 cm choke length to

energy loss as well as limiting slip back and a loss of

pump dairy waste out of a 1 cubic meter container with a

efﬁciency.

36 cm × 90 cm opening on the top. The extraction auger is

Lockup or jamming of the auger was also reduced by

being operated from a lift to simulate pumping waste out

increasing the maximum hydraulic pressure (controlled by

of the bottom of 2 m deep pit latrine.

a pressure relief valve) from 10.3 to 13.8 MPa. The increased

When the auger was ﬁrst used with dairy waste, there

torque provided by the higher pressure was able to dislodge

were repeated problems with jamming of the auger as

most material without locking up the auger. When the auger

small pieces of wood shavings became lodged between the

did jam, the hydraulic controls on the handle bar allowed

outside edge of the auger ﬂights and the pipe interior.

the auger rotation to be easily reversed and dislodge any

While the wood shavings could be easily dislodged by rever-

trapped material. In all cases, a 1–2 second reversal of the

sing the auger for 1 to 2 seconds, this reduced productivity.

auger was effective in dislodging trapped material. This allowed continued pumping of the waste without any other action. The final design generated a flow rate of 127 Lpm when dairy waste was lifted 2 m out of the bottom of a 1 m deep waste container. However, flow declined to 88 Lpm as the waste container was emptied with the lift increasing from 2.0 to 2.35 m and intake submergence decreasing from 0.7 to 0.35 m. As waste level in the tank declined below the pipe intake, a second person would need to push the waste towards the auger intake with a shovel to continue pumping. In practice, this may not be necessary since it can be beneficial to leave a small amount of waste in the bottom of the pit, enhancing degradation of future waste (Still & O’Riordan ).

FIELD TESTING Over a four-day period in March 2013, our modiﬁed design was tested on a total of six pit latrines in South Africa. The hydraulic powered extraction auger was easily carried to all of the pits investigated. While it was not possible to drive a Figure 4

Extraction auger being used to pump dairy waste 2 m upward from a 1 m3 container.

vehicle close to many latrines, access with the equipment was not a problem. Overall, the machine was effective at

T. W. Rogers et al.

Figure 5

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Extraction auger being used to empty pit latrine in Durban, South Africa.

pumping medium viscosity wastes containing a mixture of

dry layer ﬂoating on top of liquid waste. The current

liquid and solid material. Figure 5 shows the pumping oper-

design could penetrate the dry ﬂoating layer, but was not

ation. Five 75-liter containers were pumped out over about

effective in mixing the wet and dry material to generate a

20 minutes of operation.

pumpable slurry.

In general, clogging with trash and other debris was not

Additional work is needed to develop effective methods

a problem. Reversing the auger ejected most debris. While

to empty pits containing solid material that does not ﬂow

elastic bands (presumably from disposable diapers) did

into the auger intake. Potential solutions include develop-

occasionally get wrapped around the auger, they did not

ment of equipment that could mix the liquid and solid

result in jamming of the equipment. When trash was

material followed by pumping using the extraction auger

caught in the auger, the machine could be easily reversed,

and/or vacuum equipment. Additional needs include: (a)

ejecting most of the trash and other debris.

development of methods to determine the amount and char-

The major operational limitation with the auger was

acter of waste through the toilet hole, prior to breaking the

associated with extracting either very low or very high vis-

pit seal; and (b) standardized sanitation protocols to

cosity waste. As shown during the initial laboratory

reduce the spread of waste and fecal borne pathogens

testing, the auger is not efﬁcient for pumping watery, low vis-

during pit emptying.

cosity waste, and other available technologies (Gulper, Vacutug or a simple trash pump) are more effective. The equipment was also not effective in removing solid waste

CONCLUSIONS AND RECOMMENDATIONS

in dry pits without some liquid to allow the material to ﬂow to the auger intake. The auger was able to drill into

A relatively simple, low cost extraction auger was developed

high solids waste, forming a ‘posthole’ in the non-ﬂowing

to remove semi-solid sludge from waste pits. The current

solid waste. Problems were also encountered in pits with a

design is capable of lifting dairy waste over 2.5 m at ﬂow

T. W. Rogers et al.

Waste extraction auger

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

rates of 125 Lpm, allowing typical waste pits to be emptied

added to the auger inlet to reduce damage to the inlet of

in 1–2 hours. A rotational speed of at least 350 rpm is

the PVC pipe. Further modiﬁcations may be needed to

needed to lift less viscous, ‘water-like’ wastes. The auger pro-

allow manufacture using locally available materials.

duces minimal pressure at the discharge end so the waste

The most critical need is to develop methods to effec-

must be discharged directly into another container for off-

tively empty pits containing both low viscosity liquid waste

site transport. In general, jamming or clogging with trash

and dense solid waste. It is not unusual to encounter pits

and other debris is not a major problem.

containing dense waste at the bottom with one or more

The current design can be easily transported and used by

layers of liquid and then solid material above. While the

two men. A 2.7 m long unit weighs 48 kg and can provide

extraction auger can penetrate the solid material, it often

adequate ﬂow while still being maneuverable with one or

forms a ‘posthole’ in the solid material, removing only a

two operators. Wrap-around handlebars and a support

small portion of the solid waste. Better methods are

stand allow reasonably easy operation by one person. Use

needed to mix the solid and liquid material prior to removal.

of an external hydraulic drive signiﬁcantly reduced the

This will improve the performance of the extraction auger,

weight, while providing sufﬁcient power. The extraction

vacuum methods and other extraction technologies.

auger can be assembled inside a latrine structure using several shorter lengths of auger and pipe. However, this is cumbersome and can increase the risk of waste spillage

ACKNOWLEDGEMENTS

during disassembly after the pit has been emptied. The unit is also somewhat difﬁcult to maneuver in pits full of

We gratefully acknowledge the ﬁnancial support provided by

waste, so further reductions in weight would be beneﬁcial.

the Bill & Melinda Gates Foundation for this project. We also

The reverse mechanism allows waste remaining inside

wish to thank the great people at the NC State Dairy Farm

the auger to be discharged back into the pit, reducing the

and Partners in Development (PID) for their wonderful

equipment weight and waste spillage outside the pit. A

assistance with the ﬁeld-testing portions of this work.

rubber squeegee with a long handle allows the operator to wipe down the outside of the pipe during emptying, further reducing spillage. Between pits, the auger inlet and outlet can be capped to minimize exposure to the waste and reduce the time spent cleaning the equipment. When the auger needs to be disassembled, ﬂushing a steady stream of water through the outlet while running the mechanism in reverse is effective in removing most waste. However, further work is needed to develop better cleaning methods as there is an ongoing risk of waste spillage during the emptying process and associated human exposure. In general, the auger was reasonably durable and there was little evidence of damage during routine operation. Field-testing showed some scoring of the inside of the PVC pipe casing and some feathering deterioration of the edge of the HDPE auger. On one occasion, a small piece of pipe at the auger intake broke off, presumably due to material wedging between the inlet of the pipe and the edge of auger ﬂights. A 20 cm section of metal auger was used at the end of the HDPE auger to reduce exposure of the weaker HDPE material. A short section of metal pipe could also be

REFERENCES Dixon Jr, M. P. & Humphries, E. G.  Evaluation of an Archimedean screw pickle pump/conveyor. Applied Engineering in Agriculture 11 (5), 631–635. Hawkins, P.  Emptying on-site Excreta Disposal Systems in Developing Countries: An Evaluation of the Problems. WHO IRCWD, Dübendorf, Switzerland. Ideas at Work  The Gulper – a manual latrine/drain pit pump. http://www.ideas-at-work.org/pdf/Gulper_pit_emptying_ device.pdf (accessed 15 June 2013). Jere, M., Dzotizei, A. T. & Munjoma, M.  Pit-latrine-emptying using motorized equipment. In Sustainability of Water and Sanitation Systems: Proceedings of the 21st WEDC Conference (J. Pickford, ed.). 4–8 September 1995, Kampala, Uganda. Loughborough, UK. Mara, D. D.  The Design of Ventilated Improved Pit Latrines. The International Bank for Reconstruction and Development/World Bank, Washington, DC. Muller, M. S. & Rijnsburger, J.  MAPET: An appropriate latrine-emptying technology. Waterlines 13 (1), 24–27. O’Riordan, M.  Investigation into Methods of Pit Latrine Emptying. Partners in Development, Pietermaritzburg, South Africa.

T. W. Rogers et al.

Waste extraction auger

Pickford, J.  Low-cost Sanitation: A Survey of Practical Experience. Intermediate Technology, London. Radford, J.  Fluidization of Synthetic Pit Latrine Sludge. University of Cambridge, Cambridge, UK. Roberts, A. W.  An investigation of grain vortex motion in relation to the performance within vertical grain augers. Proceedings of the Institution of Mechanical Engineers 178, 293–310. Roberts, A. W.  The influence of granular vortex motion on the volumetric performance of enclosed screw conveyors. Powder Technology 104, 56–67. Roberts, A. W.  Design Considerations and Performance Evaluation of Screw Conveyors. Report prepared for Beltcon 11, International Materials Handling Conference. Retrieved from http://www.beltcon.org.za/docs/b1114.pdf (accessed 24 September 2013). Rogers, T. A.  Modification of Power Earth Augers for Pit Latrine Extraction in Developing Countries. Master of Science in Environmental Engineering, North Carolina State University, Raleigh, NC. Seal, D.  Identifying Appropriate Pit-Emptying Technologies in a Nairobi Slum Based on Sludge Characteristics and Access. Cranfield University, UK.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Still, D. A.  After the Pit Latrine is Full … What Then? Effective Options for Pit Latrine Management. Report prepared for the Biennial Conference of the Water Institute of Southern Africa. Still, D. & Foxon, K.  Tackling the Challenges of Full Pit Latrines, Volume 1: Understanding sludge Accumulation in VIPs and Strategies for Emptying Full Pits. Report to the Water Research Commission. Still, D. & O’Riordan, M.  Tackling the Challenges of Full Pit Latrines, Volume 3: The development of pit emptying technologies. Report to the Water Research Commission. Thye, Y. P., Templeton, M. R. & Ali, M.  A critical review of technologies for it latrine emptying in developing countries. Critical Reviews in Environmental Science and Technology 41 (20), 1793–1819. Tucker, J.  Design, Build and Test of a Semi Mechanised Pit Latrine Emptying System for the Developing World, Master of Engineering in Mechanical Engineering, University of Bristol, Bristol, UK. Wegelin-Schuringa, M. & Coffey, M.  A small pit emptying machine: an appropriate solution for a Nairobi slum. Schriftenr Ver Wasser Boden Lufthyg 105, 347–352.

First received 30 June 2013; accepted in revised form 5 September 2013. Available online 21 October 2013

04.1

2014

Modeling of pathogen inactivation in thermal septic tanks T. Koottatep, S. Phuphisith, T. Pussayanavin, A. Panuvatvanich and C. Polprasert

ABSTRACT Thermal application has been widely used for pathogen inactivation in various ﬁelds. The purpose of this research was to develop a model of pathogen inactivation in septic tanks operating at various temperatures. Four laboratory-scale septic tanks fed with septage were operated at temperatures of W

30, 40, 50 and 60 C and Escherichia coli (E. coli) was selected as the pathogenic indicator. The efﬁciencies of E. coli inactivation were found to increase with increasing temperatures, while the W

opposites were observed for chemical oxygen demand (COD) reduction. At 60 C, the E. coli concentrations were reduced from 9.6 × 106 to about 10 most probable number (MPN)/100 mL or 6 log reduction. The kinetics of E. coli reduction followed a modiﬁed Weibull model which could be applied to septic tank design and operation. The percentage COD removal was found to be 93, 94, 89 and 84 at

T. Koottatep S. Phuphisith T. Pussayanavin (corresponding author) A. Panuvatvanich Environmental Engineering and Management, School of Environment Resources and Development, Asian Institute of Technology, Thailand E-mail: Tatchai.Pussayanavin@ait.ac.th C. Polprasert Department of Civil Engineering, Faculty of Engineering, Thammasat University, Thailand

temperatures of 32, 40, 50 and 60 C, respectively. The results of this study suggested that pathogenic microorganisms in septic tanks could be inactivated to be at a safe level with thermal application. Key words

| E. coli, methanogens, modiﬁed Weibull model, pathogen inactivation, septic tanks, temperatures

INTRODUCTION

ABBREVIATIONS bT

temperature-dependent coefﬁcient, d 1

At present more than 2 billion people worldwide still lack

BOD5 5-day biochemical oxygen demand

access to improved sanitation and more than 15 million

COD

chemical oxygen demand

people, especially from developing countries, die every

day

year from infectious and parasitic diseases as a result of

hour

poor sanitation (WHO ). Centralized wastewater treat-

ﬁrst-order rate constant at temperatures

ment systems, practised in most developed countries, do

liter

not seem appropriate for developing countries because of

number of E. coli at time t (MPN/100 mL)

high investment and operation costs and unavailability of

initial number of E. coli (MPN/100 mL)

skilled manpower. Septic tanks are the most commonly

Weibull coefﬁcient

used onsite sanitation technologies to treat household waste-

retention time, d

waters

areas

without

centralized

sewer

systems

temperature, C

(Rybczynski et al. ). However, septic tanks constructed

TKN

total Kjeldahl nitrogen

in most developing countries usually do not have leaching

total solids

fields or drainage fields to further treat septic tank effluent.

VSS

volatile suspended solids

Instead, these septic tank efﬂuents are allowed to seep into

regression factor

surrounding soils or are directly discharged into nearby

doi: 10.2166/washdev.2013.190

T. Koottatep et al.

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

storm sewers or water courses. Owing to the short retention

04.1

2014

MATERIALS AND METHODS

time (about 1–3 days) in the septic tanks, the septic tank efﬂuent is still very polluted, containing high concentrations

This study employed laboratory-scale septic tanks and sep-

of organic matter, nutrients and pathogenic microorganisms

tage samples collected from household communities in

(Leclerc et al. ; Eriksson et al. ). Thus, these septic

central Thailand. E. coli bacteria were used as a pathogenic

tank efﬂuents can contaminate nearby surface and ground

indicator.

water resources or soil, causing water pollution and health risks to the public. Since septic tanks will continue to be an onsite sanitation technology in developing countries, methods to improve septic tank performance to alleviate the above problems should be considered. It is a well-known fact that temperature could enhance pathogen die-offs (Carrington ; Winblad & Simpson-Hébert ) and increase microbial activity or biodegradation of organic matter (Metcalf & Eddy ). Therefore, operating a septic tank at temperatures higher than ambient conditions should theoretically

produce

better

quality

septic

tank

efﬂuent.

Thermal treatment of wastewater sludge and raw sewage was previously conducted by Mocé-Llivina et al. (). Their results revealed that Escherichia coli numbers in the sludge and raw sewage samples were reduced by about W

3.6–6 log within 30 minutes of heat exposure at 60–80 C. Pandey & Soupir () found that anaerobic digesters treating dairy manure offered the highest E. coli inactivation at W

52.5 C when compared to the results obtained at 37 and W

25 C. To date, there have been very few reports on thermal application to septic tanks, only to thermophilic anaerobic digesters. The overall objective of this study was to investigate the effect of temperatures on pathogen inactivation

Operation of laboratory-scale septic tanks Four laboratory-scale septic tanks, each with a dimension of 64 cm × 25 cm × 40 cm (length × width × depth) or a working volume of 40 L, were constructed with clear acrylic plastic. Temperatures of these septic tanks were maintained at ambiW

ent (30–32 C), 40, 50 and 60 C using heated water circulating around the reactors (Figure 1(a) and 1(b)). To minimize heat loss, each septic tank was insulated with aluminum foil. An on-line temperature sensor (Thermocouple type K (CA) [series SK PCR-1)] and a gas collection chamber were installed in each reactor. To start up the operation, each reactor was fed with 40 L of septage sample collected from a household community and the anaerobic bacteria were allowed to acclimatize until biogas generation was observed. After acclimatization, the efﬂuent samples were collected for analysis of pH, chemical oxygen demand (COD), 5-day biochemical oxygen demand (BOD5), volatile suspended solids (VSS), total solids (TS), total Kjeldahl nitrogen (TKN) and E. coli concentrations. Biogas samples were collected and analyzed for the volume and percentage content of CH4.

and to develop a model to describe the pathogen inactivation efﬁciency. The speciﬁc objectives are listed below:

Experimental study on kinetics of E. coli inactivation

1. To evaluate the efﬁciencies of pathogen reduction in lab-

To determine the effects of temperatures on E. coli inac-

oratory-scale septic tanks operating at temperatures

tivation, each of the 50 mL glass bottles was fed with

varying from 30 to 60 C. 2. To determine an appropriate model for pathogen inactivation in septic tanks.

about 10 mL of septic tank sludge and 20 mL of septic tank efﬂuent collected from a household in central Thailand. To simulate conditions occurring in actual septic

3. To observe the effects of temperature variation in septic

tanks which normally have no mixing, only intermittent

tanks on the efﬁciency of COD reduction and biogas

ﬂows, these glass bottles were thoroughly mixed at the

production.

beginning and incubated under anaerobic conditions (by

4. To validate the developed inactivation model with laboratory-scale and actual septic tanks.

purging with N2 gas) in a water-bath and maintained at W

temperatures of 32, 40, 50, 60 and 70 C. After 1, 3, 6,

Figure 1

T. Koottatep et al.

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Experimental set-up: (a) layout diagram; (b) laboratory-scale septic tanks at Asian Institute of Technology, Thailand.

9, 12, 24, 48, 72, 96, 120, 168 and 216 h of incubation,

using the ﬁrst-order kinetic model (Equation (1)) and the

about 10 mL of the supernatant of each glass bottle

Weibull model (Equation (3))

were analyzed for E. coli concentrations and to determine the inactivation efﬁciencies. Analytical methods

Nt =N0 ¼ e k T t

(1)

where Nt is the number of microbial populations at any time, N0 is the number of microbial populations at initial time, t is

All physical and chemical analyses of the inﬂuent and efﬂu-

contact time and kT is ﬁrst-order rate constant at tempera-

ent septage samples were done according to Standard

ture T (Equation (2))

Methods (). The biogas composition was determined by a gas chromatograph (Agilent 7890) equipped with FID

kT ¼ k20 θ(T 20)

(2)

detector, while the E. coli concentration was analyzed by the multiple-tube fermentation technique.

where k20 is the ﬁrst-order rate constant at the standard temperature of 20 C, T is the reactor temperature and θ W

Data analysis

is a coefﬁcient which is 1.06 (1.02–1.10, Metcalf & Eddy ).

The data of the E. coli inactivation at various temperatures

Owing to various factors, the microbial inactivation may

were used to determine an appropriate inactivation model

not always follow the ﬁrst-order kinetic model and a

T. Koottatep et al.

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

modiﬁcation such as the Weibull model has been proposed

04.1

2014

Effects of temperature on septic tank performance

(Prabhakar et al. ). The performance of the laboratory-scale septic tanks at varlog Nt =N0 ¼ bT t

(3)

ious temperatures after the 60 days of operation is shown

where bT is a temperature-dependent coefﬁcient and n is the

were temperature ﬂuctuation within ±5 C in the septic

Weibull coefﬁcient.

tanks depending on hourly ambient temperatures. The

in Table 2. From the online temperature sensors, there W

The E. coli inactivation data obtained from the exper-

supernatant temperatures were about 2 C higher than

imental study described in an earlier section were tested

those in the septic tank sludge. The efﬂuent E. coli concen-

with Equations (1) and (2) to determine an appropriate

trations

inactivation model, which was later validated with the

temperatures; there were about 10 MPN/100 mL at the

laboratory-scale septic tank data obtained as described

temperatures of 50 and 60 C, suitable for discharge into

above and with the data obtained from ﬁeld investigation

a receiving water or for reuse (WHO ). It is apparent

of actual septic tank performance at some household

that the high temperatures caused some damage to the

communities in central Thailand.

E. coli cells resulting in the high inactivation efﬁciency.

were

found

decrease

with

increasing

The percentage removal of COD and BOD5 were in the ranges of 84–94% and 76–94%, respectively, comparable with the performance of conventional septic tanks and anaerobic digesters (Metcalf & Eddy ). The major

RESULTS AND DISCUSSION

mechanisms responsible for COD and BOD5 removal The characteristics of septage samples used in this study

were sedimentation of settleable solids and anaerobic

are shown in Table 1. Although the average age of these

digestion of organic matter in the settled sludge and in

septage samples was about 1 year, resulting in the BOD5/

supernatants. The TS concentrations of the settled sludge

COD ratio of about 0.32, there were still high concen-

in the four laboratory-scale septic tanks were found to be

trations of organic matter with the average BOD5 and

in the range of 49,000–53,630 mg/L. It is apparent that

total COD concentrations of 3,250 and 10,620 mg/L,

these septic tank efﬂuents need to be further treated to

respectively. The average E. coli number was 9.6 × 106

reduce the COD and BOD5 concentrations to meet the

MPN/100 mL, higher than the WHO guidelines for safe

efﬂuent standards for disposal or reuse in Thailand (120

reuse (WHO ).

and 20 mg/L, respectively) (www.pcd.go.th). In terms of biogas production, the temperature of W

40 C gave the highest biogas volume, methane content Table 1

and mL CH4/g BOD5 removed. There was less biogas

Septage characteristics

production at the temperatures of 50 and 60 C, probably Parameter

Average value

7.30 (±0.14)

sent in the septage could not survive at these high

Total COD (mg/L)

10,620 (±967)

temperatures or because of the absence of thermophilic

Soluble COD (mg/L)

1,636 (±291)

methanogenic bacteria in the septage used in this study.

Total BOD5 (mg/L)

3,250 (±354)

This hypothesis was supported by a parallel study of

BOD5/COD

0.32

Syne (), using the polymerase chain reaction tech-

TKN (mg/L)

175 (±78)

nique,

TS (mg/L)

55,173 (±5,819)

methanogenic microorganisms operating at 40 C was

VSS (mg/L)

10,947 (±1,777)

higher than in those operating at 32, 50 and 60 C. The

E. coli (MPN/100 mL)

9.6 × 10 (±3.8)

FISH analysis also showed the high number of cells

Note: standard deviation in parentheses.

because the methanogenic microorganisms originally pre-

which

found

the

DNA

concentration

hybridized with EUB338 and Arc 915 at the temperature

T. Koottatep et al.

Table 2

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Performance of laboratory-scale septic tanks at various temperatures

Supernatant concentrations

Biogas

E. coli a (MPN/

Total COD

% Total COD

BOD5

% BOD5

Biogas volume

% CH4

mL CH4/g BOD5

( C)

100 mL)

(mg/L)

removal

(mg/L)

removal

(mL)

content

removed

32b

4.6 × 103(±2.02)

93.4(±0.21)

181(±10)

94.4(±0.32)

870

Temp W

640(±22)

40.5(±7.4)

2.8

1.6 × 10 (±0.28)

696(±15)

94.0(±0.15)

208(±12)

93.5(±0.38)

1,394

45.9(±14)

6.7

12.3(±0.36)

1,144(±73)

89.2(±0.69)

438(±17)

86.5(±0.54)

781

38.8(±9.9)

2.6

13.4(±0.56)

1,324(±28)

84.8(±0.27)

760(±6)

76.5(±0.18)

809

33.8 (±15.7)

2.7

Note: standard deviation in parentheses. a

Observed on 20th day of operation.

Average ambient temperature.

of 40 C, whereas the number of these cells was much W

lower at 50 and 60 C. The ratios of mL CH4 per g BOD5 removed as shown in Table 2 were less than the theoretical value of 350 mL CH4 per g BOD5 removed (Polprasert ) because the septage contained a low BOD5/COD value of 0.32 (Table 1). Since, in principle, higher temperatures should promote higher bacterial activities and consequently higher biogas production, it is recommended that septic tanks operating at temperaW

tures of 50–60 C should be seeded with thermophilic sludge or more acclimation time should be provided to support

the

growth

thermophilic

methanogenic

bacteria. The above results show the increase of E. coli inactivation efﬁciencies in the laboratory-scale septic tanks

Figure 2

with increasing temperatures from 30 to 60 C, which

Relationship between E. coli inactivation and retention times at various temperatures.

were later used for validation of the Weibull model. It W

should be noted that this study emphasized mainly the

70 C and retention times up to 216 h were plotted accord-

effects of temperatures on E. coli inactivation in septic

ing to the Weibull model (Equation 3) as shown in

tanks and, to maintain the same experimental con-

Figure 2. Probably because of the complex interactions

ditions, no additional seeding of thermophilic bacteria

among the septage solids, E. coli cells and temperatures

was provided for the laboratory-scale septic tanks

in the experimental bottles, the E. coli inactivation data

operating at 50 and 60 C. This could be a reason for

were not found to follow the ﬁrst-order kinetic model

the low COD reduction and biogas production in the

(Equation 1). Owing to the efﬁcient heat distribution in

laboratory-scale septic tanks operating at the high

the 50 mL glass bottles, the temperature of 70 C resulted

temperatures.

in immediate inactivation of the E. coli cells and the inac-

tivation data at this temperature were not used in the E. coli inactivation efﬁciency

Weibull model calculation. The values of bT and n (appliW

cable within the temperature range of 30–60 C) and the The experimental data for E. coli inactivation in the

correlation coefﬁcient (R 2) of the Weibull model are

50 mL glass bottles at temperatures of 32, 40, 50, 60 and

given in Figure 2. From the average n value of 0.33, the

T. Koottatep et al.

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Weibull model of E. coli inactivation can be written as

Equation (4) and a modiﬁed Weibull model is proposed as

shown in Equation 4.

shown in Equation (6)

log Nt =N0 ¼ bT t0:33

(4)

log Nt =N0 ¼ γ bT tn

(6)

A relationship between bT and T derived from Figure 2 is From trial and error, the γ value of 0.5 appeared to ﬁt best

shown in Equation (5) with the R 2 value of 0.97.

the laboratory-scale E. coli inactivation data with SEE values bT ¼ 0:05 T 1:02

(5)

of 0.40–0.82, lower than those of Equation (4) (Weibull

Equation (4) was validated with the E. coli inactivation

appeared to be able to predict the E. coli inactivation efﬁciency

data obtained from the laboratory-scale septic tanks as

in the laboratory-scale septic tanks operating at the tempera-

shown in Figure 3. It is apparent that Equation (4) did not

tures of 32, 40 and 50 C satisfactorily, but not at the

ﬁt well with the laboratory-scale data with standard error

temperature of 60 C, probably owing to several factors such

of the estimate (SEE) of 0.84–1.63. The discrepancy was

as efﬁciency of heat diffusion and response of E. coli cells,

probably due to the inefﬁciency of heat diffusion into the

which may not be linear with increasing temperatures.

E. coli cells and scale-up effects. To compensate for this dis-

Investigations on heat transfer and hydraulic mixing at various

crepancy, a regression factor, γ, was incorporated into

temperatures are being conducted and will be reported later.

model), as shown in Figure 3. The modiﬁed Weibull model

Figure 3

Validation of modiﬁed Weibull model with data from laboratory-scale and actual septic tanks: (a) laboratory-scale and actual septic tanks, 32 C; (b) laboratory-scale septic tank, W

40 C; (c) laboratory-scale septic tank, 50 C; (d) laboratory-scale septic tank, 60 C.

T. Koottatep et al.

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The modiﬁed Weibull model (Equation 6) was further vali-

septic tanks constructed in urban areas usually do not have drai-

dated with ﬁeld data from ﬁve septic tanks belonging to some

nage or leaching ﬁelds, but the septic tank efﬂuent is directly

household communities in Pathumthani province, central

discharged into nearby storm drains or the adjacent environ-

Thailand (ambient temperature about 32 C). It should be

ment.

Since

septic

tank

efﬂuents

still

contain

high

noted that in Thailand and in most other developing countries,

concentrations of fecal microorganisms, this practice poses sig-

only black water (excreta and ﬂush water) is discharged into

niﬁcant health risks to the people who may have contact with

septic tanks, while grey water (wastewaters from washing,

this contaminated water or soil. From the modiﬁed Weibull

bathing and cleansing, etc.) is disposed of separately into

model (Equation 6), by raising the temperature of a convention-

nearby storm sewers or canals. This practice usually makes

al septic tank to 60 C, the E. coli concentration of the septic

the retention time (or contact time) of the septic tank 5–20

tank efﬂuent could be reduced to less than 103 MPN/100 mL,

days, longer than the conventional retention time of 1–3 days.

at a contact time of 5 days, which is suitable for discharge or

As shown in Figure 3, the predicted results of E. coli

for agricultural and aqua-cultural reuse (WHO ). Thermal

inactivation calculated from Equation (6) ﬁt satisfactorily

treatment of septic tanks can be achieved by solar heating or

with the ﬁeld data with a correlation coefﬁcient (R 2)

in combination with energy obtained from the biogas generated

value of 0.68. It is suggested that the modiﬁed Weibull

by the septic tanks themselves and future investigation on this

model (Equation 6) should be validated further with E.

practical aspect is recommended.

coli inactivation data from actual septic tanks operating W

at temperatures above and below 32 C.

In general, excreta can contain various types of pathogenic microorganisms including viruses, protozoa and

Although it is well known that increasing temperatures

helminthic ova, especially from infected persons and/or

will result in better pathogen inactivation, there have been

during disease outbreaks. Because these pathogens can be

no data about the extent of pathogen inactivation in thermal

more resistant to thermal inactivation than E. coli, further

septic tanks which could be used in the design and oper-

investigation of their inactivation efﬁciencies by thermal

ation to protect public health.

treatment is recommended.

It can be seen from the data shown in Table 2 and Figures 2

It is evident from Equation (6) and Figure 3 that without

and 3 that the extent of E. coli inactivation in the laboratory-

thermal treatment septic tanks operating at ambient tempera-

scale and actual septic tanks was less than those found in com-

ture of about 30 C or below can achieve only about 2 log E.

posting piles as reported by Feachem et al. (). For example,

coli inactivation, which could result in the septic tank efﬂuent

although the temperature of 62 C in a composting pile would

being unsafe for disposal or reuse. Owing to the socio-econ-

be able to inactivate most pathogens in about 1 h, according to

omic conditions in developing countries, sewerage or

Equation 6, it would require about 5 days to inactivate more

drainage systems are not the solution to the sanitation pro-

than 6 log E. coli in the septic tanks. The E. coli cells could sur-

blems. Therefore, efforts toward non-sewer sanitation should

vive longer in the septic tanks because they are protected from

be encouraged to improve performance of the existing on-

sun (UV light) and there are more nutrients and moisture than

site sanitation systems. As reported in this study, thermal

in compost piles, which could also contain ﬁrst level and

septic tanks are one of the improved on-site sanitation technol-

second level consumers such as protozoa or beetles that con-

ogies which have high potential in pathogen inactivation and,

sume the E. coli cells (Polprasert ).

with proper seeding and operation, could also produce biogas as an alternative energy source.

Application of results The experimental results obtained from this study suggest that

CONCLUSIONS

the efﬂuent of septic tanks could be made to contain relatively low concentrations of E. coli (pathogenic indicator) by the appli-

The results obtained from this study, which aimed to

cation of thermal treatment. In most developing countries,

improve pathogen inactivation efﬁciency in septic tanks,

owing to land limitation and other socio-economic factors,

can be summarized as follows

T. Koottatep et al.

Thermal septic tanks

Journal of Water, Sanitation and Hygiene for Development

1. The E. coli inactivation efﬁciencies were found to increase with increasing temperatures; there was more than 6 log reduction of E. coli in the laboratory-scale septic tanks W

operating at temperatures of 50 C and above. 2. The E. coli inactivation data were found to follow a modiﬁed Weibull model shown below: log Nt =N0 ¼ γ bT tn bT ¼ 0:05 T 1:02 γ ¼ 0:5 and n ¼ 0:33 The modiﬁed Weibull model could be applied to predict the extent of E. coli inactivation in actual septic tanks W

(temperature about 32 C) satisfactorily, but further validation of this model with E. coli inactivation data of septic tanks operating at temperatures above and below W

30 C is recommended. 3. The percentage COD and BOD5 removal in the laboratoryscale septic tanks operating at temperatures of 30, 40, 50 W

and 60 C were found to be in the range of 84–94 and 76–94, respectively. 4. Biogas production was found to be highest at the temperaW

ture of 40 C, while less biogas production was observed at W

temperatures of 50–60 C, probably owing to low growth of the thermophilic methanogenic bacteria.

ACKNOWLEDGMENT This study was ﬁnancially supported by the Bill & Melinda Gates Foundation, for which grateful acknowledgment is made.

REFERENCES Carrington, E. G.  Evaluation of Sludge Treatments for Pathogen Reduction. Final Report. Study contract, B4-3040, 322179, European Community, Luxembourg.

04.1

2014

Eriksson, E., Auffarth, K., Henze, M. & Ledin, A.  Characteristics of grey wastewater. Urban Water 4 (1), 85–104. Feachem, R. G., Bradley, D. J., Garelick, H. & Mara, D. D.  Sanitation and Disease: Health Aspects of Excreta and Wastewater Management. Wiley, Chichester, UK. Leclerc, H., Mossel, D. A., Trinel, P. A. & Gavini, F.  Microbiological monitoring: a new test for fecal contamination. In: Bacterial Indicators: Health Hazards Associated With Water, ASTM STP 635 (A. W. Hoadley & B. J. Dutka, ed.). American Society for Testing and Material, West Conshohocken, PA, pp. 21–31. Metcalf & Eddy  Wastewater Engineering. 3rd edn., McGraw-Hill, New York. Mocé-Llivina, L., Muniesa, M., Pimenta-Vale, H., Lucena, F. & Jofre, J.  Survival of bacterial indicator species and bacteriophages after thermal treatment of sludge and sewage. Applied and Environmental Microbiology 69 (3), 1452–1456. Pandey, P. K. & Soupir, M. L.  Escherichia coli inactivation kinetics in anaerobic digestion of dairy manure under moderate, mesophilic and thermophilic temperatures. AMB Express 1 (1), 1–10. Polprasert, C.  Organic Waste Recycling: Technology and Management. 3rd edn., IWA Publishing, London. Prabhakar Murthy, D. N., Min, X. & Renyan, J.  Weibull Models. John Wiley & Sons, Hoboken, NJ. Rybczynski, W., Polprasert, C. & McGarry, M.  Low-cost Technology Options for Sanitation. A State-of-the-Art Review and Annotated Bibliography, WB ATWSS 4. World Bank, Washington, DC. Standard Methods for the Examination of Water and Wastewater  21st edn., American Public Health Association/ American Water Works Association/Water Environment Federation, Washington, DC. Syne, N. E.  Determination of Microbial Dynamics and Pathogen Inactivation Efﬁciency from Thermophilic Septic Tanks. Master’s Thesis, Asian Institute of Technology, Pathumthani, Thailand. WHO  Health Guidelines for the Use of Wastewater in Agriculture and Aquaculture. Report of a WHO Scientiﬁc Group, Technical Report Series No. 778, WHO, Geneva. WHO  World Health Statistics 2013. www.who.int/gho/ publications/world_health_statistics/2013/en/ (accessed 15 October 2013). Winblad, U. & Simpson-Hébert, M.  Ecological Sanitation. (revised and enlarged edn)., Stockholm Environment Institute, Stockholm, Retrieved from www. ecosanres.org/ES2_download.htm (accessed 15 October 2013).

First received 11 July 2013; accepted in revised form 6 October 2013. Available online 13 December 2013

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Sustainability of water, sanitation and hygiene interventions in Central America Raquel I. Sabogal, Elizabeth Medlin, Gonzalo Aquino and Richard J. Gelting

ABSTRACT The American Red Cross and U.S. Centers for Disease Control and Prevention collaborated on a sustainability evaluation of post-hurricane water, sanitation and hygiene (WASH) interventions in Central America. In 2006 and 2009, we revisited six study areas in rural El Salvador, Guatemala, Honduras and Nicaragua to assess sustainability of WASH interventions finalized in 2002, after 1998’s Hurricane Mitch. We used surveys to collect data, calculate indicators and identify factors that influence sustainability. Regional sustainability indicator results showed there was a statistically significant decline in access to water. The presence of sanitation facilities had not changed since the beginning of the project; however, maintenance and use of latrines declined but continued to meet the goal of 75% use after 7 years. The hygiene indicator, hand washing, initially declined and then increased. Declines in water access were due to operational problems related to storm events and population changes. Sanitation facilities were still present and sometimes used even though they reached or surpassed their original design life. Changes in hygiene practices appeared related to ongoing hygiene promotion from outside organizations. These results provide useful input for making WASH programs more sustainable

Raquel I. Sabogal (corresponding author) Elizabeth Medlin Richard J. Gelting Centers for Disease Control & Prevention, National Center for Environmental Health, Division of Emergency and Environmental Health Services, 4770 Buford Highway NE, Chamblee, Georgia 30341, US E-mail: rsabogal@cdc.gov Gonzalo Aquino American Red Cross, Water and Sanitation Delegate-Latin America and Caribbean Region, Quality and Learning Unit, Albrook, Calle Jorge Bolívar Alemán Estévez, Ediﬁcio # 453, Panama City, Panama

and informing future, more in-depth research into factors inﬂuencing sustainability. Key words

| Central America, evaluation, hygiene promotion, sanitation, sustainability, water

INTRODUCTION One United Nations Millennium Development Goal (MDG)

ill-informed agencies’. When assessing sustainability, projects

is to halve, by 2015, the proportion of the world’s population

often fail to consider community capacity and needs. Various

that is without access to safe drinking water and basic sani-

estimates have documented somewhere between 30 and 60%

tation. The world reached the MDG drinking water target in

of existing water supply systems do not provide adequate ser-

2010, ﬁve years ahead of schedule. However, reaching the

vice (Brikke & Bredero ; Davis ; Lockwood & Smits

MDG sanitation target by 2015 is unlikely (UNICEF/

). ‘Water-supply and sanitation projects should not be

WHO ). One major obstacle for both water and sani-

viewed as an end in themselves, but as the initiators of beneﬁts

tation interventions is long-term sustainability.

that continue long after the projects have been handed over to

Projects often focus on providing basic infrastructure,

the communities’ (Brikke & Bredero ).

rather than on ongoing functionality. Moe & Rheingans

Organizations around the world have made signiﬁcant

() reported on many examples of ‘failed water and

progress in providing access to improved water and sani-

sanitation projects supported by well-intentioned but

tation worldwide. However, limited evidence is available on sustainability of rural water and sanitation interventions

(Montgomery et al. ). The focus of this evaluation is sustainability over the medium to long term. Our evaluation has reconﬁrmed that sustainable water, sanitation and hygiene

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

(WASH) interventions should address several elements,

practices. In providing interventions, the ARC took a partici-

including:

patory approach and integrated community participation in

• • • • •

project development and implementation at the beginning

Technical appropriateness. Continuing functionality through design life. Social acceptability to the community.

of these projects. A minimum of 80% of the population in each community had to be willing and able to participate

Economic viability.

through labor and willingness to pay a water fee. Hurri-

Protection of the environment and natural resources

cane-affected communities also provided input on the level

(Brikke & Bredero ).

of services they were able to support and selected their inter-

In this evaluation, we do not propose a universal deﬁnition for sustainability but, rather, we assess sustainability

ventions based on the costs, beneﬁts and feasibility of each option (Moll et al. ). Water system designs included projected population growth rates and per capita water needs.

using WASH indicators. The indicators integrate the sustain-

Sanitation design considered local geologic factors such as

ability elements so that if water and sanitation systems

depth to water table and soil type.

continue to function and people continue to practice posi-

In collaboration with ARC, the U.S. Centers for Disease

tive hygiene practices, then communities will meet at least

Control and Prevention (CDC) evaluated the effectiveness

some of the sustainability elements.

and the health impacts of the ARC WASH program. The

We documented the ﬁrst 7 years of a 10-year WASH sus-

evaluation occurred in eight study areas located in four Cen-

tainability evaluation in rural communities in Central

tral American countries. ARC deﬁned the study areas as

America. At the time of data collection in 2009, 19% of

either a single community or two adjacent communities

the rural population in the four Central American countries

that shared similar demographics, geographic region and

in our evaluation had no access to improved drinking water

intervention type. The ARC WASH program’s health objec-

sources; 38% of this population had unimproved sanitation

tive was a 25% reduction in childhood diarrhea from

facilities, with 14% of that population not using any type of sanitation facility (WHO/UNICEF ).

baseline (2000) to ﬁnal (2002). The 2002 ﬁnal evaluation showed that ARC’s WASH reconstruction program met this goal on a regional basis (i.e. across all study areas). Results from the 3-year health impact study have been pre-

BACKGROUND

viously documented (Moll et al. ).

In 1998, Hurricane Mitch struck Central America. It caused

impact study was limited in its ability to address longer-term

major damage to infrastructure especially throughout El Sal-

intervention sustainability. The CDC recommended follow-

vador, Guatemala, Honduras and Nicaragua and was

up evaluations every 3 or 4 years over a decade to assess

recognized as the deadliest Atlantic hurricane since the

the long-term sustainability of the WASH interventions in

Great Hurricane of 1780 (NOAA ). The devastation

these communities. After 2002, the study areas received no

left many without water, sanitation and other services,

technical or ﬁnancial assistance from ARC. This paper docu-

affecting an estimated 3.6 million persons. Some 10,000

ments the ﬁrst 7 years of that 10-year sustainability

died and nearly 100,000 homes were destroyed (USAID

evaluation.

Because of the short 3-year period, however, the health

). The American Red Cross (ARC) responded to the disaster by providing community- and household-level WASH interventions to 110 communities.

METHODS

The ARC’s goal for the program was to ‘decrease health risks associated with water and sanitation to Hurricane

For the sustainability evaluation, we revisited six study areas

Mitch survivors’. The objectives were for communities to

from the health impact study (Table 1). We excluded two of

have sustainable water systems, to have access to potable

the original eight study areas for logistical reasons (e.g. inac-

water and to learn how to improve sanitation and hygiene

cessibility), Waspam, Nicaragua and Huitzitzil, Guatemala.

R. I. Sabogal et al.

Table 1

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Study area, population size and interventions in the sustainability evaluation

Size of community (2009)

Type of community

Water intervention (year completed)

1. La Ceiba

100–105 households

Rural; existing community in hilly region

New water system: spring source, pumped to tank, gravity ﬂow to household taps (2001)

Household composting latrines (2002)

2. Las Pozas

1,004 households

Peri-urban; resettlement community

New water system: deep drilled well, pumped to tank, gravity ﬂow to household taps (2001)

Household composting latrines (2002)

Country/study area

Sanitation intervention (year completed)

El Salvador

Guatemala 3. Chiquimula (2 communities) Plan Shalaguaa

300 households

Rural; existing mountain communities

Upgrade existing water system: spring-fed, gravity ﬂow to shared communal household taps (2002)

Household VIP latrines (2002)

El Guayaboa

180 households

Rural; existing mountain communities

New water system: spring-fed, gravity ﬂow to household taps (2002)

Household VIP latrines (2002)

4. Las Lomas

500 households

Peri-urban; existing community in hilly region

Upgrade water system: new tank and source, additional household connections, springfed, gravity ﬂow system to household taps (2002)

Household pourﬂush latrines (2002)

5. Marcovia

245 households

Peri-urban; resettlement community in ﬂat area

New water system: deep drilled well, pumped to tank, gravity ﬂow to household taps (2002)

Household pourﬂush latrines (2002)

Honduras

Nicaragua 6. Nueva Segovia (2 communities) Dipilto Nuevoa

50 households

Peri-urban; existing community

New water system installed by municipality (not by ARC): spring fed gravity ﬂow system to household taps (2000)

Household dry pit latrines (2002)

Dipilto Viejoa

90 households

Peri-urban; existing community

New water system installed by municipality (not by ARC): spring fed gravity ﬂow system to household taps (2001)

Household dry pit latrines (2002)

Two communities grouped together as a study area for a sufﬁcient number of households to sample.

VIP, ventilated improved pit latrine.

The sustainability evaluation consisted of:

• • •

The U.S. Agency for International Development (USAID)

Food

and

Nutrition

Technical

Assistance

A community survey conducted with one or more

(FANTA) Project ‘Water and Sanitation Measurements

members of each community’s water committee and

Guide’ (Guide) (Billig et al. ) provided the basis for

community leaders.

the original health impact study and was used in the sustain-

A cross-sectional household survey, which included a

ability evaluations for consistency. ARC requested use of the

questionnaire, visual inspection of household water and

eight performance indicators for the health impact study.

sanitation facilities, and visual assessment of hygiene

We continued the sustainability evaluations using four of

behaviors.

eight performance indicators as we could reliably collect

An infrastructure inspection/assessment by CDC and

data for those four. Performance indicators were a consist-

ARC of the community water system and sanitation facili-

ent way to evaluate WASH interventions over time

ties to assess functionality and maintenance.

(Table 2).

R. I. Sabogal et al.

Table 2

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

USAID water and sanitation performance indicators

Intervention

Performance indicator

Description of indicator

Goal

Water infrastructure

Monitoring indicator

Percentage of households with year-round access to improved water source

100%a

Sanitation infrastructure

Monitoring indicator

Percentage of households with access to (presence of) sanitation facility

100%a

Hygiene promotion

Impact indicator

Percentage of households with appropriate handwashing behavior Percentage of population using hygienic sanitation facilities

50% increase

Impact indicator a

75% in use

Goal deﬁned by the American Red Cross; the Guide speciﬁed no goal for this indicator. See FANTA Guide (Billig et al. 1999) for how each indicator was calculated.

We used monitoring indicators to evaluate the progress of the intervention toward achieving its programmatic goal.

square odds ratios to determine differences between the years.

Monitoring indicators assess both water access and access

We collected water quality samples from stored house-

to sanitation facilities. We based the water access indicator

hold drinking water, community water sources and tap

on the FANTA Guide deﬁnition that includes connection

water from the distribution system. Given that the focus of

to a piped system, distance to water and reported year-

this paper is on sustainability in terms of the indicators

round water availability. Access to a sanitation facility

described above, we report only generalizable water

means the presence of a private/shared facility in close

sampling results.

proximity to the home. This indicator does not measure whether families use the facility, but rather the physical presence of the facility. An improved sanitation facility in rural

RESULTS

areas is a dry pit latrine, ventilated improved pit (VIP) latrine, pour-ﬂush latrine or composting latrine.

The ‘Sustainability of WASH interventions’ section contains

We used impact indicators to assess the effect the inter-

results of the sustainability evaluation using the perform-

ventions have on behavior, such as handwashing and use/

ance indicators. The household participation rate was

maintenance of sanitation facilities. Appropriate handwash-

100% in six study areas (eight communities). The ‘Factors

ing knowledge is based on both the interviewee’s self-

inﬂuencing sustainability of WASH interventions’ section

reported ability to recite – unprompted – critical times at

describes the factors we identiﬁed through evaluation of

which handwashing occurs and the interviewee’s ability to

community and household surveys.

demonstrate speciﬁc handwashing techniques. Sanitation facilities were assessed using a standard checklist in the

Results: Sustainability of WASH interventions

household survey to determine use and if they were hygienic. We determined sample size by region rather than by

Table 3 summarizes the results of the performance indicators

community. We calculated the number of households

from 2000 (baseline), 2002 (immediately post-intervention),

needed to conduct statistical analyses of handwashing beha-

2006 and 2009. We used household survey data to estimate

viors. The handwashing behavior indicator required the

indicators by percentages. Although the sample sizes were

largest sample size. The target sample size for the region

different (526 in 2000, 569 in 2002 vs. 94 in 2006 and 104

was 94 households, a range of 14–16 households in each

in 2009), these results represent the region.

of six study areas, with a probability of alpha ¼ 0.05 and 80% power. We collected data using a form developed in

Water infrastructure

EPI INFO 2002 (CDC ) and did additional statistical analyses using SAS software versions 9.1 and 9.2 (SAS Insti-

When ARC completed WASH interventions in all study

tute, Inc. –). We compared survey results using Chi-

areas in 2002, 89% of households had access to improved

Table 3

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Regional results for the sustainability evaluation

Baseline (preintervention)

2002 2006

p-value 2009 (2002 vs.

p-value (2006 vs.

p-value (2002 vs.

Goal

2006)

2009)

100%

< 0.0001 0.60

< 0.0001

Sanitation infrastructure

Households with access to 100% (presence of) sanitation facilitya

0.97

0.53

0.28

Hygiene promotion

Appropriate handwashing behavior

50% increase above 35 baseline Goal 53%b 75% in use 36

0.06

0.02

0.22

0.01

0.90

0.01

Performance indicator Intervention

description

Water infrastructure Households with yearround access to improved water source

Population using hygienic sanitation facilitiesa

Bold indicates that the indicator met the goal. Underline means statistically signiﬁcant difference. a

Both indicators represent sanitation infrastructure.

The calculated goal is 50% increase above baseline of 35%.

water sources. This level of coverage was below the ARC

signiﬁcance. Yet the period 2006–2009 saw a statistically sig-

goal of 100% access. Results in Table 3 show a statistically

niﬁcant increase from 57 to 73% in households with

signiﬁcant decline in year-round access to an improved

appropriate handwashing behavior.

water source from 2002 to 2006, dropping from 89 to 71%. This decline remained approximately the same in

Results: Factors inﬂuencing sustainability of WASH

2009, at 74% coverage.

interventions

Sanitation infrastructure

We found ﬁve factors from community and household survey data that had potential effects on WASH intervention

In 2002, 97% of households had improved sanitation present at or near the home. No statistically signiﬁcant changes in coverage occurred after 2002. Coverage in 2006 and 2009 was 98 and 95%, respectively. However, a statistically signiﬁcant decrease in use and maintenance of latrines did occur from the 2002 baseline level of 87 to 77% in 2006 and then remained at that lower level in 2009. These lower levels of use still met ARC’s original

sustainability: 1. Occurrence of natural disasters or events. 2. Population growth or decline. 3. Presence of active water committees. 4. Lifespan of WASH infrastructure. 5. Follow-up from outside organizations. Occurrence of natural disasters/natural events

goal for percentage of population using hygienic sanitation facilities of 75%.

Interviews with community leaders and water committees in 2006 and 2009 revealed that these study areas struggle with

Hygiene promotion

frequent severe weather events and natural disasters. Flooding occurs annually during the rainy season. Heavy rains

By 2002, appropriate handwashing behavior met the ARC

cause landslides and earthquakes occur at times. Not only

goal of a 50% increase over the 2000 baseline. From 2002

do these events affect community water systems, they can

to 2006, households with appropriate handwashing behav-

also damage household sanitation facilities. Table 4 shows

ior declined from 67 to 57%, although not with statistical

that all study areas reported issues with natural disasters

Table 4

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Community survey reporting of events and natural disasters damaging water and sanitation infrastructure (2006 and 2009)

Event

Occurrence after event

Impacted communities

Hurricanes/Storms

Flooding damaged water system Strong winds damaged latrines Landslide damaged water system

La Ceiba, Chiquimula, Las Lomas, Nueva Segovia La Ceiba, Las Pozas, Chiquimula, Marcovia, Nueva Segovia Chiquimula (Plan Shalagua only)

Earthquakes

Damaged water system and latrines

Las Pozas

Deforestation

Landslides affected water quality

Nueva Segovia (Dipilto Nuevo only)

or events affecting their water and sanitation infrastructure.

Community survey results in 2006 and 2009 found problems

All study areas reported issues in 2006, and ﬁve of the six

with pit latrines (dry pit/VIP) and pour-ﬂush latrines. The

study areas reported storm-related damage in 2009.

latrines would fill and overflow, especially during the rainy season. The interviewers observed and confirmed these con-

Population growth/decline

ditions in the household surveys.

Interviewers obtained population estimates as part of the com-

Follow-up from outside organizations

munity survey. Although populations both grew and declined after the ARC WASH interventions, in four of the six study

After completion of the WASH programs, communities

areas community populations increased. From 2002 to 2009,

received no ARC support. Water and sanitation infrastruc-

signiﬁcant population growth occurred in two communities

ture, hygiene promotion and the health impact study were

– Las Lomas, Honduras and Plan Shalagua, Guatemala –

ﬁnished in 2002. We found that ﬁve of the six study areas

with 131 and 233% growth, respectively. In contrast, Las

reported that they received some type of follow-up edu-

Pozas, El Salvador had a 88% population decrease.

cation from outside organizations after 2002. Our survey results identiﬁed the local Red Cross National Societies,

Active water committees

non-governmental organizations (NGOs) and local ministries of health providing follow-up.

At the time of the 2009 data collection, all six study areas continued to have designated water committees and most

Water quality results

were active, except for Plan Shalagua in Chiquimula, Guatemala. The committee, however, had just reconvened after a

General water quality results showed that microbial water

long period of inactivity to address their water source pro-

quality was better in chlorinated water systems. In addition,

blem. As of 2009, most of the six study areas had set up

water quality degraded through household water manage-

bank accounts for water fee deposits. Chiquimula, Guate-

ment. We will report complete water quality results in a

mala and Nueva Segovia, Nicaragua (only Dipilto Nuevo)

future paper.

did not have any savings in their accounts. Lifespan of WASH infrastructure

DISCUSSION

The percentage of households with access to an improved

Occurrence of severe natural events/natural disasters

sanitation facility remained relatively constant (Table 3) from 2002 to 2009. However, the percentage of those

We found that natural disasters and events in every study

using a hygienic sanitation facility decreased signiﬁcantly

area were responsible for substantial damage to community

in 2006 and remained at that lower level in 2009.

water systems and sanitation facilities and reduced their

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

sustainability (Table 4). The climate in this region in Central

spend available funds annually on water system repairs.

America has a dry and rainy season. Daily rains during the

New homes still received access to the water system,

regular rainy season contributed to rivers overﬂowing their

which put a greater demand on the system causing water ser-

banks and causing local ﬂooding in addition to other natural

vice issues for the entire community. The water system

disasters or events.

expansion did not keep up with community growth. Com-

Damage to water systems in these instances limits access

munity growth surpassed the system’s capacity to provide

to an improved water source. For example, in 2005, rains

24 hours per day water service. The lesson is that there

from Hurricane Stan caused a landslide in Plan Shalagua,

must be a balance between planned projection for growth

Guatemala that greatly reduced the quantity of water pro-

needs and a community’s circumstances.

duced by the spring source. In El Guayabo, Guatemala,

In contrast, Las Pozas, El Salvador, had a decrease in

the conduction pipeline from the source to the storage

population after 2002. Hurricane Mitch destroyed the orig-

tank is constantly prone to damage from falling trees

inal coastal community of Las Pozas residents, and they

during annual rainstorms and strong winds. El Guayabo

relocated inland during ARC’s post-disaster community

constructed this conduction pipeline above ground owing

reconstruction in 2000–2002. Yet despite the improved

to rocky terrain. Water system design in such a region may

water supply and the improved sanitation, the 2006 house-

have to be more robust to avoid damage and annual wash-

hold survey found that lack of economic opportunity

outs of pipelines during the rainy season.

forced many residents to leave this community. Residents

The percentage of households with the presence of an

returned to the original location closer to the coast to seek

improved sanitation facility near the home did not decrease.

jobs in the ﬁshing and shrimping industries. Previous work

However, the percentage of the population using hygienic

has shown that resettlement in a new location after a natural

latrines did show a statistically signiﬁcant decline. Our

disaster can pose a socio-economic challenge to relocated

observations and comments reported during the community

families, particularly with regard to employment and

and household surveys found structurally damaged latrines.

income (Badri et al. ). This signiﬁcant population

Households at times were not able to repair their latrines,

decrease in Las Pozas left many homes empty, latrines

rendering them either unusable, lacking privacy or unhygie-

unused, and because of fewer paying households, an

nic owing to waste seepage from cracked slabs or absorption

increased ﬁnancial strain on the water system. Post-disaster

tanks. Because of these issues, the better indicator of sani-

resettlement or transition can continue for several years

tation sustainability was not access (which did not

until the community has regained its social and economic

change), but use and maintenance, which showed declines.

production systems (Partridge ; Oliver-Smith ).

Sanitation system design in this region should also consider

This situation led to issues regarding sustainability of both

severe natural events and disasters to enhance sustainability.

the water and sanitation interventions.

Population growth/decline

Active water committees

Owing to population growth, water systems in some study

Well-maintained and functioning water systems invariably

areas were unable to keep up with consumer demand

had active water committees that met regularly to resolve pro-

while population decline put water system sustainability

blems and to make repairs. These water committees were

into question in other areas. In Las Lomas, Honduras,

responsible for collecting and depositing water fees into a

ARC planned for projected growth at the initiation of the

bank account, and operating and maintaining the water

project, expecting that families would be attracted to this

system. A well-run water system leads to more paying custo-

community owing to the water service. From 2002 to

mers and to more revenue for maintenance and repair.

2009, there was 131% population increase. Water system

Water committees that were not diligent in collecting water

expansion, however, was limited owing to seasonal wash-

fees had no water bank account and had more difﬁculty

outs of water pipelines that required the community to

making needed repairs. Chiquimula, Guatemala (Plan

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Shalagua) and Nueva Segovia, Nicaragua (Dipilto Nuevo)

follow-up after 2002 and showed a decline in the hygiene

both had active water committees but had no water system

promotion indicators.

bank account. When communities do not repair water sys-

During both the community and household surveys, inter-

tems in a timely manner, disruptions in service, inadequate

viewees emphasized the importance of follow-up assistance

water quantity and poor water quality result, and dissatisﬁed

and education. Water committees reported a need for ongoing

consumers refuse to pay their water fees.

assistance and technical development of members on water system management to improve service to the community or

Lifespan of WASH infrastructure

to repair their water systems after storm or earthquake-related damage. Many household interviewees also reported a need

According to the ARC, depending on the design and type of

for ongoing technical assistance and materials on how to

latrine (dry pit/VIP, pour-ﬂush, composting), design life is

maintain and continue using their speciﬁc type of latrine

subject to considerable variation. ARC based design life on

(e.g. composting latrines) or how to build a new one after

local criteria (e.g. soil type, depth to groundwater and

their latrines reach the end of their design life.

number of family members using the latrine) (WHO ;

Changes in hygiene practices, both positive and nega-

USAID ). ARC-constructed pit latrines had a design

tive, appeared related to the presence or absence of

life of 5 to 10 years, while ARC expected composting latrines

ongoing hygiene promotion from other organizations.

to last up to 20 years. A composting latrine requires constant

Although that follow-up education is not likely at the same

maintenance to function properly and last through its desig-

level of programming as the ARC pre-2002, it can be a posi-

nated design life span. Household survey results showed

tive inﬂuence. Previous work has shown that behavior

that after 4 years, communities that received pit latrines

change requires consistent messaging to communities

(dry pit/VIP) and pour-ﬂush latrines were experiencing pro-

(Arnold et al. ; Luby et al. ) to be effective.

blems. When their latrines were ﬁlled or overﬂowing during the rainy season, household members used a relative’s or neighbor’s sanitation facility or simply defecated outdoors.

LIMITATIONS

The lesson here is that when investing in long-term, sustainable sanitation – and particularly when there is a need to

Several limitations in this sustainability evaluation could

construct new latrines in the future – organizations should

inﬂuence our results. Owing to limited resources, the CDC

consider appropriate latrine design factors as well as com-

designed overall sample sizes for the 2006 and 2009 sustain-

munity education that teaches how to use readily available

ability evaluations to detect expected differences in the

local materials for repairs or construction.

USAID WASH indicators (Table 2) over the entire region rather than at the study area level. Sample size calculations

Follow-up education and assistance

did not account for clustering and the design effect could be large. Thus, sample sizes only allowed for statistical ana-

A lack of follow-up hygiene promotion may help explain the

lyses at a regional level, across all study areas combined,

decrease

whereas only trends in the USAID WASH indicators are

appropriate

handwashing

behavior

and

decreased use of hygienic latrines from 2002 to 2006. How-

observable at the study area level.

ever, the provision of follow-up from 2006 to 2009 may

Additionally, limitations due to the random selection of

explain the regional improvement in handwashing behavior

households within the study areas might have changed some

and lack of change in use of hygienic sanitation facilities

of the evaluation results. New households might never have

over that time. Several study areas (Las Pozas, El Salvador;

received certain ARC WASH interventions (e.g. water

Las Lomas and Marcovia, Honduras; and Nueva Segovia,

system access, sanitation facilities, hygiene promotion)

Nicaragua) received hygiene promotion from organizations

before or after 2002. When conducting the surveys for the

other than ARC after 2002. Chiquimula, Guatemala (Plan

2006 and 2009 sustainability evaluations, these households

Shalagua) was the only study area that did not receive any

were included in the random selection. These households

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

may not have known who provided WASH interventions to

The primary factors that inﬂuenced the sustainability of

their communities. If these households did not receive cer-

water systems were severe storm events, population changes

tain interventions and were unaware of who provided

and active water committees. Our results suggest that

them, it would be difﬁcult to assess accurately the sustain-

improving water system design by accounting for such

ability of these interventions.

weather events could promote sustainability. If available,

Self-reporting of data is also a limitation in this evalu-

using localized population growth estimates rather than

ation. CDC included self-reported knowledge and observed

standard rules of thumb or national growth estimates,

practice of appropriate handwashing behavior in each

could also help promote sustainability. Active water commit-

household according to the FANTA Guide. Ram ()

tees also enhanced water system sustainability, indicating

recognized that self-reporting of handwashing knowledge

that developing or identifying appropriate community-level

is not a valid measure of actual behavior; however, we col-

capacity to manage water systems is an important consider-

lected data in this manner to be consistent from survey to

ation (Gelting & Ortolano ).

survey. This indicator introduces bias since observed and

We can best evaluate sanitation interventions by looking

evaluated respondents may have modiﬁed their normal

at indicators for both access and use. ARC initially provided

handwashing techniques. To lessen the impact of this limit-

latrines to households, which require proper maintenance

ation, interviewers received thorough training in the proper

and use. Simply looking at access did not provide an adequate

way to conduct this survey.

measure of sustainability. In some areas, latrines were still

Finally, we did not initially use these indicators to

accessible and in use but were damaged or past their usable

measure sustainability. CDC used the same indicators from

life, and could not therefore be considered sustainable.

the health impact study and added questions to the survey

Thus, organizations beginning a sanitation intervention

to identify the factors associated with long-term sustainabil-

should consider what would happen at the end of the sani-

ity. Two indicators, for example, measure sanitation access

tation

and use. Our results show that although access to sanitation

replacement or pit emptying could be included at project

appears sustainable (monitoring indicator), use of sanitation

initiation. Replacement could result from new interventions

facilities did decline (impact indicator) as we have noted in

in the future or by training community members in techniques

our results. We must consider indicator results collectively

for constructing new latrines when needed, or making repairs

to get a better picture of the sustainability of each interven-

to existing latrines. Most helpful to these communities will be

tion. Despite such limitations, our evaluation was a unique

developing training materials that emphasize the use of

opportunity to follow the same study areas over a number

locally obtained materials for repairs or construction.

of years.

facilities’

design

life.

Planning

for

eventual

Rural communities, with limited resources and expertise, can beneﬁt most from follow-up by local organizations once the initial WASH program is complete. The follow-up hygiene promotion that occurred in some of these commu-

CONCLUSIONS

nities, appeared in every instance to improve WASH intervention sustainability (appropriate handwashing behav-

The goal of this evaluation was to observe and measure the

ior). Hygiene promotion programs involve teaching materials

sustainability of infrastructure and hygiene interventions

and/or technical assistance as opposed to capital investment

once communities began the operation and maintenance

in infrastructure. Organizations can accomplish ongoing

of these interventions, with no additional ARC follow-up.

follow-up hygiene promotion programs with relatively little

We wanted to identify possible factors that inﬂuenced

expense using local resources available from municipalities,

WASH intervention sustainability. Community participation

local health clinics or NGOs.

and input did not vary widely across the region and

Our sustainability evaluation results can potentially

would not have a differential impact on sustainability in

guide future program start-up toward ensuring and enhan-

this study.

cing the long-term sustainability of these types of projects

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

and lead to programmatic change in international development organizations. The ARC has incorporated our results to make changes in their current WASH programming to reﬂect this. In addition, our results provide the background research for future studies to explore sustainability further, such as investigating which WASH interventions are most sustainable and identifying additional factors contributing to sustainability. If interventions are not sustainable, health and other beneﬁts derived from them are lost. Thus for all water and sanitation interventions, future investigators need to address sustainability issues.

DISCLAIMER The ﬁndings and conclusions in this journal article are those of the author(s) and do not necessarily represent the views of the Centers for Disease Control and Prevention.

ACKNOWLEDGMENTS The authors would like to thank the American Red Cross, Red

Cross

National

Societies

water

and

sanitation

delegates, and all community members who participated in our study. Their support ensured our success with this study. American Red Cross (Sergio Denegri, Mark Toy) and Red Cross National Societies (El Salvador-Mirna Zelaya, Salvador Molina, Adan Rivas; Guatemala-Maria Teresa Estrada, Adolfo Saguil; Honduras-Joel Duron, Jorge Pinel; Nicaragua-Carmen Maria Garcia Bucardo, David Maxwell, Ramona Caceres). We would also like to thank the additional CDC investigators who participated in past data collections (Lana Corrales MPH, Victoria Cuellar MPH, Kristin Delea MPH, Jane Horton MPH, Carolyn Monteilh PhD).

REFERENCES Arnold, B., Arana, B., Mausezahl, D., Hubbard, A. & Colford, J. M.  Evaluation of a pre-existing, 3-year household water treatment and handwashing intervention in rural Guatemala. Int. J. Epidemiol. 38 (6), 1651–61. Epub 2009 Jul 2, 1–11.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Badri, S. A., Asgary, A., Eftekhari, A. R. & Levy, J.  Postdisaster resettlement, development and change: a case study of the 1990 Manjil earthquake in Iran. Disasters 30 (4), 452–468. Billig, P., Bendahmane, D. & Swindale, A.  Title 2 Indicator Guidelines: Water and Sanitation Indicators Measurement Guide. US Agency for International Development (USAID) Food and Nutrition Technical Assistance (FANTA) Project, Washington, DC. Available from: http://watsanmission assistant.wikispaces.com/ﬁle/view/WatSan+Indicators +Measurement+Guide+(USAID).pdf. Brikke, F. & Bredero, M.  Linking Technology Choice with Operation and Maintenance in the Context of Community Water Supply and Sanitation. A Reference Document for Planners and Project Staff. World Health Organization and IRC Water and Sanitation Centre, Geneva, Switzerland. Centers for Disease Control and Prevention (CDC)  EpiInfo 2002. Revision 2, January 30. CDC Division of Public Health Surveillance and Informatics. Davis, S.  Improve International. About Us. Available at: http://improveinternational.wordpress.com/about-us/ Gelting, R. & Ortolano, L.  A model describing performance of rural drinking water systems in Honduras. Water Resources Development 14 (2), 199–215. Lockwood, H. & Smits, S.  Supporting Rural Water Supply: Moving Towards a Service Delivery Approach. Practical Action Publishing, Rugby, UK. Luby, S. P., Agboatwalla, M., Bowen, A., Kenah, E., Sharker, Y. & Hockstra, R.  Difﬁculties in maintaining improved handwashing behavior. Karachi, Pakistan. Am. J. Trop. Med. Hyg. 81 (1), 140–145. Moe, C. & Rheingans, R. D.  Global challenges in water, sanitation, and health. Journal of Water and Health 4(Supp 1), 41–57. Moll, D. M., McElroy, R. H., Sabogal, R. I., Corrales, L. F. & Gelting, R. G.  Health impact of water and sanitation infrastructure reconstruction programmes in eight Central American communities affected by Hurricane Mitch. Journal of Water and Health 5 (1), 51–65. Montgomery, M. A., Bartram, J. & Elimelech, M.  Increasing functional sustainability of water and sanitation supplies in rural sub-Saharan Africa. Environmental Engineering Science 26 (5), 1017–1023. National Oceanic and Atmospheric Administration (NOAA)  Mitch: The Deadliest Atlantic Hurricane Since 1780. Available at: http://www.ncdc.noaa.gov/oa/reports/mitch/ mitch.html (accessed May 25 2010). Oliver-Smith, A.  Successes and failures in post-disaster resettlement. Disasters 15 (1), 12–23. Partridge, W.  Involuntary resettlement in development projects. Journal of Refugee Studies 2 (3), 373–384. Ram, P.  Practical Guidance for Measuring Handwashing Behavior. Global Scaling Up Handwashing Project. Water and Sanitation Program: Working Paper.

R. I. Sabogal et al.

Sustainability of water, sanitation, hygiene interventions, Central America

SAS Institute, Inc – The SAS System for Windows. Release 9.1 and 9.2. SAS Institute, Inc., Cary, North Carolina. UNICEF and World Health Organization (UNICEF/WHO)  Progress on Drinking Water and Sanitation. 2012 Update. WHO/UNICEF Joint monitoring Program for Water Supply and Sanitation. Available at: http://www.unicef.org/media/ ﬁles/JMPreport2012.pdf. US Agency for International Development (USAID)  Community Sanitation Improvement and Latrine Construction Program Training Guide. Technical Report No. 83, March. Available at: http://pdf.usaid.gov/pdf_docs/ PNABN929.pdf.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

USAID  Bureau for Humanitarian Response/Ofﬁce of U.S. Foreign Disaster Assistance Annual Report, FY 1999. Washington, DC. Available at: http://pdf.usaid.gov/pdf_ docs/pdabs450.pdf. WHO  A Guide to the Development of on-site Sanitation. Available at: http://helid.digicollection.org/en/d/Jh0210e/3. 2.1.html. WHO/UNICEF  WHO/UNICEF Joint Monitoring Programme (JMP) for Water Supply and Sanitation. 2003–2013. Data resources and estimates for El Salvador, Guatemala, Honduras and Nicaragua 2009. Available at: http://www. wssinfo.org/data-estimates.

First received 7 February 2013; accepted in revised form 8 August 2013. Available online 21 October 2013

100

04.1

2014

Participatory multi-criteria evaluation of alternative options for water supply in cyclone-prone areas of Bangladesh Abu Hena Mustafa Kamal Sikder and Mashﬁqus Salehin

ABSTRACT Availability of safe drinking water is considered a key challenge in the coastal region of Bangladesh. High concentrations of salinity, iron and arsenic, and the unavailability of suitable aquifers, have deterred the exploitation of groundwater resources. In addition the cyclonic storm surge is a major threat to this system. Cyclones accompanied by storm surges in the coastal area cause signiﬁcant deterioration of drinking water supply and sanitation. Water professionals have launched some initiatives to promote small-scale, alternative safe water sources (e.g. rainwater harvesting, pond sand ﬁlters and piped water techniques) to provide sustainable solutions to the problem. However, a systematic evaluation of the alternatives that considers social, technical and economic criteria has

Abu Hena Mustafa Kamal Sikder (corresponding author) Ka-86/C, Khapara, Khilkhet, Dhaka, Bangladesh E-mail: sikder_cee@yahoo.com Mashﬁqus Salehin Institute of Water and Flood Management, BUET, Dhaka-1000, Bangladesh

not been carried out so far. The present study is an attempt to evaluate the alternative options for drinking water supply in a cyclone-prone area. The authors conducted a multi-criteria analysis and reached the conclusion that rainwater harvesting is the most suitable option for the area. Moreover, the ﬁnal result was shared with the users to obtain their feedback to ensure sustainability of the water source. Key words

| coastal area, cyclone, drinking water supply, participatory multi-criteria analysis, water supply option

INTRODUCTION The integrated water resource management (IWRM)

Rahman ). As a result, lack of stakeholder participation

approach is gaining traction in the ﬁeld of water resource

is one of the causes of the failure of many schemes to func-

planning and management. The aim of this interdisciplinary

tion as expected (Chadwick & Datta ).

approach is to consider technical, social, economic and

Public sector planning and management problems are

environmental aspects of a given problem (Zarghaami

always challenging and complex. Decision makers have to

). Multi-criteria decision analysis (MCDA) is an effec-

deal with a number of alternatives and evaluate them

tive tool for decision-making using the IWRM approach.

based on some incompatible criteria. To ensure social

This ensures the active participation of stakeholders and

equity, technical feasibility and economic efﬁciency, the

experts from different disciplines in the decision-making

conventional cost–beneﬁt analysis method is not very effec-

process. To make a water resource project successful and

tive as it applies the valuation technique and converts all the

sustainable it is important to involve users, government

impact into monetary value (Munda ). Stakeholders

agencies, nongovernmental organizations and other devel-

need to be involved as early as possible in the analysis and

opment stakeholders in the decision-making process. In

to focus that involvement on reﬁning objectives and criteria,

Bangladesh, however, stakeholder participation during plan-

rather than on adjusting a proposed solution (Bruen ).

ning is not a very common phenomenon (Chowdhury &

The MCDA approach provides a ﬂexible way to deal with

doi: 10.2166/washdev.2014.094

101

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

both qualitative and quantitative multidimensional effects of

Then users were asked to suggest sustainable solutions to

a given problem. Using this technique, practitioners look at

equity issues that were identiﬁed during the questionnaire

complex problems characterized by any mixture of monet-

survey. Finally, some recommendations were made that

ary and non-monetary objectives, break the problem into

would ensure sustainability of the intervention.

more manageable pieces, and apply data and judgments to the analysis of each component. After this exercise, the

Study area

pieces are reassembled to present a whole coherent picture to decision makers (DCLG ).

The study was conducted in Barguna, one of the coastal dis-

The coastal zone of Bangladesh is a place of multiple

tricts of Bangladesh. The villages of Gajimahamud and

vulnerabilities and opportunities. Records of the last 200

Nishan Baria in Barguna Sadar subdistrict and the villages

years show that about 70% of the cyclones that originated

of Padma and Shingra Bunia in Patharghata subdistrict

in the Bay of Bengal have made landfall in Bangladesh

were selected because of their vulnerability to cyclone and

(Islam & Ahamed ). Rural water supply in Bangladesh

storm surge. All four villages are located in the exposed

is mainly dependent on groundwater. But establishing a

coastal zone deﬁned by the national Coastal Development

safe and reliable source of drinking water is one of the

Strategy (Water Resources Planning Organization )

major challenges in the coastal areas because of salinity in

and along the bank of the tidal river Bishkhali. The data

groundwater (PSU ). The situation deteriorates further

on demographics and total land area of the study area

when a cyclone strikes. To establish a reliable and consistent

were collected from the respective Union Council Ofﬁces

source of water, government and other organizations have

(local government at Union level). The total population is

installed and constructed different water supply options

12,718 (2,574 households) and total land area is 6.8 km2.

(Paudyal ). This study is an attempt to apply MCDA

Among the villages, Nishan Baria has the largest population

in evaluating the existing water supply options in the study

with a total of 857 households and Padma has the smallest,

area, ensure stakeholder participation and develop a rank-

511 households. Average total annual rainfall is 2,659 mm,

ing of the alternatives.

and 90% of this rainfall occurs during monsoon season, which is from June to October (Uddin & Rahman ). In the coastal area brackish groundwater is available

MATERIALS AND METHODS

within 0 to 2.5 m below the ground surface. In some regions low saline groundwater is available in deep aquifers at a

Multi-criteria analysis is a well-known tool to use when sta-

depth greater than 200 m (Ahmed ). The area is only a

keholder participation and conﬂict management are among

few feet above sea level and the entire southern part is

the objectives of the decision-making process. Some

exposed to the Bay of Bengal (Uddin & Rahman ). Geo-

examples of applying multi-criteria analysis successfully in

graphical positioning has made the district highly vulnerable

environmental and water resource management issues are

to cyclone and storm surge (Islam & Peterson ).

found in Raju et al. (), Hostmann et al. () and Herath (). A participatory multi-criteria analysis

Origin and extent of the problem

method was applied to evaluate the alternative water supply options of the area. The study also assessed technical

In the Coastal Development Strategy (Water Resources

and social feasibility of the different options. Questionnaire

Planning Organization ) of the Bangladesh Govern-

surveys and participatory rural appraisal tools were applied

ment, lack of safe drinking water was identiﬁed as a key

to identify the existing alternatives and also to derive the

concern. Water supply in the area is mainly dependent on

evaluating criteria from the stakeholders. Multi-criteria

hand pump tubewells. Pond water is also used especially

analysis was conducted using an impact matrix, in which

where groundwater is either saline or unaffordable. Data

different options were ranked following both quantitative

from a recent multiple indicator cluster survey showed

and qualitative analysis of options against the criteria.

that 85.7% of the population has access to deep tubewells

102

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

(BBS and UNICEF ). But in recent years, the ground-

(incorporating partial lists of criteria and asking the respon-

water-based water supply in coastal areas has suffered

dent to accept or reject and add new criteria if necessary). A

from a number of major problems, primarily arsenic con-

total of 100 households were surveyed from the four villages.

tamination, lowering of the water table, salinity and

Twenty-ﬁve households from each village were included in

nonavailability of a suitable aquifer (Ahmed & Rahman

the survey, and at least four households representing each

). The problem becomes more critical when a cyclone

type of water option were surveyed. After the researchers

or storm surge strikes the area and damages the water

identiﬁed users of the different water supply options, the

supply sector critically. All the ponds of the inundated

respondents for the survey were selected randomly. The cri-

area become contaminated by saline and turbid surge

teria

water. Surge water also deteriorates the quality of the tube-

implementation cost, operation and maintenance cost, econ-

well water. Saline water goes through the suction pipe and

omic impact of the facility over the users, water availability,

contaminates the water in the aquifer. In Barguna a total

water quality, social acceptance, and cyclone resistance. The

of 1,343 tubewells were damaged by the devastating Cyclone

criteria were then grouped into three categories: economic,

Sidr in 2007 (MoFDM ). Without proper treatment the

technical and social. The study took place with the approval

water cannot be used for drinking purposes. After a cyclone

of the Committee for Advanced Studies and Research of

strikes, the population of the coastal area suffers a massive

Bangladesh University of Engineering and Technology.

deﬁciency of safe drinking water and this results in the outbreak of waterborne diseases (Rahman & Bux ).

listed

the

questionnaire

survey

were:

In the economic group three criteria (implementation cost, operation and maintenance cost, and economic impact) were included. The implementation cost includes

Selection of alternatives and criteria

the expenses associated with the installation of the water supply option. Operation cost is the annual expenditure to

To meet the drinking water demand, various water supply

maintain a functional facility, including spare parts and

technologies were introduced in the area. Formerly, people

labor. Economic impact also seeks to explore whether the

drank rainwater during the monsoon, and they stored the

option has any ﬁnancial implication, either positive or nega-

water in large-sized earthen pitchers during this season

tive, on the users.

(Hussain & Ziauddin ). During the dry season the

The technical group comprises water availability, water

main source of drinking water was ponds, which were

quality and cyclone-resistance criteria. Water availability

specially conserved for drinking purposes (Ahmed ).

deals with the issue of whether the water source produces

Then shallow and deep tubewells were introduced. The

water year-round and if not for how many months. Water

tubewell reduced the prevailing problem of water supply

quality criterion investigates the odor, color and taste of

and quality to some extent but created some new problems

the water sources. And cyclone resistance is the ability of

as referred to in the section above. Options such as pond

the water source to withstand cyclonic storm surge and pro-

sand ﬁlters (PSF) and rainwater harvesting systems

duce safe water.

(RWHS) were also introduced in the area to minimize

Lastly, the only criterion in the social group (social

dependency on groundwater. For this study all ﬁve water

acceptance) indicates the users’ perceptions about a given

supply options in the community were considered. They

water supply technology – if there is any presumptive

are: hand pump connected with deep (DTW) and shallow

notion among users about any of the technologies. The cri-

(STW) tubewells, PSF, RWHS, and natural ponds.

teria and their groups are shown in Table 1.

The criteria are the measures of performance by which the alternatives will be judged. A measurement or judgment needs to specify how well each alternative meets the objec-

MULTI-CRITERIA ANALYSIS

tives expressed by a criterion (DCLG ). The criteria for the analysis were selected ensuring the active participation

A weighted sum method was used in this study to perform

the multi-criteria analysis. The alternatives were calculated

the

stakeholders

through

questionnaire

surveys

103

Table 1

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

impacts of all the options (no economic loss due to the

Evaluation criteria according to group

water option) except the users of PSF. This difference can

Economic

Technical

Social

Implementation cost

Water availability

Social acceptance

Operation and maintenance cost

Water quality

Economic impact

Cyclone resistance

be explained by the prohibition of commercial ﬁsh cultivation in ponds using PSF, which limits owners’ income generation options. In the technical group, the water availability criterion was determined by using both survey data and technical analysis results. To deﬁne the criterion of water quality both user perception and data from the relevant literature

based on the score value and respective weight assigned to them using Equation (1) V(A) ¼

wi vi ðai Þ

was used (Ahmed ; Islam et al. ; Kawahara et al. ; Rahman & Jahra ). Water from the PSF and RWHS scored well as both the options are capable of sup-

(1)

where, wi is the weight of the criterion i, vi (ai) is the score of

plying safe water if designed and maintained properly (Ahmed ; Ferdausi & Bolkland ; Islam et al. ; TWDB ; Harun & Kabir ) and the users were also satisﬁed with the quality. Because the water from DTW is

one alternative with respect to criterion i, and V(A) is the

bacteriologically safe, this water option scored higher than

resultant value of the alternative A. Both qualitative and

the STW, which are at risk of bacterial contamination

quantitative data were used as the scores so standardization

from the nearby pit latrines, but lower than PSF and

was necessary to convert all scores into the same unit.

RWHS because of salinity. Survey data were used to deter-

The standardization was performed by assigning a score

mine cyclone resistance. Respondents were asked about

of zero to the worst strategy and 1 to the best by using

the experience of their water source after the last cyclone

Equation (2)

(Cyclone Sidr). The technical feasibility of the water alterna-

STDk,j ¼ ACTk,j WORSTk,j = BSTk,j WORSTk,j

tives were also compared while scoring. RWHS scored the (2)

highest because the storage chamber used with this option is made of strong plastic, ferrocement or brick masonry,

where STDk,j is the standardized score value of the kth cri-

which keeps water safe even during tropical cyclones

terion and jth alternative. ACTk,j is the actual value,

accompanied by signiﬁcant rainfall (Shahed & Sikder

WORSTk,j and BSTk,j are the worst and best value of the

). DTW and STW are vulnerable as the surge water inun-

kth criterion. Finally the ranking of the alternatives was

dates the base of the pump. The main source of water in PSF

developed by using the total score (V(A)) obtained by the

is the pond, so the pond and PSF are equally vulnerable to

analysis.

cyclone damage; however, PSF user committees take preventative measures by building embankments around their

Impact matrix

ponds so they are more protected than natural ponds. Considering this fact the PSF scored higher than natural ponds.

The impact matrix for the analysis was formed by using both

For the social acceptance criterion, DTW, PSF and

quantitative and qualitative (linguistic) variables. Under the

RWHS received good scores from respondents. A few inter-

economic group, the implementation cost was calculated by

viewees were not satisﬁed with STW because of high salinity

the estimation of material costs for each option. Mainten-

and unavailability of water year-round. And users of pond

ance costs were obtained from the questionnaire survey

water said they do not want to drink it because it is not

data. An average value for each of the options was used as

safe and requires further treatment.

the cost of maintenance was not constant for all the intervie-

Qualitative data in the impact matrix are presented

wees. Qualitative data were collected and used to determine

using ﬁve different linguistic variables: very good (VG),

economic impact. Respondents reported positive economic

good (G), moderate (M), bad (B) and very bad (VB).

104

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

A scale of 1 to 5 was used to deﬁne the scores from VB to

through the following steps: (i) the numerator was calcu-

VG. The monetary values for the ﬁrst two criteria were

lated by subtracting the worst (maximum) value from the

scored in Tk (currency of Bangladesh). Table 2 shows the

actual value, i.e. 70,000–95,000; (ii) the denominator was

score of alternatives with respect to the criteria.

determined by subtracting the worst value form the best

During the survey respondents were also asked to prior-

(minimum), i.e. 25,000–95,000; and (iii) the score was

itize the evaluating criteria. The economic and technical

obtained by dividing the nominator by the denominator

criteria were considered higher priority than the social. So

and multiplying the result with the weight for the criterion,

economic and technical were weighted at 1.5 (except the

i.e. ( 25,000/ 70,000) × 1 ¼ 0.36.

implementation cost) and the social criterion was weighted at 1. The implementation cost is not fully borne by the user.

Equity issues

Either the government or another organization recompenses the major portion; a small percentage of the total cost is

Users were asked to identify sustainable solutions to equity

taken from the user to develop ownership (LGD ). So

issues that were identiﬁed during the survey. Discussions

the weight of the implementation cost was determined to

focused on ﬁve main concerns: (i) distribution of mainten-

be 1. Table 3 illustrates the standardized scores of the

ance cost among the users; (ii) ensuring options for

alternatives and the weights. For instance, the score of

women and adolescent girls that take into account easy

implementation cost of DTW (Table 3) was determined

access, harassment and dignity; (iii) obstruction by the

Table 2

Scores of alternatives with respect to criteria

Alternatives Groups

Criteria a

DTW

STW

PSF

RWHS

Pond

Economic

Implementation cost (Tk) Operation cost (Tk)a Economic impact

70,000 3,875 G

30,000 1,880 G

95,000 2,478 M

55,000 3,329 G

25,000 1,655 G

Technical

Water availability Water quality Cyclone resistance

G M M

M B M

G G B

M G G

M VB VB

Social

Social acceptance

G: good (score 4); M: moderate (score 3); B: bad (score 2); VB: very bad (score 1); DTW: deep tubewell; STW: shallow tubewell; PSF: pond sand ﬁlter; RWHS: rainwater harvesting system. a

77.64 Tk ¼ 1 US$.

Table 3

Standardized scores of the alternatives

Alternatives Groups

Criteria

Weights

DTW

STW

PSF

RWHS

Pond

Economic

Implementation cost Operation cost Economic impact

1.0 1.5 1.5

0.36 0.00 1.50

0.93 1.35 1.50

0.00 0.94 0.00

0.57 0.37 1.50

1.00 1.50 1.50

Technical

Water availability Water quality Cyclone resistance

1.5 1.5 1.5

1.50 1.00 1.00

0.00 0.50 1.00

1.50 1.50 0.50

0.00 1.50 1.50

0.00 0.00 0.00

Social

Social acceptance

1.0

1.00

0.00

1.00

0.00

6.36

5.28

5.44

6.44

4.00

Total

105

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

land owner during water collection; (iv) discrimination of

memorandum between the users and the land owner

rich and poor in decision-making; and (v) the inﬂuence of

should be put into place for communal options. Priority

the rich in site selection of the facility. Some of the users felt that maintenance costs should only be borne by the afﬂuent to ensure the smooth functioning of the facility. Most of the users felt that women, as the primary handlers of water,

•

should be given to ensure maximum coverage. The users emphasized the need to establish ownership of the options to ensure sustainability, which requires much more social mobilization and community participation.

must be ensured free access to options. Some respondents complained that land owners obstructed access to water. On the other hand, households that had their own facility worried that if they allow everyone to collect water, their

MAJOR FINDINGS AND RECOMMENDATIONS

facility will become damaged through misuse and not last long. Criticism also surfaced about discrimination of rich

The scores of the alternatives based on the impact matrix are

and poor during site selection of a new facility and about

presented in Table 3. The scores for DTW, STW, PSF, RWHS

the formation of caretaker committees, which are respon-

and natural ponds are 6.36, 5.28, 5.44, 6.44 and 4.00, respect-

sible for function and maintenance of water sources.

ively, when calculating the weighted sum. It is clear that

Seemingly, if only the local elite are involved in the care-

RWHS secured the highest score at 6.44. The ranking from

taker committee, it may not function as expected and the

highest to lowest based on the scores of alternatives is:

users may encounter difﬁculties in compelling the commit-

RWHS, DTW, PSF, STW and natural ponds. During the

tee to take the necessary steps to ﬁx the problem.

FGD, the ranking was discussed with the community. Partici-

A focus group discussion (FGD) with 24 users of different

pants also agreed that RWHS is the best option as a source of

options was conducted in the community to clarify the equity

drinking water supply for the study area. As the groundwater

issues. The users were asked to evaluate the alternatives with

is saline and PSF requires sophisticated maintenance such as

respect to the above-mentioned equity criteria. The equity

changing the ﬁlter materials and replacing them after a cer-

evaluation reﬂected the qualitative assessment of alternatives

tain period, RWHS was viewed as the best choice. The

by the stakeholders. Also the ranking of the alternatives based

equity issues were evaluated by the qualitative assessment

on the impact matrix was shared with the users.

of the stakeholders. Some suggestions also came up during

The major suggestions of the FGD were:

the user discussion regarding the sustainability and smooth

•

• •

functioning of facilities. A major portion of the participants Individual options such as household RWHS function

complained of discrimination between rich and poor during

better and require less maintenance than communal

the site selection process and formation of the caretaker com-

options because no one takes the responsibility to ﬁx

mittees. They also suggested considering the gender issue

communal facilities if something is broken or stolen.

during site selection and design of facilities. For instances,

Though some newly constructed facilities have caretaker

women are not comfortable to fetch water from facilities

committees in place, most committees are inactive.

located in public places with a high male assemblage. Simi-

It was strongly recommended that during site selections

larly, the staircases of raised platforms and the height of the

the interests of women should take priority as they are

collection tap should be designed considering easy access

prime users of water options. Sites should be selected in

for women and children. The necessity of developing owner-

places where women feel comfortable, with no threat of

ship

physical harassment, and their privacy is ensured.

maintenance of the facilities also came up during the discus-

Training needs to be enhanced to maintain new technol-

sion. Participants also made strong recommendations

ogies such as PSF and RWHS, which are more

regarding the installation of individual facilities.

and

the

importance

proper

operation

and

Considering the criteria to evaluate the existing options,

sophisticated and require more technical expertise. During site selection and formation of caretaker commit-

RWHS were found to be the most suitable alternative in the

tees,

area. It is important to consider the value judgments of the

all

users

should

considered

equally.

106

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

users during the selection of a water supply option. This will

Drinking water supply in the coastal area of Bangladesh

ensure the sustainability of the facility. During the design of

is a major challenge. Drinking water is the most critical

a facility, user suggestions should be given high priority. Dis-

issue especially in the rural areas closer to the sea where

crimination among the rich and poor should be eliminated

groundwater is saline. The government is trying to promote

from the process. Instead of selecting afﬂuent households,

small-scale alternative water supply sources. But the sustain-

sites should be selected based on speciﬁc criteria such as

ability of these facilities depends on the users to a great

number of users covered, distance from the target house-

extent. So the determination of the user’s perception

holds and not inside someone’s private residence. This will

about a facility is essential. If an option is selected based

ensure easy access of all the users as well as longevity of

only on the technical feasibility, it might not be sustained

the facility. During site selection the gender perspective

if the users are reluctant to receive the technology. To

should be given signiﬁcant weight and water options

ensure the predicted performance of any water supply facil-

should be installed where women and children have easy

ity, all the criteria related to the water supply should be

and safe access. By providing necessary training and devel-

considered properly. This study was an attempt to perform

oping awareness, the regular maintenance of a facility can

a participatory assessment of water supply facilities in the

be ensured. Independent options for each household could

cyclone-prone coastal area and develop a ranking of the

address many of these issues as the users will no longer

options. RWHS emerged as the most preferred option for

have to collect water from a distance and others will not

the study area.

be able to restrict access. However, this will require huge ﬁnancial support. Participation of both beneﬁciary and implementer (government or other stakeholders) in the

REFERENCES

initial cost can reduce this burden. Discussion with the stakeholders should be open, and the decision-maker should address all the issues raised. Feasibility analysis of technical and economic criteria along with social perspectives will ensure the sustainability of a water supply facility. Since rainwater harvesting was chosen as the most preferred option, special attention should be given to facility design. The storage tank should be designed to preserve water for the dry season. The operation and maintenance of the facility should be simpliﬁed to ensure smooth functioning and longevity. Thus use of low-cost, long-lasting and locally available material that would ensure easy water collection, tank cleaning and ﬂow diversion may popularize the system to the users. Additionally, as participants mentioned, the operation cost of RWHS is relatively high compared with other options; while designing the facility this should be taken into account and components that are more durable and do not require annual repair or replacement should be used. RWHS received the highest score in the cyclone resistance criterion. While all the other water sources are contaminated by saline surge water, RWHS can serve as a source of safe water due to the accompanying rain of the cyclone. This unique feature of the technology makes it a suitable option for water supply in the cyclone-prone areas of Bangladesh.

Ahmed, M. F.  Coastal water supply in Bangladesh: reaching the unreached: challenges for the 21st century. In 22nd WEDC Conference, New Delhi, India. Ahmed, M. F.  Rainwater harvesting potentials in Bangladesh, Integrated development for water supply and sanitation. In 25th WEDC Conference, Addis Ababa, Ethiopia. Ahmed, M. F.  Alternative water supply options for arsenic affected areas of Bangladesh. In Workshop on Arsenic Mitigation in Bangladesh, 14–16 January 2002, Dhaka. http://www.physics.harvard.edu/~wilson/arsenic/ conferences/Feroze_Ahmed/, accessed 2 December 2013. Ahmed, M. F. & Rahman, M.  Water Supply and Sanitation: Rural and Low Income Urban Communities. ITNBangladesh, BUET, Dhaka, Bangladesh. BBS and UNICEF  Bangladesh Multiple Indicator Cluster Survey 2009. Technical Report Volume 1. Bangladesh Bureau of Statistics and UNICEF, Dhaka, Bangladesh. Bruen, M.  Multiple criteria and decision support systems in water resources planning and river basin management. Paper presented at the National Hydrology Seminar 2002, Dublin, Ireland. http://www.opw.ie/hydrology/data/speeches/ National%20Hydrology%20Seminar%202002%20No% 204%20-%20MULTIPLE%20CRITERIA%20AND% 20DECISION%20SUPPORT%20SYSTEMS.pdf, accessed 2 December 2013. Chadwick, M. & Datta, A.  Water Resource Management in Bangladesh: A Policy Review. Livelihood–Policy Relationships in South Asia. Working Paper 1. Department of International Development (DFID), London. Retrieved

107

A. H. M. K. Sikder & M. Salehin

Evaluation of water supply options in Bangladesh

from http://citeseerx.ist.psu.edu/viewdoc/download? doi=10.1.1.62.3923&rep=rep1&type=pdf, accessed 6 December 2013. Chowdhury, R. K. & Rahman, R.  Multicriteria decision analysis in water resources management: the Malnichara Channel improvement. International Journal of Environmental Science Technology 5 (2), 195–204. DCLG  Multi-Criteria Analysis: A Manual. Department for Communities and Local Government, London. http://www. communities.gov.uk/documents/corporate/pdf/1132618. pdf, accessed 2 December 2013. Ferdausi, S. A. & Bolkland, M. W.  Design improvement of pond sand ﬁlter. Water sanitation and hygiene: challenges of the millennium. 26th WEDC Conference, Dhaka, Bangladesh. Harun, M. A. Y. A. & Kabir, G. M. M.  Evaluating pond sand ﬁlter as sustainable drinking water supplier in the Southwest coastal region of Bangladesh. Applied Water Science 3 (1), 161–166. Herath, G.  Incorporating community objectives in improved wetland management: the use of the analytic hierarchy process. Journal of Environmental Management 70 (3), 263–273. Hostmann, M., Borsuk, M., Reichert, P. & Truffer, B.  Stakeholder values in decision support for river rehabilitation. Archiv für Hydrobiologie, Supplementband Large Rivers 15 (1–4), 491–505. Hussain, M. D. & Ziauddin, A. T. M.  Rainwater harvesting and storage technique from Bangladesh. Waterlines 10 (3), 10–12. Islam, M. R. & Ahamed, M.  Living in the Coast: Problems, Opportunities and Challenges. Integrated Coastal Zone Management Plan. Water Resource Planning Organization, Ministry of Water Resources, Government of People’s Republic of Bangladesh. http://www.warpo.gov.bd/rep/liv/ living2.pdf, accessed 31 January 2014. Islam, T. & Peterson, R. E.  Climatology of landfalling tropical cyclones in Bangladesh 1877–2003. Natural Hazards 48 (1), 115–135. Islam, M. S., Siddika, A., Khan, M. N. H., Goldar, M. M., Sadique, M. A., Kabir, A. N. M. H., Huq, A. & Colwell, R. R.  Microbiological analysis of tube-well water in a rural area of Bangladesh. Applied and Environmental Microbiology 67 (7), 3328–3330. Islam, M. M., Kabir, M. R. & Chou, F. N.-F.  Feasibility study of rainwater harvesting techniques in Bangladesh. In Rainwater and Urban Design 2007, Engineers Australia, Barton, ACT. Kawahara, K., Nakamura, J., Rahman, M. M., Uddin, M. S. & Latif, M. A.  Pond Sand Filter, JICA/AAN Arsenic Mitigation Project, Report 3. Dhaka, Bangladesh: Asia Arsenic Network. http://www.aan-bangladesh.org/index. php?option=com_content&view=article&id=77% 3Apublications&catid=36%3Apublications&Itemid=74. LGD  National Cost Sharing Strategy for Water Supply and Sanitation in Bangladesh. Local Government Division, Ministry of Local Government, Rural Development &

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Cooperatives, Government of the People’s Republic of Bangladesh. MoFDM  Cyclone Sidr in Bangladesh: Damage, Loss, and Needs Assessment for Disaster Recovery and Reconstruction. Report prepared by the Government of Bangladesh. http:// gfdrr.org/docs/AssessmentReport_Cyclone%20Sidr_ Bangladesh_2008.pdf, accessed 6 December 2013. Munda, G.  Multi-Criteria Evaluation in Fuzzy Environment: Theory and Application in Ecological Economics. PhysicaVerlag, Heidelberg, Germany. Paudyal, G. N.  HYSAWA fund: a unique approach to sustainable and decentralised water supply and sanitation services in rural Bangladesh. In 34th WEDC International Conference, Addis Ababa, Ethiopia. PSU  National Strategy for Water and Sanitation: Hard to Reach Areas of Bangladesh. Policy Support Unit, Ministry of Local Government, Rural Development and Cooperatives. Rahman, M. M. & Bux, M. K.  Post disaster situation of water supply and sanitation. Journal of the Civil Engineering Division, Institute of Engineers Bangladesh CE23 (01), 91–102. Rahman, M. M. & Jahra, F.  Challenges for Implementation of Rain Water Harvesting Project in Arsenic Affected Areas of Bangladesh. http://www.wepa-db.net/pdf/0612sympo/ paper/Md.MaﬁzurRahman.pdf, accessed 2 December 2013. Raju, K. S., Duckstein, L. & Arondel, C.  Multicriterion analysis for sustainable water resources planning: a case study in Spain. Water Resource Management 14 (6), 435–456. Shahed, I. M. & Sikder, A. H. M. K.  Evaluation of water supply facility in the cyclone shelters and the feasibility of rainwater as an alternative source. Session paper B3, Regional Conference on Disaster Risk Reduction and Emergency Response in a Rapidly Changing World, Dhaka, Bangladesh. TWDB  Effect of Roof Material on Water Quality for Rainwater Harvesting Systems – Additional Physical, Chemical, and Microbiological Data. Report, Texas Water Development Board, Austin, Texas. Uddin, A. H. M. & Rahman, A. B. M.  District Information: Barguna. Integrated Coastal Zone Management Plan. Water Resource Planning Organization, Ministry of Water Resources, Government of People’s Republic of Bangladesh. http://www.warpo.gov.bd/pdf/barguna.pdf, accessed 31 January 2014. Water Resources Planning Organization  Coastal Development Strategy. Ministry of Water Resources, Government of People’s Republic of Bangladesh. http://www.warpo.gov.bd/pdf/ coastalDevPolicy.pdf, accessed 6 December 2013. Zarghaami, M.  Integrated water resources management in Polrud irrigation system. Water Resources Management 20 (2), 215–225.

First received 1 August 2013; accepted in revised form 21 November 2013. Available online 27 January 2014

108

04.1

2014

Health in perspective: framing motivational factors for personal sanitation in urban slums in Nairobi, Kenya, using anchored best–worst scaling Carl Johan Lagerkvist, Suvi Kokko and Nancy Karanja

ABSTRACT Severe health, safety and environmental hazards are being created by the growing population of urban poor in low-income countries due to lack of access to sanitation and to inadequate existing sanitation systems. We developed a multi-faceted motivational framework to examine the constituents that explain user motivation regarding a personalised sanitation system. In 2012 we interviewed slum dwellers in Nairobi, Kenya, to estimate individual motivational factor importance rankings from anchored best–worst scaling (ABWS) using hierarchical Bayesian methods. We found that personal safety, avoidance of discomfort with shared toilets, cleanliness and convenience for children were ranked of highest importance. Motivational factors related to health were only relatively highly ranked. Thus factors contributing to overall individual wellbeing, beyond health

Carl Johan Lagerkvist (corresponding author) Suvi Kokko Department of Economics, Swedish University of Agricultural Sciences, PO Box 7013, 75007 Uppsala, Sweden E-mail: carl-johan.lagerkvist@slu.se Nancy Karanja Department of Land Resource Management and Agricultural Technology, University of Nairobi, PO Box 30197, Nairobi 00100, Kenya

benefits, drive adoption and use of the low-cost personal sanitation solution studied. This suggests that non-health benefits of low-cost sanitation solutions should be better acknowledged and communicated to raise awareness and encourage adoption of improved sanitation in urban slums. These findings may help develop policies to promote personal sanitation, improve public health and safety and reduce environmental risks. Key words

| anchored best–worst scaling, factor importance, motivation, personal sanitation, slums

ABBREVIATIONS ABWS anchored best–worst scaling

provided in an ad hoc manner, a problem often ignored by

BWS

best–worst scaling

municipal authorities. For instance in our study area,

coefﬁcient of variation

Kibera slum in Nairobi, Kenya, between 50 and 90% of

hierarchical Bayesian (model)

households do not have access to adequate sanitation and

KSh

Kenyan shilling

child mortality rates are among the worst in the world,

independent-samples median

with 1 in 5 children not surviving beyond 5 years of age

M-W

Mann–Whitney

(Schouten & Mathenge ). Slum sanitation is characterised by shared or communal facilities as a result of high population density, lack of space,

INTRODUCTION

high poverty levels and non-feasibility of constructing conventional sewage systems ( Jenkins & Scott ; Katukiza

The demand for sanitation infrastructure is a growing chal-

et al. ; Isunju et al. ). However, investment in commu-

lenge, especially in urban slums of low-income countries.

nal sanitation facilities has proven to be an unsustainable

Owing to the informal nature of slum settlements, sanitation

solution, not meeting the needs of urban slum dwellers

infrastructure is typically not planned and services are

(Joshi et al. ). Studies in urban slums in Kenya (Schouten

doi: 10.2166/washdev.2013.069

109

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

& Mathenge ), Uganda (Katukiza et al. ; Isunju et al.

collection methods are also needed to help inform public

) and Mozambique (Carolini ) indicate that sanitation

policy makers about individuals’ preferences for sanitation.

facilities have higher use frequencies than they are designed for and suffer from lack of resources for emptying and main-

Background

tenance. Physical constraints due to hilly topography and dense housing add to the burden of emptying (Isunju et al.

For the purposes of this study, we selected the Peepoo toilet

), making such facilities health and environmental

bag to illustrate a low-cost, new technology solution for per-

hazards (Dellström Rosenqvist ).

sonal sanitation. Peepoo is similar to the ‘ﬂying toilet’ in its

Local authorities and governments generally consider

functional characteristics. The ‘ﬂying toilet’ is a polythene

public health an important driver for investments in sani-

bag used for defecation, especially in slum areas, and dis-

tation (Isunju et al. ). However, the unsustainable

posed of in the approximate environment (drainage,

features of current communal systems in low-income

roadsides, rooftops, etc.), often causing an environmental

countries suggest that there could be a demand for solutions

and health hazard when in contact with drinking water

that better match the way users value, for example, accessi-

supply and humans. However, Peepoo is biodegradable

bility, cleanliness, comfort and personal safety of sanitation

and self-sanitising because it contains urea, which inacti-

facilities, and the underlying social and environmental fac-

vates and breaks down harmful human faecal pathogens

tors. Despite many attempts to address the water,

into ammonia and carbonates within 2–4 weeks, allowing

sanitation and hygiene challenges in low-income countries,

faeces to be safely used as fertiliser. It was ﬁrst introduced

little attention has been paid to individual motivation for

in the Kibera slum in Nairobi, Kenya, in 2010 and currently

improved sanitation in relation to the mix of sanitation chal-

an estimated 4,500 individuals and 30 schools use the bag in

lenges faced by slum dwellers, not least access to a personal,

the area (Peepoople ). The bags are sold by self-

instead of a communal, sanitation solution.

employed women (sales ladies) and by street kiosks, for 3

This study investigated the concept of individual motiv-

Kenyan shillings (KSh). Used bags can be dropped off at a

ation in the adoption and use of a personal sanitation

drop point or are collected by women who have started a

solution. Based mainly on previous ﬁndings on sanitation

used Peepoo bag collection service. At the drop points the

adoption and use in sub-Saharan Africa (Jenkins & Curtis

used bags are refunded at KSh1 each. The used bags are

; Jenkins & Scott ; Schouten & Mathenge ;

then further developed into marketable fertiliser. Sales

Isunju et al. ; Whaley & Webster ), we developed a fra-

events and supply to schools are accompanied by training

mework of hypothetical factors motivating choice to use the

in personal hygiene, such as hand washing with clean

sanitation solution studied. The main aim of the study was

water and soap and correct use and disposal of the bag.

to extend the stream of research on household demand for improved sanitation in low-income countries to include moti-

Motivational domains for personal sanitation

vational factors for adoption of a low-cost solution for personal sanitation in urban slum settings, where ﬁnancial

Motivation is an inner drive that reﬂects goal-directed arou-

and geographical constraints do not allow for construction

sal (Jenkins & Curtis ) with the result being a drive to

of home toilets. The overall aim was to provide an understand-

satisfy needs and wants, in this case through the adoption

ing of driving factors for sanitation adoption and use, so that

and use of a personal sanitation service. When examining

policies to increase demand for sanitation can be further

preferences and decision-making that explain household

developed and directed to meet the needs of the urban slum

demand for sanitation, determination of individual attitudi-

population in the above-mentioned settings. The examination

nal and structural determinants instead of socio-economic

of user-orientated priority settings for a multi-faceted concept

characteristics is recommended (Jenkins & Scott ). Con-

such as personal sanitation required reliable and valid

sideration should also be given to how biological drivers

methods to obtain individual-based concept measurements.

such as disease avoidance are shaping hygienic behaviour

Straightforward,

towards sanitation in general and defecation in particular.

accurate

and

easy-to-administer

data

110

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The disgust emotion has been found to function as an adap-

Curtis ). In this study we took the presence of ﬂies

tation for disease avoidance (Curtis & Biran ).

and odour as indicators of uncleanness.

Prototypical objects of disgust are waste products of the human body such as faeces (Phillips et al. ) and dirty

Convenience and comfort

latrines (Curtis & Biran ). Furthermore, the need for comfort and cleanliness, perception of safety and disgust

Convenience and comfort are mostly considered endow-

and mental and physical avoidance of excreta are guided

ment factors within the literature (i.e. being most valued

by social ideas and perceptions. These perceptions have

by those who have already considered or actually chosen

been shown to be difﬁcult to change (Dellström Rosenqvist

to invest in sanitation; e.g. Jenkins & Scott ). The

), making adoption of new sanitation solutions challen-

most common motivation for constructing a toilet in rural

ging as well as contextual. Furthermore, physiological

and peri-urban Ghana is to provide a toilet for sick or old

needs concerning sanitation are in general considered sec-

relatives (Jenkins & Scott ). We assumed that this

ondary to status-related needs. Solutions to overcome

type of convenience consideration for others may extend

established ideas and perceptions of sanitation should there-

to children, since women household members are likely to

fore be introduced together with a solution to status, comfort,

assist children with using a toilet. A personal, in-house sani-

safety and physical needs (Dellström Rosenqvist ).

tation solution could thus save time for other activities. In

Hence a wider approach beyond physical health,

addition, convenience typically relates to aspects of ease of

encompassing physical, mental and social wellbeing, is

physical use and accessibility. Previous studies have shown

necessary for analysing the motivation for adoption of sani-

that technology itself plays a role in sanitation adoption.

tation (Jenkins & Curtis ; Isunju et al. ). We

The toilet design and its management have been found to

identiﬁed six motivational categories which together con-

be important factors for users of communal sanitation facili-

tribute to how sanitation can enhance quality of life within

ties in Kibera (Schouten & Mathenge ). For individual

the framework studied. These categories are hygiene and

sanitation systems like the Peepoo bag, the size and design

cleanliness; comfort and convenience; value for money;

should thus be adapted to how people are expected to use it.

safety; dignity and self-esteem; and health.

Distance to communal toilets and restricted hours of operation were the most determining factors of latrine

Hygiene and cleanliness

usage in urban poverty environments in Bhopal, India

Regardless of their social status or geographical location,

among non-adopters of a household toilet in rural and

(Biran et al. ) and were associated with dissatisfaction ‘people try to use as clean toilets as possible’ (Dangert

peri-urban Ghana (Jenkins & Scott ). These factors

, 5). The urban poor are especially aware of cleanliness,

were drivers for latrine adoption in rural Benin ( Jenkins &

since stigmatisation as a dirty person is associated with pov-

Curtis ). Accessibility is particularly relevant in Kibera,

erty ( Joshi et al. ). Dangert () reported a positive

where public toilets are open only during daytime hours.

association between hygienic perception of toilets and the

Ability to avoid public toilets and easy disposal of Peepoo

absence of visible faeces. Generally speaking, sanitation

enhance convenience for users.

adoption is associated with cleanliness and a better home environment. Cleanliness of the sanitation facility proved

Value for money

to be the most important factor for users in a study assessing communal sanitation facilities in Kibera (Schouten & Math-

Value for money is conceptualised to mean the good or bad

enge ), whereas in rural and peri-urban Ghana bad

feeling arising from the perceived quality of service or utility

odour and dirtiness were related to dissatisfaction with the

provided by the sanitation technology relative to the price

facilities (Jenkins & Scott ). Similar ﬁndings have

paid for it. Therefore the price paid for a sanitation service

been reported from the Philippines (Cairncross ) and

may relate to technology uptake. In general, user fees for

from latrine adopters in rural areas in Benin (Jenkins &

communal facilities burden the poor (WSP ) and the

111

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

high costs of constructing private toilets hinder toilet adop-

a factor of dissatisfaction to users in rural and peri-urban

tion ( Jenkins & Scott ). Affordability of communal

Ghana (Jenkins & Scott ), whereas in the Kibera slum

facilities is an important factor in urban slum settings,

people would prefer to have the sanitation facility in their

such as Kibera (Schouten & Mathenge ). Furthermore,

house (Schouten & Mathenge ), hence reﬂecting a

reduced healthcare costs due to improved sanitation may

desire for privacy. Privacy includes ‘visual screening, social

raise value. Regarding Peepoo, there are two types of

comfort and restricting access to personal information’ (Jen-

value-for-money aspects to consider. Firstly, users can be

kins & Curtis ); seeing other people’s faecal matter

expected to attribute a value to the bag because the cost is

causes embarrassment and reduces the feeling of privacy.

partly refunded upon return and may attribute additional

Sounds and odours may also harm social relations (Dell-

value for the amount they are compensated. Secondly, the

ström Rosenqvist ). In a study of 10 communities in

opportunity to utilise used bags as fertiliser to grow veg-

Zimbabwe, Whaley & Webster () found that adoption

etables provides the potential to save money otherwise

of sanitation is driven by a need for conformance with pre-

spent on purchasing vegetables, or offers otherwise unaf-

vailing social norms and gaining of social acceptance.

fordable vegetables to the household at a low cost. Health Safety Exposure to odour and appearance of excreta are conSafety was ranked of high importance in previous studies,

sidered indicators of health hazards and poor social status

for instance in both rural and peri-urban Ghana (Jenkins

(Dangert ; Jenkins & Curtis ; Jenkins & Scott

& Scott ) and in the Kibera slum (Schouten & Math-

; Aiello et al. ; Whaley & Webster ). Despite

enge ). This is most likely due to people’s preference

the human instinct to avoid faecal matter, avoidance of

for defecating in a safe location (Dellström Rosenqvist

faecal-oral transmission of diseases has been a less signiﬁ-

), for reasons of physical protection from accidents,

cant driver for adoption of sanitation in both rural and

injuries and violence and associated emotional fears.

urban settings (Cairncross ; Jenkins & Curtis ;

Restricted access to toilets for women, who often rely on

Schouten & Mathenge ). Nevertheless, the public sec-

the early morning and night hours to defecate (Cairncross

tor’s demand for sanitation is usually associated with

) and are subjected to risks of sexual violence in the

health in terms of absence of disease (Jenkins & Sugden

vicinity of communal toilets (Amnesty International ),

; Aiello et al. ; Whaley & Webster ).

places this group in a vulnerable position.

A broader approach to health, such as quality of life and wellbeing, may be a more suitable driver for demand for

Dignity and self-esteem

sanitation (Isunju et al. ). It may therefore be relevant to consider the extent to which users of a personal toilet

Toilets tend to convey social dignity and therefore empower

recognise a wellbeing beneﬁt and a link between disease pre-

people (Biren et al. ). A multi-country study in peri-urban

vention and quality of life indicators such as school

areas in East Africa and Sweden showed that men are

attendance.

driven by prestige and status to adopt latrines (Dangert ), whereas women in rural and peri-urban Ghana prefer wellbeing factors such as convenience, comfort and

METHODS

privacy ( Jenkins & Curtis ). Cairncross () found that privacy and avoided embarrassment toward visitors

Methods to investigate and compare the relevance of

were highly ranked in terms of desire for a latrine in various

motivational factors

urban communities in low-income countries. Such positive qualities may invoke feelings of pride, which can further

One of the primary objectives of this study was to examine

accentuate technology adoption. Having to share a toilet is

slum residents’ reasons for choosing to use the Peepoo bag

112

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

sanitation service. Therefore information on each individ-

the method excludes information useful for comparing indi-

ual’s assessment of the importance of the motivational

vidual preferences or for dividing respondents into groups

factors was needed.

with distinctive preferences. However, compared with

Best–worst scaling (BWS) is a theory-driven preference scaling procedure (Marley & Louviere ) commonly

rating or ranking techniques, use of BWS may reduce social desirability, as well as extreme response biases.

used in health treatment (Flynn et al. ) and social care

Anchored best–worst scaling (ABWS) overcomes the

(Potoglou et al. ). Based on an assumption that all moti-

relativism of BWS by introducing a dual-response format to

vational factors included can be ordered transitively (i.e.

augment the standard BWS method (Lagerkvist et al. ).

from most to least important), BWS is an extension of

In Figure 1, the dual format is introduced by asking the inter-

paired comparison, but provides more information and

viewees, for each choice set respectively, an additional

requires less input, since the choice sets contain a larger

choice question to consider the extent of importance for all

set of factors. The method allows for further discrimination

ﬁve factors jointly using three alternatives. Across all

among factors and is devised to invoke trade-offs in the

choice sets, this question establishes each individual respon-

choices between the relevant factors competing for the con-

dent’s threshold between ‘important’ and ‘unimportant’

cerns of the interviewees. A BWS study is based on an

ratings. By assigning the value ‘zero’ to each respondent’s

experimental design which generates choice sets containing

threshold point, a common scale origin or anchor is created

factors under consideration. Figure 1 shows an example of a

that allows different respondents’ importance ratings to be

BWS choice task used in this study. For each choice set,

compared and linked.

interviewees are asked to indicate which factor they like most and which they like least. The analysis of BWS data

Study site and sample description

generates individual estimates of the probability that a given factor is being chosen as most important relative to

The study site, Silanga village in the Kibera slum in Nairobi,

a single reference factor. However, the relativism of BWS

has a population of approximately 17,363 (Kenya National

means that the degree of factor importance is only compar-

Bureau of Statistics ). Existing sanitation options

able within a respondent, but not across respondents

include: open defecation, ﬂying toilets, biogas latrines, pit

(Lagerkvist et al. ). Thus a limitation of BWS is that

latrines, ventilation-improved pit latrines, pour ﬂush toilets

Figure 1

Example of ABWS scaling choice set.

113

Table 1

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Sample description

Category

Characteristic

Age (years) Gender (share) Marital status (share)

Employment (share)

Standard deviation

Min

Max

10.5

Female

0.75

Male

0.25

Married

0.61

Single

0.37

Widow

0.02

No. of household members No. of children in household

Mean n ¼ 122

4.7

2.54

< 2 years old

0.5

0.61

2–5 years

0.7

0.91

5–15 years

1.2

1.32

Unemployed

0.33

Work less than once per week

0.12

Work 2–3 days/w

0.21

Work 4–5 days/w

0.09

Work 6–7 days/w

0.19

No information

0.06

No. of Peepoo bags used per day by household

5.6

3.25

Usage ratio (no. of bags used divided by household size)

1.27

0.74

Usage ratio < 1 (share)

0.471

Usage ratio > 1 (share)

0.521

4.5

and WCs with sewer connection. Lack of space, poor soil

unemployed. The respondents used on average one (1.27)

quality and layout of the area make it difﬁcult to construct

Peepoo per household member per day.

more communal sanitation facilities or to transport construction material (Schouten & Mathenge ). In 2012, a convenience sample of 122 individuals (men

Empirical application

n ¼ 31; women n ¼ 91) was recruited from among Peepoo users (see Table 1). Recruitment was conducted by the

We started the study with a ‘voice of the consumer’ perspec-

Peepoo sales ladies who informed their clientele about the

tive, to examine the face validity of the measurement model.

study. The main criterion for selection was prior experience

A list of 27 factors, representing the six motivational areas of

of Peepoo. The aim was to obtain a gender-balanced sample

the framework (Table 2) and drawn from existing research

of adult users covering a wide range of ages. Literacy was

and communication with local scientists and sanitation

not necessary, since each choice set was read out loud to

experts, was subjected to focus group discussions with

the participants and two facilitators helped with translation,

Peepoo users (n ¼ 7). The focus group discussions provided

clariﬁed questions and helped illiterate respondents ﬁll in

a preliminary insight into perceived factor importance. Face

the socio-economic part of the questionnaire. Oral informed

validity of the list of factors was considered good, i.e. repre-

consent was obtained from participants when introducing

sentative of the experiences users associated with the

the survey. A participation fee of KSh200 was paid. Respon-

sanitation solution. No changes were made to the list of

dents lived predominantly in households of ﬁve members

27 factors and respondents reported the content and presen-

sharing one room and one-third of the respondents were

tation of the factors to be comprehensible.

114

Table 2

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Motivational factors, rank and factor importance of Peepoo

Best–worst scaling rank (n ¼ 122)

Motivational category

Motivational factor

1. Hygiene and cleanliness

10) Peepoo makes going to the toilet cleaner 23) The surroundings of my house and my house itself are cleaner when using Peepoo 24) The use of Peepoo reduces the number of ﬂies in my house/ garden 25) The use of Peepoo reduces smells in my house/garden

Best–worst factor importance weight (%)

3 5

4.81 4.44

3.66

3.70

1) It is easy for me to access Peepoo bags 5) It is easy for me to dispose of the bag at a certain point of collection 6) I save time by not needing to walk to public toilets or queue for public toilets 9) Peepoo is easy to use 11) Peepoo allows me to go to the toilet whenever I need to 12) Peepoo is convenient to use when outside my home 13) It is easy for me to dispose of the bag to a collector 14) The size of the bag is suitable for my use 17) Peepoo makes it easier for old/sick people to go to the toilet 18) Peepoo makes it easier for children to go to the toilet 19) Peepoo is convenient to use at home 22) Using Peepoo allows me to avoid the discomfort of common toilets or ‘ﬂying toilets’

12 18

3.87 3.45

4.11

10 6 27 20 23 8 4 9 2

4.27 4.37 2.10 3.28 3.05 4.35 4.53 4.33 4.88

3. Value for money

2) The price paid for the bag offers ‘value for money’ 3) I receive compensation for a used Peepoo bag 4) The price I receive currently for the used bag is sufﬁcient for me 15) My household’s healthcare costs decrease when using Peepoo 16) I can use Peepoo as a fertiliser when I grow vegetables

19 25 24 15 16

3.34 2.34 2.59 3.64 3.49

4. Dignity and selfesteem

7) I have more privacy when using Peepoo for going to the toilet 8) I have a more modern life when using Peepoo 20) I feel proud of using Peepoo

21 22 17

3.25 3.16 3.47

5. Safety

27) Using Peepoo allows me to avoid dangers at night and at public toilets

4.89

6. Health

21) My health and that of my household members is better when using Peepoo 26) Using Peepoo allows my children to attend school more frequently

4.36

2. Comfort and convenience

2.19

Average percentage explained (APC)

47.0

Standard deviation of percentage explained

8.25

Chance-ratio

2.82

An ABWS survey was prepared to measure the relative

understand the technique for choosing most important and

importance of the 27 motivational factors. Choice sets in

least important within a set of choices. Five daily items

the ABWS experiment were speciﬁed using the MaxDiff

were placed on the table for the respondents to choose

designer v.2.0.2. (Sawtooth Software). The design meant

the object that was most and least important to them. In

that 17 choice sets of ﬁve factors were generated. Each

groups of 8–13 participants, respondents individually

factor was presented on average 3.2 times and each pair

ﬁlled in the ABWS questionnaire, presented in Swahili.

of factors appeared together on average 0.5 times (std

Respondents had the possibility to ask for clariﬁcation

0.5). Before administering the ABWS survey, a short exer-

and translation to English. Sessions lasted on average

cise was undertaken to allow the respondents (n ¼ 122) to

45 min.

115

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Data analysis

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

without covariates for segmentation of sub-samples of respondents. In this study, a chance model has a predictive

Hierarchical Bayesian (HB) models have recently been

power of 16.7% (one out of six choice options).

shown to outperform aggregate (multinomial logit) and latent class methods in estimations of best–worst choice data (Lagerkvist et al. ). Estimation of the HB models

RESULTS AND DISCUSSION

was performed using CBCHB v.5.0.4 (www.sawtoothsoftware.com), resulting in 80,000 iterations before results were

Anchored best–worst scaling

used and an additional 20,000 iterations to calculate, for each interviewee, the share of preferences for each factor,

Table 2 presents the estimated factor importances, which

respectively, i.e. the predicted probability that each factor is

are based on 5,331 observations (43.7 observations per

picked as the most important. For ease of interpretation,

respondent). The anchored dual-response alternative ‘all

the importance weight of factor k within the full set of factors

ﬁve of these are important’ was selected 1,225 times, leaving

in Table 2 was obtained by dividing the share of preference

only 13 observations for the ‘none of these are important’

for factor k by the sum of share of preferences over all factors

alternative. This pervasive result indicates the difﬁculty

for each interviewee, after which results were averaged over

respondents had in discriminating between the factors.

interviewees. The factor importance measure in Table 2

The estimated ABWS model had an average percentage

was then expressed on a common ratio scale (with minimum

explained of 47.0, suggesting that the ABWS model had a

0 and maximum 1), meaning that if one factor has an impor-

predictive accuracy that was 2.82 times higher than that of

tance value twice that of another factor, the former factor is

a pure chance model. The estimated standard deviation of

twice as important as the latter. This procedure followed

8.4 (coefﬁcient of variation (CV) ¼ 0.17) indicates relatively

Lagerkvist et al. (), from which ranks were obtained.

low levels of heterogeneity in the sample in terms of impor-

To predict the effectiveness of a system for personal

tance placed on motivational factors.

sanitation, it is important to understand whether, and to what extent, different groups of users differ in their motiva-

Priority to safety

tional structures (i.e. factor importance). To this end, we examined differences in factor importance due to gender,

Overall, the results of the ABWS and of the focus group dis-

work status and level of Peepoo bag use (to capture extent

cussions corroborated ﬁndings from previous studies on

of user experience).

demand for sanitation. Focus group discussions conﬁrmed

Categorical covariates were used when examining differ-

that sexual violence after sunset is feared by both male

ences in factor importance across segments of respondents

and female sanitation users. The human preference for defe-

based on the ﬁndings by Seaman & Richardson (). To

cation in a safe location (Dellström Rosenqvist ) was

examine pair-wise equality in medians and distributions of

conﬁrmed by the Peepoo users choosing a safety-related

importance shares across segments of respondents based

factor (27) as most important. The importance of safe sani-

on the HB estimations with covariates, the independent-

tation in terms of avoiding danger and violence therefore

samples median (M) and Mann–Whitney (M-W) rank tests

seems to be similar among shared and personal toilet users.

were employed. The average percentage certainty (APC) measure

Personal sanitation provides comfort and convenience

(Hauser ), obtained as the difference between the log likelihood of each model and the log likelihood of a

The second most important factor (22) represents a motiv-

chance model, was used to assess model ﬁt. In addition, a

ation for comfort. This could be connected to the Peepoo

chance ratio measure deﬁned as APC divided by the predic-

bag users’ desire for a proximate and personal sanitation sol-

tive power of a chance model was used to compare the

ution, considering the non-existence of other options for

predictive accuracy between model speciﬁcations with and

household-owned toilets and having to share with strangers

116

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

the poorly maintained public toilets. The convenience of

ﬁndings in other studies on sanitation adoption in rural and

being able to use the Peepoo bag whenever necessary was

urban settings (Cairncross ; Jenkins & Curtis ; Jen-

also rather highly ranked (6). This is due to saved effort in

kins & Scott ; Joshi et al. ). Despite Peepoo being

walking to and queuing at a communal toilet and avoiding

a bag, much like the ‘ﬂying toilet’, users appreciate the

open defecation at night-time. In the focus group discus-

hygiene and cleanliness it provides relative to their alterna-

sions, the easy access to Peepoo and the availability of

tives of public or ﬂying toilets. Training in use of the bag

Peepoo at any time were mentioned as important by the dis-

may be a signiﬁcant contributor to the correct use of the

cussants, who said that they can call their sales lady at any

bag and perceived cleanliness. However, factors related to

time to come and replenish their Peepoo stocks. In addition,

elimination of odour were ranked as only of average impor-

the convenience of Peepoo when suffering from diarrhoea

tance, suggesting either that those aspects are not fully

was named by a focus group discussant. Despite similarity

recognised by the users, or actually conﬁrming the anecdo-

to a ﬂying toilet, the hygienic character of Peepoo could

tal evidence that the odour remains after some time. This

be considered to provide comfort to users. Providing con-

was a somewhat unexpected result, since Peepoo allows

venience and comfort for sick and old relatives is among

users to avoid the sight and smell of their own and others’

the top three motivations for constructing a household

faecal matter, thus providing an alternative to dirty and

toilet in rural and peri-urban Ghana ( Jenkins & Scott

often full communal facilities (Schouten & Mathenge ).

), whereas for Peepoo users it was the eighth most important motivational factor, although it was rated of

Health, value for money and usability away from home

higher importance in the focus group discussions. Consider-

of secondary importance

ing that an adult is incapacitated an average of 2 days when suffering from diarrhoea and sick children require care for

When considering health aspects, the overall evaluative

an average of 2 hours per day (WSP ), it is surprising

aspect of perceived improved health (factor 21) was

that this convenience factor was not ranked higher.

ranked as the seventh most important motivational factor,

Factor 18 ‘ease of use by children’ was perceived to be

demonstrating a link between user experience and perceived

among the top ﬁve in importance. Focus group discussions

improved health. Future research could examine the health

conﬁrmed that it is inconvenient for mothers to have to

impacts of using personal sanitation, including the causal

assist children at communal and shared toilets (most often

relationship between personal sanitation and improved

pit latrines), whereas Peepoo can be used at home by the

health. Surprisingly, the health-related factor relating to

children themselves.

improved school attendance by children (26) was ranked

Saving time in walking to public toilets was only the

next to last in importance, while the factor reduced health-

11th most important motivational factor, whereas in other

care costs was ranked of moderate importance. The

studies distance to defecation sites emerged as an important

moderate connectivity between general health conse-

factor of dissatisfaction and a motivator to build a house-

quences and economic outcome may be explained by a

hold latrine in rural and peri-urban Ghana and rural

shortage of household funds for medical services, respon-

Benin (Jenkins & Curtis ; Jenkins & Scott ). The

dents’ poor general health status (e.g. other pathways of

low signiﬁcance of saved time from using Peepoo may be

faecal-oral transmission), requiring other treatments or the

due to the proximity of communal toilets or plot toilets in

free medical care provided by non-government organis-

this slum setting, compared with the rural and peri-urban

ations. However, the focus group discussions conﬁrmed

settings in the majority of other studies.

the reasons to be related to lack of causal identiﬁcation. Value-for-money aspects of the sanitation solution were

Desire for cleanliness

not considered of high importance. The low importance of receiving a refund may be explained by the fact that

Hygiene and cleanliness was the third most important moti-

people are already saving money using the bags, which are

vational category for Peepoo users, which is consistent with

cheaper than public toilets and also more convenient.

117

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

(This potential explanation was suggested by a reviewer for

factor 2 (Median test statistic (M) ¼ 4.817, p ¼ 0.028;

which we are grateful.) Hence, the value of the refund is not

M-W ¼ 2.501, p ¼ 0.012); factor 13 (M ¼ 4.817, p ¼ 0.028;

sufﬁcient to compensate for the extra effort and possible

M-W ¼ 2.667, p ¼ 0.008); factor 14 (M-W ¼ 1.946, p ¼

inconvenience of storing used bags and taking them to

0.052); and factor 26 (M ¼ 3.224, p ¼ 0.073; M-W ¼ 2.545,

drop-off locations. Another potential explanation is fear of

p ¼ 0.011).

negative social evaluation (i.e. shyness) when presenting used bags for refund, which is a situation that can invoke

Limitations and lessons learned from using ABWS in the

disgust in others. The focus group discussions suggested

study setting

that people prefer to simply give the bags to collectors, who then obtain the refund. The collectors present in the

It was a challenge to recruit a gender-balanced sample in the

focus group discussions reported that this refund is an

study area, since men seemed to work during the day time

important source of income for them. Thus whether the

and administering the survey after sunset was considered

low importance given to value for money can be assigned

too dangerous. Saturday was the only day male Peepoo

to altruism or negative social evaluation remains an open

users could participate in the survey. This made the pro-

question.

portion of male respondents in the study smaller than

The least important factor was convenience when away

wished for and the sample was therefore unbalanced. How-

from home (factor 12), due to lack of privacy when outside

ever, the choice data revealed relatively little heterogeneity

the home and the need for a plastic or tin bucket to hold the

across sub-groups of the sample, including gender. This,

bag.

together with the advantage of HB in generating individ-

Motivational factors are not gender-sensitive in the

), should make generalisations of sub-group character-

sampled group

istic-related ﬁndings less problematic. Moreover, on the

ual-speciﬁc data from sparse data sets (Lagerkvist et al.

same argument, generalisation of the results over the Unexpectedly, no gender differences in importance were

whole population of Kibera or even Silanga village should

revealed in motivational factor importance. Lack of dignity,

not be problematic. A more important caveat is that the

privacy and safety due to inadequate sanitation, but also

data for this study were obtained from existing users. The

caring for incapacitated children and the elderly, place a

gap between preferences, intentions and choice, as noted

greater burden on women than men. Women are also less

by Jenkins & Scott (), makes generalisation of results

likely to leave the slum for work during the day and there-

to current non-users the most challenging aspect when seek-

fore

The

ing to understand drivers of adoption of this personalised

segmentation based on covariates showed that the hypoth-

sanitation system. However, we believe that even a small

eses of equality in medians and distributions for factor

sample can provide information on the importance of the

importance between men and women, and that of equality

different factors for a larger population, since people living

between respondents with a usage ratio higher and lower

in the same area are likely to share the socio-cultural context

than one, could not be rejected.

and interpersonal associations leading to socially con-

less

accustomed

alternative

toilets.

Interestingly, the overall model ﬁt was somewhat lower

structed meanings, including emotions.

when categorical covariates were considered in the esti-

We believe that the ABWS exercise was rather simple to

mation. The APC was reduced to 43.6, corresponding to a

perform for the respondents. The survey was in written

chance ratio of 2.62 and a CV of 0.19. Factor importance

format but each choice set was read aloud to illiterate

equality for the hypotheses of equality of medians and distri-

respondents (which were only two elderly respondents).

butions, respectively, was rejected for some employment-

The literate respondents ﬁlled in the questionnaire at their

related covariates. Speciﬁcally, respondents who were

own pace and the illiterate participants followed the reading

unemployed or worked a maximum of three days per

of the choice sets at the pace of the slowest respondent. The

week were found to assign low levels of importance to

two facilitators were a necessary resource, especially for the

118

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

illiterate respondents and for clarifying any questions that

communicate these non-health beneﬁts, which contribute to

were raised by the respondents in general. The simple

meeting basic human needs, raising awareness of health con-

choice exercise we conducted prior to starting with the ques-

sequences and inducing adoption of new sanitation systems.

tionnaire helped the respondents understand the kind of

Employing the straightforward, accurate and easy-to-

choice decision being requested.

administer data collection methods used in this study could help public policy makers become informed about public preferences for sanitation. The ABWS method pro-

CONCLUSIONS

vides an understanding of motivational needs for personal sanitation based on information on why households adopt

This study examined people’s motivation to use a personal

such a personal solution. This can further guide sustainable

sanitation solution, the Peepoo bag. A framework of motiva-

sanitation planning and public health management and

tional factors for personal sanitation was developed. The

facilitate the marketing of low-cost sanitation solutions to

focus on behavioural motivations can and should form an

the poor.

important complement to infrastructure-oriented interventions. Anchored best–worst scaling was used to determine the importance attributed by urban slum dwellers in Nairobi

ACKNOWLEDGEMENT

to the personal toilet. The results were slightly different from those of previous studies in that no support was found for the

Senior authorship is shared by Carl Johan Lagerkvist and

presence of gender differences in motivational factors for per-

Suvi Kokko. This research was funded by core research

sonal sanitation. However, the small sample size (122) and

funding from the Swedish University of Agricultural

small number of men (31) participating in the study may

Sciences.

have contributed to this result. All respondents ranked highly factors of wellbeing such as privacy, comfort and convenience, and hygiene and cleanliness. Status, in terms of

REFERENCES

pride in having such a sanitation solution and in contributing to a modern lifestyle, which was a leading motive for latrine adoption amongst men in other studies, was not supported in this study, most likely due to the low and unmanly proﬁle of the Peepoo bag. Construction of pit latrines and similar technologies is relatively costly, making it reasonable to generate certain endowment values (such as prestige) at the community level, as well as social pressure at the individual level. However, the less expensive, and more accessible, personalised Peepoo system means that personal safety and convenience are leading motivational factors. As in previous studies, hygiene and cleanliness were ranked high in importance, presumably because of their connection with dignity, social acceptance and ﬁtting in with society. Motivational factors directly related to a general sense of health improvement were relatively highly ranked, but health seemed to be secondary to other aspects of wellbeing, such as safety, cleanliness and comfort. Policies and investments directed toward adoption and use of low-cost personal sanitation systems should acknowledge and further

Aiello, A. E., Larson, E. L. & Sedlak, R.  Hidden heroes of the health revolution, sanitation and personal hygiene. American Journal of Infection Control 36 (10 Suppl), 128–151. Amnesty International  Risking rape to reach a toilet: women’s experiences in the slums of Nairobi, Kenya. Index AFR 32/006/2010. Biran, A., Jenkins, M. W., Dabrase, P. & Bhagwat, I.  Patterns and determinants of communal latrine usage in urban poverty pockets in Bhopal, India. Tropical Medicine and International Health 16 (7), 854–862. Cairncross, S.  Sanitation and Water Supply: Practical Lessons from the Decade. International Bank for Reconstruction and Development, World Bank, Washington, DC. Carolini, G.  Framing water, sanitation, and hygiene needs among female-headed households in periurban Maputo, Mozambique. American Journal of Public Health 102 (2), 256–262. Curtis, V. & Biran, A.  Dirt, disgust, and disease: is hygiene in our genes? Perspectives in Biology and Medicine 44 (Winter), 17–31. Dangert, J-O.  Norms and attitudes towards Ecosan and other sanitation systems. Report 2004–5, EcoSanRes Publications Series. EcoSanRes Programme and Stockholm Environment Institute, Stockholm.

119

C. J. Lagerkvist et al.

Framing motivational factors for personal sanitation in slums

Dellström Rosenqvist, L. E.  A psychosocial analysis of the human-sanitation nexus. Journal of Environmental Psychology 25 (3), 335–346. Flynn, T. N., Louviere, J. J., Peters, T. J. & Coast, J.  Best–worst scaling: what it can do for health care research and how to do it? Journal of Health Economics 26 (1), 171–189. Hauser, J. R.  Testing and accuracy, usefulness and signiﬁcance of probabilistic choice models: an informationtheoretic approach. Operations Research 26 (3), 406–421. Isunju, J. B., Schwartz, K., Schouten, M. A., Johnson, W. P. & van Dijk, M. P.  Socio-economic aspects of improved sanitation in slums: a review. Public Health 125 (6), 368–378. Jenkins, M. W. & Curtis, V.  Achieving the ‘good life’: why some people want latrines in rural Benin. Social Science & Medicine 61 (11), 2446–2459. Jenkins, M. W. & Scott, B.  Behavioral indicators of household decision-making and demand for sanitation and potential gain from social marketing in Ghana. Social Science & Medicine 64 (12), 2427–2442. Jenkins, M. W. & Sugden, S.  Rethinking sanitation – lessons and innovation for sustainability and success in the new millennium. Sanitation Thematic Background Paper commissioned by the UNDP for the Human Development Report 2006. Joshi, D., Fawcett, B. & Mannan, F.  Health, hygiene and appropriate sanitation: experiences and perceptions of the urban poor. Environment and Urbanization 23, 91–111. Katukiza, A. Y., Ronteltap, M., Oleja, A., Niwagaba, C. B., Kansiime, F. & Lens, P. N. L.  Selection of sustainable sanitation technologies for urban slums – A case of Bwaise III in Kampala, Uganda. Science of the Total Environment 409 (1), 52–62.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Kenya National Bureau of Statistics  2009 Kenya Population and Housing Census. Kenya National Bureau of Statistics Nairobi, Kenya. Lagerkvist, C. J., Okello, J. J. & Karanja, N.  Anchored vs. relative best–worst scaling and latent class vs. hierarchical Bayesian analysis of best–worst choice data: investigating the importance of food quality attributes in a developing country. Food Quality and Preference 25 (1), 29–40. Marley, A. A. J. & Louviere, J. J.  Some probabilistic models of best, worst, and best–worst choices. Journal of Mathematical Psychology 49 (6), 464–480. Peepoople  Urban slums. Available at: http://www.peepoople. com/we-are-all-peepoople/urban-slums/ (accessed 25 April 2012). Phillips, M. L., Fahy, T., David, A. S. & Senior, C. (eds.)  Disgust: the forgotten emotion of psychiatry. The British Journal of Psychiatry 172, 373–375. Potoglou, D., Burge, P., Flynn, T., Netten, A., Malley, J., Forder, J. & Brazier, J. E.  Best–worst scaling vs. discrete choice experiments: an empirical comparison using social care data. Social Science & Medicine 72 (10), 1717–1727. Schouten, M. A. C. & Mathenge, R. W.  Communal sanitation alternatives for slums: a case study of Kibera, Kenya. Physics and Chemistry of the Earth 35 (13), 815–822. Seaman, S. R. & Richardson, S.  Bayesian analysis of casecontrol studies with categorical covariates. Biometrika 88 (4), 1073–1088. Water and Sanitation Programme (WSP)  Economic Impact of Poor Sanitation in Africa – Kenya. Water and Sanitation Programme, World Bank, Washington, DC. Whaley, L. & Webster, J.  The effectiveness and sustainability of two demand-driven sanitation and hygiene approaches in Zimbabwe. Journal of Water, Sanitation and Hygiene for Development 1 (1), 20–36.

First received 3 May 2013; accepted in revised form 9 September 2013. Available online 21 October 2013

120

04.1

2014

School-based intervention: evaluating the role of water, latrines and hygiene education on trachoma and intestinal parasitic infections in Ethiopia Bizu Gelaye, Abera Kumie, Nigusu Aboset, Yemane Berhane and Michelle A. Williams

ABSTRACT We sought to evaluate the impact of a hygiene and sanitation intervention program among schoolchildren to control active trachoma and intestinal parasitic infections. This longitudinal epidemiologic study was conducted among 630 students in rural Ethiopia. Baseline and follow-up surveys were conducted to evaluate the impact of a three-pronged intervention program: (i) construction of ventilated improved pit latrines; (ii) provision of clean drinking water; and (iii) hygiene education. Socio-demographic information was collected using a structured questionnaire. Presence of trachoma and intestinal parasitic infections were evaluated using standard procedures. At baseline, 15% of students had active trachoma, while 6.7% of them were found to have active trachoma postintervention (p < 0.001). Similar improvements were noted for parasitic infections. At baseline, 7% of students were reported to have helminthic infections and 30.2% protozoa infections. However, only 4% of students had any helminthic infection and 13.4% (p < 0.001) of them were found to have any protozoa infection during follow-up surveys. Improvements were also noted in students’ knowledge and attitudes towards hygiene and sanitation. In summary, the results of our study demonstrated that provision of a comprehensive and targeted sanitation intervention program was successful in reducing the burden of trachoma and intestinal parasitic infection among schoolchildren. Key words

| Ethiopia, intervention, parasitic infection, sanitation, school, trachoma

Bizu Gelaye (corresponding author) Department of Epidemiology, University of Washington School of Public Health, Seattle, WA, USA E-mail: bgelaye@hsph.harvard.edu Bizu Gelaye Michelle A. Williams Department of Epidemiology, Harvard School of Public Health, 677 Huntington Ave, K 501, Boston, MA, 02115 USA Abera Kumie Faculty of Medicine, Addis Ababa University, Addis Ababa, Ethiopia Abera Kumie Yemane Berhane Addis Continental Institute of Public Health, Addis Ababa, Ethiopia Nigusu Aboset Family Health International, Addis Ababa, Ethiopia

INTRODUCTION Trachoma is a leading cause of preventable infectious blind-

of a mechanical vector of C. trachomatis, the eye-seeking

ness in the world (WHO ). It is caused by infection with

Musca sorbens, by reducing the amount of exposed human

Chlamydia trachomatis bacterium, one of the most common

feces, the breeding media of M. sorbens (Emerson et al.

human pathogens (Schachter et al. ). The World Health

). To combat the burden of trachoma, the WHO

Organization (WHO) estimates that trachoma is endemic in

endorsed an integrated four components strategy known

56 countries, most within Africa and the Middle East, and

as SAFE (Surgery, Antibiotics, Face washing and Environ-

causes 3.6% of all blindness (Resnikoff et al. ). It is cur-

mental

rently reported that close to 1.2 million individuals are blind

improvement of latrines, access to clean water and encour-

from trachoma while about 21.4 million suffer from active

aging the use of water for face washing are two of the

trachoma. Factors most consistently associated with tra-

important parts (‘F’ and ‘E’) of the WHO-recommended

choma include inadequate access to water and sanitation

SAFE strategy for trachoma control (WHO ).

sanitation)

(WHO

).

Provision

and

facilities (Baggaley et al. ; Polack et al. ). Latrines

Intestinal parasitic infections, comorbid with trachoma,

have been proposed to play a critical role in the exposure

are the second most important causes of outpatient

doi: 10.2166/washdev.2013.060

121

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

morbidity in Ethiopia (Alemu et al. ). Importantly, intes-

their learning potential and their educational achievement

tinal parasitic infections destroy the well-being and learning

(WHO ). Therefore, in this intervention study we

potential of millions of schoolchildren in developing

sought to evaluate the impact of a hygiene and sanitation

countries (WHO ). Although a number of local and

intervention program among Angolela primary education

international

students in controlling active trachoma and intestinal para-

health

organizations

have

implemented

deworming programs in hardest-hit areas, sustained beneﬁts

sitic infections.

have been elusive, in part, because interventions were provided in isolation rather than in combination with sanitary improvements and hygiene education programs (Knopp

METHODS AND MATERIALS

et al. ). Prior studies have shown that inadequate sanitary conditions and poor hygiene practices play a major

This longitudinal epidemiologic study was conducted in

role in the increased burden of gastrointestinal infections

Angolela Woreda, North Showa zone during October

(WHO ).

2008 and May 2009. Angolela Woreda is located 140 km

Studies conducted in Ethiopia documented high levels

from Addis Ababa (capital city) (Figure 1).

of active trachoma (Berhane et al. ) and intestinal parasitic infections (Belyhun et al. ; Alemu et al. ) and signify the need for timely and sustainable efforts aimed at preventing new infections and treating prevalent cases. Despite the magnitude of these problems, to the best of our knowledge, comprehensive sanitation intervention programs have not been systematically evaluated in Ethiopia. Control of active trachoma and intestinal parasitic infections alleviates suffering, reduces poverty and supports equal opportunities among young children. A schoolbased hygiene and sanitation program is one of the most widely used and effective approaches to control trachoma and parasitic infections (Lewallen et al. ). Notably, school enrollment is on the rise in most parts of the world (UNESCO ). Ethiopia is on track to achieve universal primary education, one of the United Nations Millennium Development Goals, with enrollment reaching up to 95.9% (MoFED ). With the vast majority of school-age children now enrolled, schools present an opportunity to reach thousands of children with safe

Baseline survey The ﬁrst survey was conducted in October 2008 to establish baseline prevalence estimates of active trachoma, intestinal parasitic infection as well as estimates of knowledge, attitudes and practices (KAP) of hand washing among students. The survey questionnaire consisted of: demographic information (grade, gender); mother and father literacy (no, yes); and frequencies of bathing, washing feet/ hair, brushing teeth and changing clothes. Students were asked whether their hands were washed during the day prior to interview (no, yes); their reasons for washing hands (after defecation, before meals); and what materials were used for hand washing (soap and water, water only). All 669 students attending grades 1–6 at Angolela Primary School were included in the survey (median age ¼ 11 years old). Following the baseline survey, improvements in hygiene and sanitation were implemented by the school.

water and hygiene and health messages. Targeting schools for water and sanitation improvement and hygiene training

Intervention components

has multifaceted beneﬁts including reaching children from households at all socio-economic levels (Nagpal ) as

The school implemented the following improvements in

primary education is very highly subsidized or completely

sanitation and hygiene.

free (UNESCO ; MoFED ; Nagpal ). Institut-

1. Low-cost, eight-seat ventilated improved pit (VIP)

ing health promotion and disease prevention programs in

latrines were newly constructed for female students.

schools beneﬁts not only the students but also the family,

Another existing eight-seat pit latrine was refurbished;

community and country as a whole. Improved health

contents were emptied and became functional for use

among students will improve their school attendance,

by male students.

122

Figure 1

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Map of Angolela Woreda, Ethiopia (created using the WHO health mapper software).

2. Two ﬁberglass water tankers each with capacities of three

After the baseline survey, students who were found to be

cubic meters and with three taps were installed, separ-

positive for parasites and active trachoma (described below)

ately for both males and females. Tankers were ﬁlled

were treated for free with appropriate doses of nationally

every other day with clean water from a nearby protected

recommended drugs: albendazole for intestinal parasitic

spring water source using jerry cans of 10 and 20 liters

infection (a WHO-recommended drug and noted to be

capacity to promote hand washing after toilet use as

highly

well as to encourage students to wash their faces. The

(Vercruysse et al. ) and tetracycline eye ointment for

protected spring had an effective protection around it.

active trachoma.

efﬁcacious

against

round-

and

hookworms)

In addition, set back distance, based on travel time, was established to protect against possible contaminant

Post-intervention survey

sources such as latrines. 3. A student-led health club was formed to educate students

Since October 2008 all students have had access to all the

every morning before class starts during morning parades

intervention components. A follow-up study of students

and using special sessions about basic hygiene and sani-

enrolled in the baseline survey was conducted in May

tation. Members of the club and the teachers heading

2009 to evaluate the extent to which prevalence estimates

the club were oriented on proper hygiene and disease

of infections, as well as estimates of KAP had changed

prevention. Different types of posters and printed

after interventions were implemented.

materials on hygiene and sanitation were supplied. The health club’s main functions included: supervising the

Socioeconomic and personal risk factors survey

availability of water inside tankers; supervising the cleanliness of latrines; instructing students on how to use

A structured questionnaire (translated from English and

latrines and water stands; and communicating health

printed in the local Amharic language) was used to collect

messages to students.

data.

The

questionnaire

included

information

123

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

socioeconomic, sanitary, environmental and demographic

who were found to harbor an intestinal parasite were treated

risk factors. It also included questions concerning parental

with albendazole for free at a single dose of 400 mg orally

(mother and father) literacy, students’ gender, age and

(WHO ). All treatments were provided without charge

grade in school. Prior to use, the questionnaire was pre-

to students or their families.

tested in Dalcha Elementary School in Basona Worena Woreda, Ethiopia to assess the suitability of the questionnaire with regard to duration, language appropriateness

ETHICAL CONSIDERATION

and question comprehensibility. The questionnaire was administered by trained research personnel who performed

Ethical approval for all study procedures was granted by the

in-person interviews of all participating students.

Institutional Review Board of Addis Continental Institute of Public Health (IRB of ACIPH) in Addis Ababa, Ethiopia.

Eye examination

Approval from the Woreda Health Ofﬁce and the Woreda

All students were examined for clinical signs of active

ment of this study. Verbal consent was obtained from

trachoma using the WHO simpliﬁed clinical grading

parents or appropriate guardians of eligible children and

scheme by trained and standardized ophthalmic nurses.

assent was obtained from eligible children before they

Each ophthalmic nurse was standardized against a highly

were included in the study in accordance with the principles

experienced ophthalmologist in trachoma diagnosis (gold

of the declaration of Helsinki. Written consent was not

standard). The same nurses were used for baseline and

deemed appropriate, given the low literacy rate in Angolela

post-intervention surveys. In accordance with the WHO

Woreda and the research involved no more than minimal

Education Ofﬁce was also granted prior to the commence-

grading scheme (WHO ), active trachoma was deﬁned

risk to the subjects. Study procedures including the verbal

as the presence of either trachomatous follicular (TF) or tra-

consenting process were approved by the IRB of ACIPH.

chomatous inﬂammation intense (TI). Each student was also

Documentation of verbal assent was initiated and dated by

examined for trachomatous trichiasis (deﬁned as either in-

the interviewers on data collection forms as approved by

turned eyelashes rubbing on the eye or evidence of pre-

the IRB. Before analysis, personal identiﬁers were removed

viously removed lashes) and trachomatous scarring (TS).

from each data set. The Human Subjects Division of the University of Washington, USA granted approval to use

Parasitological examination

the de-identiﬁed and anonymized data set for analysis.

About 3 grams of fresh stool samples were collected in plas-

Statistical analysis

tic cups and were labeled with the students’ unique study identiﬁcation number. Following stool collection, samples

Data were entered in the program EPI-INFO (Version

were preserved in a tube containing 10% formalin in

3.3.2), a public access software made available from the

0.85% saline. Samples were then taken to Debre Birhan

US Centers for Disease Control and Prevention (CDC

Hospital, 10 km from the study site, for processing, using

Atlanta, GA, USA), and the analysis was completed using

the ether concentration technique for fecal examination to

SPSS (version 17.0, SPSS, Inc., Chicago, IL, USA). Fre-

diagnose infections with intestinal worms and protozoa, fol-

quency distributions of students’ socio-demographic and

lowing WHO standard operating procedures for the

behavioral characteristics according to gender were exam-

parasitological examination of feces (WHO ). To evalu-

ined. Continuous variables were expressed as mean ±

ate diagnostic accuracy, 5% of the fecal samples were

standard error of mean. Categorical variables were

randomly re-processed by a separate lab technician from

expressed as number (percentage, %). Since the same stu-

Debre Birhan Hospital and the results were compared

dents were involved in the pre- and post-intervention

with the results made by the original lab technician. In

assessments, we used McNemar’s test for paired data to

accordance with the WHO recommendation, students

evaluate signiﬁcance while accounting for correlations

124

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

between the two time periods. All reported tests of statistical signiﬁcance were two-sided set at α ¼ 0.05.

RESULTS The socio-demographic and personal characteristics of Angolela schoolchildren during baseline and post-intervention surveys are presented in Table 1. At baseline a total of 669 students with a median age of 11 years (interquartile range 9.8–13.0) participated in the study while 630 students participated during the post-intervention survey with a median age of 11 years (interquartile range 9.7–13.0). There was no signiﬁcant difference in the gender distri-

Figure 2

Knowledge about sanitation and hygiene practices.

bution, age and parental literacy rates. As shown in Figure 2, there was a marked improvement in KAP toward sanitation among students after intervention. At baseline, approximately, 61% of students felt that boiling water kills germs, while 74% of them reported this postintervention. At baseline 73% of students thought human

Table 1

feces contain germs, whereas during the follow-up survey 88% of students reported this. Similarly, as shown in Figure 3, signiﬁcant improvements were noted across all objectively observed personal hygiene characteristics (all p-value <0.01). As depicted in Figure 4, improvements in hand washing practices were also noted.

Personal and household characteristics

As shown in Table 2, there was a signiﬁcant decrease in

Pre-intervention

the prevalence of active trachoma (p < 0.001). At baseline

Post-intervention

n (total ¼ 669)

n (total ¼ 630)

110

16.4

125

19.8

188

28.1

165

26.2

114

17.0

14.3

10.3

11.3

100

16.3

100

15.9

11.7

12.5

Female

326

48.7

304

48.3

Male

343

51.3

325

51.7

399

59.7

345

54.8

Yes

263

39.3

269

42.7

1.0

2.5

210

31.4

168

26.7

Yes

437

65.3

436

69.2

3.3

4.1

Characteristics

15% of students had active trachoma while, notably, only 6.7% of students were found to have active trachoma

Grade

during the post-intervention survey.

Gender

Similar improvements were noted for intestinal parasitic infections (Table 3). At baseline 7% of students were diagnosed with any helminthic infection and 30.2% were found to have any protozoa infection. These prevalence estimates, however, were signiﬁcantly reduced (p ¼ 0.04) postintervention. Only 4% of students had any helminthic infection and 13.4% of them were found to have any protozoa infection.

Mother literate

Don’t know Father literate

Don’t know

DISCUSSION The results of our study demonstrated that provision of a comprehensive and targeted sanitation intervention project was successful in reducing the burden of trachoma and intestinal parasitic infection among schoolchildren in Angolela, Ethiopia. To date, we are not aware of any published

125

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Figure 3

Objectively observed personal hygiene characteristics.

Figure 4

Hand washing practices.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

article that systematically evaluated the impact of access to

The ﬁndings from our study on the positive impact of

water, latrines and hygiene education on eye infection tra-

access and use of latrines on trachoma are in general agree-

choma and intestinal parasitic infections in Ethiopia.

ment with others (Rabiu et al. ). Because our study is one

126

Table 2

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

with a 3.3 -fold increased odds of trachoma. Moreover, pres-

Prevalence of trachoma in study population

Pre-intervention

ence and use of a pit latrine in households, even when full

Post-intervention P-value

Sign of trachoma n (total ¼ 669)

n (total ¼ 630)

628

91.9

610

96.8

Yes

6.1

3.2

and unscreened, was associated with a 14% reduction in trachoma prevalence (Courtright et al. ). Similarly in Ankober, Ethiopia, Golovaty et al. (), using a

TI 0.05

community-based study of children aged 1–9 years, reported that those without access to a latrine were nearly ﬁve times more likely to have active trachoma. Other investigators

TF No

596

89.1

599

95.1

Yes

10.9

4.9

<0.001

have found protective effects of access and use of latrines on the incidence of trachoma among young children (Emerson et al. ; Burton et al. ; Montgomery et al. ).

TS No

620

92.7

609

96.7

Yes

7.3

3.3

0.02

In contrast to prior ﬁndings, Stoller et al. (), in their study among 24 communities in Northern Ethiopia, found no signiﬁcant reduction in eye infection trachoma among

Any active trachoma

children due to latrine construction. Reasons for these differ-

568

84.9

588

93.3

Yes

101

15.1

6.7

<0.001

ences are unclear. Additionally we found that the prevalence of TS decreased from a baseline 7.3% to 3.3%

TF, trachomatous follicular; TI, trachomatous inﬂammation intense; TS, trachomatous scarring.

during post-intervention. Scarring is not a sign of active infection, but rather it indicates that an individual has had repeated trachoma infections in the past. It is possible that

Table 3

some of the students had repeated infections since interven-

Prevalence of parasitic infections in study population

Pre-intervention

Post-intervention

Intestinal parasitic

Number

infection

infected

Ascaris lumbricoides Hookworm

infected

3.6 11

tion (WHO , ). Collectively, the results of our study and those of others underscore the importance of latrine P-value

provision and use in trachoma prevention. Our study results indicated an overall reduction in hel-

1.7

0.15

minthic and protozoa infection rates. The lack of

0.5

0.81

statistically signiﬁcant reduction in some of the helminthic

1.5

0.2

0.01

infection is, in part, due to small sample size. At baseline

Hymnolepis dimunita

0.3

0.0

Trichuris trichiura

0.3

0.5

0.99

any protozoa infection was found among 30% of students.

Hymenolepis nana

–a

the prevalence of any helminthic infection was 7% while

Enterobius vermicularis

1.6 10

1.6

0.60

These prevalence estimates, however, were signiﬁcantly

Any helminthic infection

7.0 27

4.3

0.04

reduced (p ¼ 0.04) post-intervention. Only 4% of them had

Giardia intestinalis

7.6 10

1.6 <0.001

179

27.8 77

12.3 <0.001

Any protozoa infection 202

30.2 85

13.4 <0.001

Entamoeba

McNemar’s test was not computed as no cases were found post-intervention.

any helminthic infection and 13.4% of them were found to have any protozoa infection. Although the effects of treating all schoolchildren at regular intervals with deworming drugs have been extensively studied, the knowledge base for the effects of screening and then treating infected children is limited (Taylor-Robinson et al. ). In their study among

of the ﬁrst to investigate the impact of water and sanitation

children in Bangladesh, Hall & Nahar () evaluated the

programs on trachoma and parasitic infection in school set-

effectiveness of a single dose of albendazole on Giardia.

tings, our ﬁndings can only tentatively be compared with

They noted that a single dose of albendazole treated 62%

community-based studies. For instance, in their study

of infected children indicating the moderate efﬁcacy of

among Egyptian households, Courtright et al. () found

single doses on affected children (Hall & Nahar ).

that an absence of a latrine in a household was associated

Given that a single dose of albendazole may not be as

127

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

efﬁcacious as multiple doses for protozoa infections, pro-

Given that we have administered tetracycline treatments

vision of clean water, hand washing and other personal

to those students who were found to be positive at baseline,

hygiene measures might have contributed to this reduction

it is difﬁcult to delineate the exact contribution of each inter-

(Speich et al. ). Speich et al., in their recent study

vention component on trachoma infection. Although we do

among school-aged children in Tanzania, evaluated the

not have reinfection rates among schoolchildren, studies

effect of a single dose albendazole, nitazoxanide or albenda-

have shown the moderate success of antibiotic treatment

zole–nitazoxanide treatment. No signiﬁcant effect was

on active trachoma (Hu et al. ). Infection rates were

found in the mean intensity of intestinal protozoa infection

initially falling immediately post-treatment, but increasing

3 weeks after treatment. The authors speculated that a

within 12 months of treatment (Hu et al. ). A study con-

single-dose albendazole or nitazoxanide or a combination

ducted in Egypt, The Gambia and Tanzania noted that the

of the two drugs do not have sufﬁcient efﬁcacy against

prevalence of infection at 1 year after mass treatment was

pathogenic intestinal protozoa (Speich et al. ). In our

substantially lower than at baseline, but was higher than

study, it is possible that the changes in the attitudes and

the prevalence at the 3-month follow-up (Schachter et al.

practices of personal hygiene among students might have

). Collectively, however, research thus far has shown

contributed to the parasitic infection reduction. In compari-

that use of latrines blocks mechanical transmission of tra-

son to baseline assessment, all objectively observed personal

choma by reducing ﬂy breeding (Kumie & Ali ; Stoller

hygiene characteristics were signiﬁcantly improved during

et al. ; Rabiu et al. ). This important environmental

post-intervention assessment.

intervention thereby decreases the amount of contact

Studies have demonstrated that those who used open

between the reservoir of young children with active tra-

ﬁelds for defecation have markedly higher parasite infection

choma through contact with discharge from the eye and

rates than those who used some type of latrine although sus-

possibly the nose via the ﬂy vector. It is important to note,

tainable results can be achieved if coupled with hand

however, that access to a latrine does not necessarily trans-

washing practices (Alum et al. ). Most investigators,

late to increased latrine use (O’Loughlin et al. ).

though not all, have shown previously that availability and

Furthermore, improper use and maintenance of latrines,

use of sanitation facilities is associated with reduced intesti-

particularly shared or communal latrines, may exacerbate

nal

transmission.

parasitic

infections.

recent

meta-analysis

Ziegelbauer et al. () found that improved sanitation is

The ﬁndings of our study should be interpreted in light of

associated with a reduced risk of transmission of parasitic

some study limitations. It is possible that the observed

infections. However, Yajima et al. (), in their study

reduction in trachoma and parasitic infection may be due

among a rural agricultural community in Vietnam, found

to the group treatment effect where there was no control com-

no association between helminthic infections and presence

parison group. The concordance of our results with those

and use of latrines. Intestinal parasitic infections occur at

from prior studies, however, serves to attenuate some of

a critical time in life with a maximum intensity in the age

these concerns. In addition, it has been reported that there

range of 5 to 14. Their adverse health consequences are

is seasonal variation in density of eye-seeking ﬂies in Ethiopia

enormous including permanent organ damage, anemia,

(Taye et al. ). In the month of May (when the post-inter-

poor physical growth, poor intellectual development and

vention survey was conducted), the population of eye-seeking

impaired cognitive function (WHO ). School-based

ﬂies is expected to be highest (Taye et al. ). As a result, the

deworming is one of the widely recommended strategies to

magnitudes of the reported effects of interventions on active

combat intestinal parasitic infections among schoolchildren

trachoma are likely to be conservative. Additionally, it is

(Matthys et al. ). Although deworming programs are

possible that the students’ exposure to trachoma and parasitic

commendable, we and others recognize that sustained

infections might differ by place of residence. Future studies

beneﬁts can only be achieved if coupled with a multi-

need to examine the effect of both school and residential

pronged approach of hygiene education and sanitation

exposures and family contacts for the identiﬁcation and con-

improvements (Knopp et al. ; Stoller et al. ).

trol of trachoma. Finally, it is possible that some educational

128

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

efforts conducted at a national and regional level might have

collection and analysis, decision to publish, or preparation

contributed to increased awareness about hand and face

of the manuscript. The authors wish to thank Feed the

washing among students.

Children Ethiopia, Angolela Primary School and Addis

In conclusion, as evidenced in our study, provision of

Continental Institute of Public Health for providing

latrines, water and hygiene education was found to be protec-

facilities and logistics support throughout the research

tive of trachoma and intestinal parasitic infections. In

process. The authors would also like to thank Debre

developed countries where major improvements in water,

Birhan Hospital, the Woreda Health Ofﬁce and the

sanitation and hygiene behaviors have occurred, trachoma

Woreda Education Ofﬁce for providing testing facilities

and intestinal parasitic infections ceased to be public health

and granting access to conduct the study.

problems. This change happened in the absence of any trachoma or parasitic infection-speciﬁc interventions such as antibiotic treatment (Burton et al. ) . The solution for

REFERENCES

these problems remains simple, basic and a fundamental core of public health – improved sanitation. Schools are important environments to initiate and promote healthy behaviors and improved sanitation practices. Improving water and sanitation in schools will improve health, intellect and school attendance. Consequently, school performance will improve, child mortality will decline and economic productivity will increase (Taylor-Robinson et al. ). As illustrated in this study, improving sanitary infrastructure and provision of clean water and hygiene education is feasible with modest cost. The sustainability of these programs depends on health and economic beneﬁts perceived by various stakeholders, commitment of teachers and school principals, involvement of community members and support of governmental and non-governmental organizations to promote and ﬁnance water and sanitation improvements (Hutton et al. ). Clearly, mobilizing local resources to improve school sanitation facilities should be actively sought by relevant stakeholders.

ACKNOWLEDGMENT Provision of VIP latrines and water tankers was made possible through the donation from Scott and Ann Marie Robertson, an anonymous donor and funds from the Brotman Award In partnership with Feed the Children (FTC) Ethiopia and Addis Continental Institute of Public Health (ACIPH). The research was supported in part by an award from the National Institutes of Health, National Institute of Minority Health and Health Disparities (T37MD001449). The funders had no role in study design, data

Alemu, A., Atnafu, A., Addis, Z., Shiferaw, Y., Teklu, T., Mathewos, B., Birhan, W., Gebretsadik, S. & Gelaw, B.  Soil transmitted helminths and Schistosoma mansoni infections among school children in Zarima town, northwest Ethiopia. BMC Infect. Dis. 11, 189. Alum, A., Rubino, J. R. & Ijaz, M. K.  The global war against intestinal parasites: should we use a holistic approach? Int. J. Infect. Dis. 14 (9), e732–e738. Baggaley, R. F., Solomon, A. W., Kuper, H., Polack, S., Massae, P. A., Kelly, J., Safari, S., Alexander, N. D., Courtright, P., Foster, A. & Mabey, D. C.  Distance to water source and altitude in relation to active trachoma in Rombo district, Tanzania. Trop. Med. Int. Health 11 (2), 220–227. Belyhun, Y., Medhin, G., Amberbir, A., Erko, B., Hanlon, C., Alem, A., Venn, A., Britton, J. & Davey, G.  Prevalence and risk factors for soil-transmitted helminth infection in mothers and their infants in Butajira, Ethiopia: a population based study. BMC Public Health 10, 21. Berhane, Y., Worku, A., Bejiga, A., Adamu, A., Alemayehu, W., Bedri, A., Haile, Z., Ayalew, A., Adamu, Y., Gebre, T., Kebede, T. D., West, E. & West, S.  Prevalence of Trachoma in Ethiopia. Ethiopian J. Health Dev. 21 (3), 211–215. Burton, M. J., Holland, M. J., Makalo, P., Aryee, E. A., Sillah, A., Cohuet, S., Natividad, A., Alexander, N. D., Mabey, D. C. & Bailey, R. L.  Profound and sustained reduction in Chlamydia trachomatis in The Gambia: a five-year longitudinal study of trachoma endemic communities. PLoS Negl. Trop. Dis. 4 (10). Courtright, P., Sheppard, J., Lane, S., Sadek, A., Schachter, J. & Dawson, C. R.  Latrine ownership as a protective factor in inflammatory trachoma in Egypt. Br. J. Ophthalmol. 75 (6), 322–325. Emerson, P. M., Cairncross, S., Bailey, R. L. & Mabey, D. C.  Review of the evidence base for the ‘F’ and ‘E’ components of the SAFE strategy for trachoma control. Trop. Med. Int. Health 5 (8), 515–527. Emerson, P. M., Lindsay, S. W., Alexander, N., Bah, M., Dibba, S. M., Faal, H. B., Lowe, K. O., McAdam, K. P., Ratcliffe, A. A., Walraven, G. E. & Bailey, R. L.  Role of flies and

129

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

provision of latrines in trachoma control: cluster-randomised controlled trial. Lancet 363 (9415), 1093–1098. Golovaty, I., Jones, L., Gelaye, B., Tilahun, M., Belete, H., Kumie, A., Berhane, Y. & Williams, M. A.  Access to water source, latrine facilities and other risk factors of active trachoma in Ankober, Ethiopia. PLoS One 4 (8), e6702. Hall, A. & Nahar, Q.  Albendazole as a treatment for infections with Giardia duodenalis in children in Bangladesh. Trans. R. Soc. Trop. Med. Hyg. 87 (1), 84–86. Hu, V. H., Harding-Esch, E. M., Burton, M. J., Bailey, R. L., Kadimpeul, J. & Mabey, D. C.  Epidemiology and control of trachoma: systematic review. Trop. Med. Int. Health 15 (6), 673–691. Hutton, G., Haller, L. & Bartram, J.  Global cost-benefit analysis of water supply and sanitation interventions. J. Water Health 5 (4), 481–502. Knopp, S., Stothard, J. R., Rollinson, D., Mohammed, K. A., Khamis, I. S., Marti, H. & Utzinger, J.  From morbidity control to transmission control: time to change tactics against helminths on Unguja Island, Zanzibar. Acta. Trop. Kumie, A. & Ali, A.  An overview of environmental health status in Ethiopia with particular emphasis to its organization, drinking water and sanitation: a literature survey. Ethiopian J. Health Dev. 19 (2), 89–103. Lewallen, S., Massae, P., Tharaney, M., Somba, M., Geneau, R., Macarthur, C. & Courtright, P.  Evaluating a schoolbased trachoma curriculum in Tanzania. Health Educ. Res. Matthys, B., Bobieva, M., Karimova, G., Mengliboeva, Z., Jean-Richard, V., Hoimnazarova, M., Kurbonova, M., Lohourignon, L. K., Utzinger, J. & Wyss, K.  Prevalence and risk factors of helminths and intestinal protozoa infections among children from primary schools in western Tajikistan. Parasit. Vectors 4, 195. MoFED  Ethiopia: 2010 MDGs Report. Trends and Prospects for Meeting MDGs by 2015, Ministry of Finance and Economic Development (MoFED). Available at: http://www. undp.org/content/dam/undp/library/MDG/english/MDG% 20Country%20Reports/Ethiopia/ethiopia_september2010. pdf (accessed 14 August 2013). Montgomery, M. A., Desai, M. M. & Elimelech, M.  Comparing the effectiveness of shared versus private latrines in preventing trachoma in rural Tanzania. Am. J. Trop. Med. Hyg. 82 (4), 693–695. Nagpal, T.  Clean Start: Focusing on School Water, Sanitation, and Hygiene: A Reflection From GWC. Available at: http://www.globalwaterchallenge.org/resources/ cleanstartschoolreflection.pdf (accessed 14 August 2013). O’Loughlin, R., Fentie, G., Flannery, B. & Emerson, P. M.  Follow-up of a low cost latrine promotion programme in one district of Amhara, Ethiopia: characteristics of early adopters and non-adopters. Trop. Med. Int. Health 11 (9), 1406–1415. Polack, S., Kuper, H., Solomon, A. W., Massae, P. A., Abuelo, C., Cameron, E., Valdmanis, V., Mahande, M., Foster, A. & Mabey, D.  The relationship between prevalence of

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

active trachoma, water availability and its use in a Tanzanian village. Trans. R. Soc. Trop. Med. Hyg. 100 (11), 1075–1083. Rabiu, M., Alhassan, M. B., Ejere, H. O. & Evans, J. R.  Environmental sanitary interventions for preventing active trachoma. Cochrane Database Syst. Rev. 2, CD004003. Resnikoff, S., Pascolini, D., Etya’ale, D., Kocur, I., Pararajasegaram, R., Pokharel, G. P. & Mariotti, S. P.  Global data on visual impairment in the year 2002. Bull. World Health Organ. 82 (11), 844–851. Schachter, J., West, S. K., Mabey, D., Dawson, C. R., Bobo, L., Bailey, R., Vitale, S., Quinn, T. C., Sheta, A., Sallam, S., Mkocha, H. & Faal, H.  Azithromycin in control of trachoma. Lancet 354 (9179), 630–635. Speich, B., Marti, H., Ame, S. M., Ali, S. M., Bogoch, I. I., Utzinger, J., Albonico, M. & Keiser, J.  Prevalence of intestinal protozoa infection among school-aged children on Pemba Island, Tanzania, and effect of single-dose albendazole, nitazoxanide and albendazole-nitazoxanide. Parasit. Vectors 6, 3. Stoller, N. E., Gebre, T., Ayele, B., Zerihun, M., Assefa, Y., Habte, D., Zhou, Z., Porco, T. C., Keenan, J. D., House, J. I., Gaynor, B. D., Lietman, T. M. & Emerson, P. M.  Efficacy of latrine promotion on emergence of infection with ocular Chlamydia trachomatis after mass antibiotic treatment: a cluster-randomized trial. Int. Health 3 (2), 75–84. Taye, A., Alemayehu, W., Melese, M., Geyid, A., Mekonnen, Y., Tilahun, D. & Asfaw, T.  Seasonal and altitudinal variations in fly density and their association with the occurrence of trachoma, in the Gurage zone of central Ethiopia. Ann. Trop. Med. Parasitol. 101 (5), 441–448. Taylor-Robinson, D. C., Maayan, N., Soares-Weiser, K., Donegan, S. & Garner, P.  Deworming drugs for soil-transmitted intestinal worms in children: effects on nutritional indicators, haemoglobin and school performance. Cochrane Database Syst. Rev. 11, CD000371. UNESCO  Education For All by 2015: Will We Make It? UNESCO Publishing, Paris; Oxford University Press, Oxford. Vercruysse, J., Behnke, J. M., Albonico, M., Ame, S. M., Angebault, C., Bethony, J. M., Engels, D., Guillard, B., Nguyen, T. V., Kang, G., Kattula, D., Kotze, A. C., McCarthy, J. S., Mekonnen, Z., Montresor, A., Periago, M. V., Sumo, L., Tchuente, L. A., Dang, T. C., Zeynudin, A. & Levecke, B.  Assessment of the anthelmintic efficacy of albendazole in school children in seven countries where soil-transmitted helminths are endemic. PLoS Negl. Trop. Dis. 5 (3), e948. WHO  World Health Organization Model Prescribing Information: Drugs used in Parasitic Diseases, 2nd edn. WHO, Geneva. WHO  Future Approaches to Trachoma Control: Report of a Global Scientific Meeting. Available at: http://whqlibdoc.who. int/hq/1996/WHO_PBL_96.56.pdf (accessed 20 April 2009). WHO  Report of the First Meeting of the WHO Alliance for the Global Elimination of Trachoma. Available at: http:// whqlibdoc.who.int/hq/1997/WHO_PBL_GET_97.1.pdf (accessed 20 June 2008).

130

B. Gelaye et al.

School-based trachoma intervention in Ethiopia

WHO  Report of the Fourth Meeting of the WHO Alliance for the Global Elimination of Trachoma. World Health Organization, Geneva. WHO  Blood Safety and Clinical Technology: 2000–2003 Strategy. Dept. of Blood Safety and Clinical Technology, World Health Organization, Geneva. WHO  The Millennium Development Goals. The Evidence is in: Deworming Helps Meet the Millennium Development Goals. Available at: http://whqlibdoc.who.int/hq/2005/WHO_CDS_ CPE_PVC_2005.12.pdf (accessed 7 November 2013). WHO  A Guide: Trachoma Prevention through School Health Curriculum Development. World Health Organization, Geneva.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

WHO  The Trachoma Grading Cards Slide 1/2 SAFE Documents. Available at: http://www.who.int/blindness/ causes/trachoma_documents/en/index.html (accessed 20 June 2008). Yajima, A., Jouquet, P., Do, T. D., Dang, T. C., Tran, C. D., Orange, D. & Montresor, A.  High latrine coverage is not reducing the prevalence of soil-transmitted helminthiasis in Hoa Binh province, Vietnam. Trans. R. Soc. Trop. Med. Hyg. 103 (3), 237–241. Ziegelbauer, K., Speich, B., Mausezahl, D., Bos, R., Keiser, J. & Utzinger, J.  Effect of sanitation on soil-transmitted helminth infection: systematic review and meta-analysis. PLoS Med. 9 (1), e1001162.

First received 25 March 2013; accepted in revised form 11 October 2013. Available online 12 December 2013

131

04.1

2014

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation: assessing the design, management and functionality of sanitation in poor communities of Dar es Salaam, Tanzania M. W. Jenkins, O. Cumming, B. Scott and S. Cairncross

ABSTRACT This study assessed sanitation access in rapidly expanding informal settlements in Dar es Salaam (Dar) against eight proposed indicators of hygienic safety, sustainability and functionality, and in relation to the Millennium Development Goal (MDG) ‘improved’ sanitation deﬁnition. Information was collected on toilet facility designs, management and functionality through a structured interview and observations at 662 randomly selected residential properties across 35 unplanned, low-income subwards of Dar. Trends in access and associations with sharing, occupancy, latrine replacement, income, education and location factors were considered through statistical analyses. Surveyed subwards were open-defecation free. While 56% of households used a facility that met the MDG improved technology deﬁnition, only 8% had a functional facility that could be considered as hygienically safe and sustainable sanitation. Safe, sustainable, functioning sanitation access was 2.6

M. W. Jenkins (corresponding author) Civil and Environmental Engineering, University of California, Davis, CA, USA E-mail: mwjenkins@ucdavis.edu M. W. Jenkins O. Cumming S. Cairncross Environmental Health Group, Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine B. Scott UK Department for International Development

times greater among the richest quintile than the two poorest quintiles. Very poor sanitation services among Dar’s urban poor arise from widespread lack of access to hygienically safe pit emptying services, unhygienic designs and functionality problems (affecting 67, 55 and 29%, respectively). As new goals and targets beyond 2015 are discussed, these ﬁndings may have important implications for deﬁning what constitutes ‘improved’ sanitation for poor populations living in unplanned informal settlements. Key words

| equity, fecal sludge management, indicators, monitoring and evaluation, pit emptying, shared sanitation

INTRODUCTION Sanitation is an important foundation for health, economic

when they constitute small pockets within enumeration

development and well-being (Bartram & Cairncross ;

areas, are poorly deﬁned, or enumerators are frightened

WSP ). In developing countries, urban areas are esti-

to visit them (Fry et al. ; Hancioglu & Arnold ).

mated to have higher sanitation coverage than rural areas

A further limitation is that access within a city is rarely dis-

(WHO/UNICEF a) but these statistics often mask the

aggregated spatially and economically (Fry et al. ;

severity and complexity of sanitation challenges affecting

UNICEF ).

towns and cities and the urban poor in particular. Sani-

The current generalized approach of deﬁning sani-

tation realities in slums and informal settlements where

tation access as single household use of an ‘improved’

the urban poor live are inevitably diluted in representative

toilet technology (WHO/UNICEF ) may be less appro-

national results of ofﬁcial monitoring surveys or may

priate for rapidly growing cities where on-site sanitation

be under-represented or completely missed, for example

technologies such as pit latrines, pour ﬂush latrines and

doi: 10.2166/washdev.2013.180

132

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

septic tanks continue to be used despite ever-increasing

ﬁlariasis, dysentery and diarrhea in Dar and other Tanza-

urban densities. In such situations, sanitation facilities are

nian cities (UNICEF ). A third of children under ﬁve

likely to be emptied rather than moved (Thye et al. )

years of age in Dar were ill in the four weeks preceding

so that the safety of sanitation systems depends at least as

the 2007 National Household Budget Survey, suffering

much on provisions for safe fecal sludge management

from fever (50%), malaria (40%), diarrhea (12%), in addition

including emptying, removal, treatment and disposal or

to other symptoms of infection (NBS ). More recent

reuse, as on safe fecal capture and containment (i.e. the

data for Tanzania indicate that diarrhea among children

design of the facility) (Barrenberg & Stenström ). Eval-

under ﬁve living in urban areas has increased markedly

uating whether on-site sanitation facilities are functional, a

between 2004 and 2010 from 10 to 18% (two-week preva-

measure of users’ ability to maintain them, is also crucial

lence), surpassing the rate in rural areas (2010 two week

for assessing access.

prevalence of 13%) (UNICEF ). Cholera cases continue

This paper presents a detailed assessment of household

to occur annually in Dar with the last major outbreak of

sanitation facilities in some of the poorest unplanned settle-

over 9,000 cases in 2006, concentrated in wards with the

ments of Dar es Salaam (Dar), Tanzania’s largest city. It

greatest poverty, population densities and percentages of

proposes and applies a set of new indicators to characterize

informal residents (Penrose et al. ).

and assess the hygienic safety and sustained functioning of existing on-plot systems, including locally available pit emptying and disposal (or waste reuse) methods, based on

METHODS

household survey data. New approaches are needed to properly characterize sanitation services in cities where on-plot

An extensive survey undertaken in Dar in June 2008 (the

sanitation continues to be the dominant service model for

start of the dry season) of household sanitation facilities

the urban poor. The study demonstrates this need for new

and associated practices, costs, preferences, perceptions

urban sanitation monitoring approaches and offers one

and plans forms the basis for this study. The survey was

example for how this might be done.

designed to gather consumer insights and a baseline to

Dar’s population was estimated at 2.5 million in 2002 with a growth rate of 4.3% (DCC ). Approximately

develop demand-responsive sanitation upgrading strategies for Dar’s urban poor.

70–80% of residents live in unplanned settlements that lack public infrastructure, basic services, proper roads and

Sampling design

layout, and are often densely populated, resulting in very poor living conditions (UN-Habitat ). Poverty continues

Thirty-ﬁve unplanned, low-income, high-density sub-wards

to be a major problem; 50% of the population live on <$1

(average 379 persons/hectare) were identiﬁed by Dar’s

per day and recent data indicate the urban share of Tanza-

three municipal councils (MC) for upgrading, representing

nians living below the basic needs poverty line has risen

approximately 20% of the unplanned population. Residen-

sharply from 19.6% in 2000 to 25.2% in 2007 (NBS ).

tial properties were chosen as the sampling and analysis

The sewer system reached just 13% of the city’s population

unit, since on-plot sanitation facilities are organized and

in 2004 (DCC ) and 80% of all sanitation facilities in

managed

Dar were pit latrines (NBS ). The extent to which facili-

sampling, following a two-stage process, was used to select

ties are safe and sustainable remains unknown, although the

a representative random sample of 662 residential proper-

Joint Monitoring Programme (JMP) reports that 40% of

ties across the 35 sub-wards (2% of all properties). Each

urban Tanzanians used an improved sanitation facility in

sub-ward (i) was assigned its share (Si) of the sample,

2010, half of whom shared with more than one household

according to its 2002 population. Starting at a random

(WHO/UNICEF a).

10-cell group of houses, a skipping interval was used to

There is a persistently high environmental disease burden

associated

with

cholera,

malaria,

lymphatic

property

level.

Population

proportional

select Si 10-cells in sub-ward i. A single property was randomly selected from each 10-cell list.

133

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indicators of improved and safe and sustainable

Household survey

sanitation The survey consisted of a structured questionnaire administered verbally to the property owner or the oldest tenant in

Table 1 describes eight indicators developed to assess the

the absence of the owner, supplemented by structured spot

safety and sustainability of on-site sanitation systems in

observations of the toilet facility and plot. Following

study communities. The ﬁrst three (1–3) assess the technical

exploratory focus group discussions with owners and

design of the facility, the next two (4 and 5) assess avail-

tenants, the survey was designed to characterize sanitation

ability and access to safe fecal waste management services,

facilities (below and above ground design), age and con-

and the last three (6–8) assess the functionality of the facility

dition;

at the time of the survey.

document

latrine

use,

operating

and

maintenance practices; and to collect past and anticipated

Inclusion of pit lining as a safe and sustainable design indi-

expenditures and plans for construction, replacement,

cator (#3) accounts for two considerations. First, lining is

improvement and pit emptying. The survey also assessed

structurally necessary to prevent pit collapse in areas with

facility design and emptying preferences and perceptions

ﬂooding, poor drainage or high water tables, conditions

of sanitation conditions and problems. At each property,

widely present in poor neighborhoods of Dar (see Table A1

GPS coordinates and respondent socio-economic character-

in the Supplemental Material available online at http://www.

istics were collected. Trained professional surveyors from

iwaponline.com/washdev/004/180.pdf) and frequent charac-

Research International conducted the survey. This study

teristics of undesirable lands where the urban poor reside.

draws on a sub-set of the survey data related to facility

Second, lining also facilitates emptying of full pits while

design, management and functionality to assess the safety

unlined pits, for the most part, cannot be emptied using hygie-

and sustainability of on-site sanitation systems in study

nic mechanized suction methods and risk damage or collapse

settlements.

even when manual means are used (Koné & Strauss ).

Structured observations

use in Dar. Two are considered sanitary services, the vacuum

‘Flooding out’, also known as ‘vomiting’ is a method of par-

hygienically suctioned, transported and disposed of at author-

tial emptying of pits which involves inserting a drain or

ized municipal treatment and disposal sites. Three are unsafe

opening into an exposed or elevated portion of the latrine

methods where fecal sludge is drained into a temporary pit

pit wall, below the slab in order to release fecal sludge

(pit diversion), manually excavated and released or buried

into the open environment to be washed away by storm

on-site (manual emptying), or intentionally released into the

water during rains. In some cases, rising water tables

neighborhood (ﬂooding out). Access to vacuum tanker or

during rain events and excess ﬂooding may increase pit

Vacutug services (indicators 4 and 5) is thus necessary for

sludge levels to the level of the opening, where it is divulged

hygienic emptying and sanitary disposal of pit waste in Dar.

A preliminary review found ﬁve pit-emptying methods in tanker and Vacutug (Thye et al. ), in which fecal sludge is

or ‘vomited’ out. In light of growing concerns over this unsanitary pit-emptying practice, surveyors were trained to look

Analyses

for and record the presence of a ‘ﬂooding out’ pit waste drain pipe during structured spot observation of each facil-

Facility design, waste management and functionality indi-

ity. They also observed the functional state of facilities in

cators were evaluated and combined to measure JMP

terms of slab structural condition, fullness of the waste pit/

improved and safe and sustainable sanitation access across

tank and status of the superstructure. Pit fullness was

the study population. Associations between access and facil-

judged by observing the height of void space between the

ity sharing, occupancy type, replacement latrine, income,

slab or cover and the surface of the sludge. To understand

education and location-related factors were examined.

barriers to safe pit emptying, surveyors observed physical

Using self-reported monthly income, respondent households

accessibility of the property to a small car or tanker vehicle.

were divided into income quintiles, to examine access

134

Table 1

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indicators for classifying on-site sanitation systems as improved (I) (JMP deﬁnition), hygienically safe and sustainable (SS), and functioning (F) in low-income, unplanned communities in Dar es Salaam

Indicator System aspect

Indicator description

type

Deﬁnition and measurement applied in this study

Facility design

1. Pit with slab or better

I, SS

2. Waste contained in pit/tank

I, SS

3. Below ground pit/tank lined

Above-ground technology is basic pit with slab, ventilated improved pit, WC, ceramic bowl (pour ﬂush), but not drum/tyre Technology has waste pit, septic tank, or connects to sewer with no exterior waste drain pipe observed by enumerator (assumed if unable to observe) Below-ground technology is part or fully lined, septic tank, or is connected to sewer, to allow for safe waste emptying, and protect shallow groundwater

4. Hygienic emptying service locally available

5. Plot accessible to hygienic emptying service vehicles

6. Structurally safe to use

7. Pit not completely full of waste

8. Facility has half height walls and half height door or more 8A. (high standard) Facility has roof, full height walls & door

Waste management (emptying, transport, disposal)

Functional condition

Vacuum tanker or Vacutug service to extract pit waste in sealed tanks and dispose into municipal treatment system reported as locally available, or user intends to use service to empty in near future Enumerator observation of plot physical accessibility (car, tanker or tug), or actual use of tanker/tug to empty within last 3 years Enumerator observation that slab/ﬂoor is not collapsing into pit nor in state prohibiting safe use (assumed if unable to observe and not reported) Enumerator observation of pit fullness by measuring depth from slab/top to surface of sludge (assumed if unable to observe and not reported) Enumerator observation of toilet facility superstructure walls, roof, door presence and height Enumerator observation of toilet facility superstructure walls, roof, door presence and height

equity. Tests of association were performed in SPSS v.19

(39%) and tenant-only (absentee landlord) (10%) proper-

using the chi-squared likelihood ratio test for differences in

ties.

proportions and the chi-squared linear-by-linear association

including 2.2 children 3 years, resided at each property.

test of trends in proportions. Differences in group means

Family-owned/occupied properties compared with land-

were tested with the analysis of variance F-statistic. Unless

lord–tenant and tenant-only properties, respectively, had

average,

3.5

households

and

people,

noted otherwise, signiﬁcant results are reported at p 0.05

signiﬁcantly fewer households (2.7 vs. 4.3 and 4.1), non-

for a 2-sided test. Where noted, odds ratios and 95% conﬁ-

child residents (7 vs. 8.6 and 8.3) and total residents

dence intervals were calculated using binary logistic

(9.0 vs. 10.8 and 10.9). They represented 46% of the

regression. Monetary values are adjusted to 2008 values

study population. A communal tap was the most

and presented in US dollars (TSH1,175 ¼ US$1 in 2008).

common main water source (50.4%); just 10.6% had a house tap. Respondents were mostly the property owner (88%) or a tenant (12%). Tenant households paid average

RESULTS AND DISCUSSION

rent of $12/month (range $3–128). Median monthly household income was $106, ranging from $40 (quintile 1,

Sample description

poorest) to $255 (quintile 5, richest). Eighty-two per cent reported monthly income below the Dar 2007 average

A majority of properties (51%) were family-owned and

of $184 while 17% had incomes below the 2007

occupied, with the balance consisting of landlord–tenant

national basic needs poverty line of $12 (NBS ).

135

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Descriptive statistics are provided in Table A1 (this can be

claim of Tanzania’s long-standing urban open defecation

found in the Supplemental Material available online at

rate of 2% (WHO/UNICEF a).

http://www.iwaponline.com/washdev/004/180.pdf). Facility replacement Defecation places At 41% of residences the toilet facility was a replaceTwo residences (0.3%) had no toilet facility while at three

ment for the original facility. Replacements reﬂect re-

(0.5%) the facility was inoperable. Residents at these proper-

investment by owners in a new facility to replace an

ties reporting using a neighbor’s toilet (n ¼ 4) or public

old, full or otherwise no longer functioning latrine,

facility (n ¼ 1). Residents at all other properties used a pri-

while original construction reﬂects population growth

vate toilet facility (99%), sharing it with an average of nine

(new residences). Dar’s high growth rate and sustained

other users (range 1–29) and 3.5 households (range 1–14).

rates of facility replacement since the 1980s are visible

These ﬁndings align with 2007 data indicating 1.1% of Dar

in the breakdown by decade built of the current stock

households had no toilet (NBS ) and 2010 Tanzania

of sampled toilets into original and replacement facilities

JMP data reporting 2% urban open defecation (WHO/

(see Figure A1, which is available online in the Sup-

UNICEF a).

plemental

Material

http://www.iwaponline.com/

washdev/004/180.pdf).

Under-reporting of the use of a neighbor’s toilet seemed likely. At 5.6% of properties, neighbors were cur-

Above and below ground design

rently using the toilet facility (without paying). On the other hand, under-reporting of open defecation seemed unlikely, given that enforcement by environmental health

The above (AG) and below (BG) ground design of each

ofﬁcers of by-laws requiring toilets and prohibiting open

facility (Table 2) and slab material were used to assess

defecation were mentioned frequently during preliminary

facilities in terms of improved and safe and sustainable

focus group discussions. Essentially, poor, unplanned

fecal capture. Simple slab over pit designs (‘traditional pit

areas of Dar appear to be free of open defecation. This

latrine’) were the most common AG design. About 36%

ﬁnding may be the result of many years of campaigns for

of facilities had unlined pits. Most facilities were operated

latrine building and use in Tanzania during the colonial

as wet systems owing to anal cleansing with water and

and Ujamaa periods (UNICEF ) and supports the

Table 2

separate

bathing

place,

posing

increased

fecal

Residential sanitation facility designs in use in low-income, unplanned communities of Dar es Salaam (June 2008)

Above-ground design Shares (% total) Below-ground design

Traditional pit latrine

VIP pit latrine

Ceramic bowl (pour ﬂush)

Drum/tyre latrine

Othera

Unlined pit

35.3

0.6

–

0.2

Partially lined pit

31.9

2.3

1.2

–

Fully lined pit

18.4

4.2

0.5

–

Row total

Drum/tyre lining

0.6

–

1.5

0.2

Septic tank

–

0.3

0.2

–

0.5

Sewer

–

0.5

0.2

–

0.6

Don’t know

1.4

–

0.2

1.5

1.1

Column total

VIP ¼ Ventilated improved pit. a

Includes WC (ﬂush toilet), seat and don’t know.

100

136

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

contamination risks to shallow groundwater (Howard et al.

sanitation facilities in study areas. Similar problems occur

) which is a source of drinking water for 6–23% of Dar

in rural Tanzania (McCubbin ).

residents (NBS ). Water seal facilities (‘ceramic bowl’, ‘WC’) represented about 3% of AG designs, while 1% of BG designs were septic tank and <1% were connected to a sewer. Facility construction expenditures Expenditures for on-plot sanitation facilities, including initial construction and subsequent improvement, were evaluated for facilities built in the 10 years preceding the survey (see Table A2, which is available online in the Supplemental Material at http://www.iwaponline.com/washdev/004/180. pdf). The average reported capital cost per latrine varied little across income quintiles, reflecting a consistent cost for construction of a basic pit latrine ($272). Unimproved drum or tyre latrines required significantly less investment ($112) while water-based, pour-flush and flush technologies required substantially more ($450–500). Facility condition Over 40% of facilities were full or nearly full (<25 cm of full); only 5% were >1 m empty. One in five had a badly cracked or collapsing slab, or was in a state that prohibited safe use. Three-quarters lacked a roof, 19% lacked a door and 6% lacked walls. Owners of facilities in poor structural condition or with inadequate superstructures had significantly lower incomes compared with others. No income

Pit emptying Given the ten-year median age of toilet facilities, it was surprising that 64% had never been emptied. Among facilities that had been emptied, however, nearly half (44%) had been emptied multiple times. While nearly two-thirds reported local availability of a hygienic emptying service, only 23% of emptied latrines used one of the two hygienic methods to empty the last time. Nearly half (43%) of residents reported that flooding out was practiced in their community. Flooding out was the third most frequently reported emptying method, used on 12% of emptied latrines, following pit diversion (59%) and vacuum tanker (18%). While just 4% of respondents admitted to flooding out their latrine, 28% of latrines were observed to have an emptying pipe, suggesting the practice may be more common than reported. Under-reporting likely reflects awareness that the practice is both illegal and considered anti-social, and because it tends to occur in high water table and frequently flooded areas in Dar with little effort by owners, apart from installation of a flood-out pipe (UN-Habitat ). About one in 20 latrines was emptied manually, mainly just the top portion. Average emptying expenditure, including slab repair (a common additional expense incurred), was $57. It varied by method, but differences were not significant.

difference was found between owners with and without completely full pits, suggesting factors beyond ability to

Access to safe and sustainable facilities

pay account for the high rate of full pits. Sanitation facility designs were assessed against indicators Interrupted use

1–3 (Table 1) to estimate safe and sustainable facility access rates (Table 3). While 59% of facilities met the JMP

More than one in six toilet facilities had been unusable at

improved technology deﬁnition, fewer (41%) met our hygie-

some point in the past because it had collapsed, the pit

nically safe and sustainable criteria. In population terms,

was full or improvements were being made (see Table A3,

45% had access to a safe and sustainable sanitation facility.

which is available online in the Supplemental Material at

Pit waste drain pipes (28% prevalence) associated with

http://www.iwaponline.com/washdev/004/180.pdf). Non-

ﬂooding out were a signiﬁcant cause of failure to meet

usable durations averaged 26 days, varying from 10

both improved and safe and sustainable design criteria,

(improvements) to 55 (full pits). Long periods of interrupted

while use of unlined pits further reduced safe and sustain-

use indicate barriers to residents’ ability to maintain their

able facility access.

137

Table 3

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indicator prevalence and rates of access to improved (I), safe and sustainable (SS) and functioning (F) sanitation facilities, waste management and systems in 35 low-income, unplanned areas of Dar es Salaam (June 2008)

Safe, sustainable and Safe and sustainable (SS)

functional (SSF)

Improved (I) Indicator (see Table 1)

Functional (F)

% with

Design

Waste mgmt

Sanit. system

Facility

Sanit. system

1 Pit with slab

82.5

2 Waste contained

72.3

3 Below-ground lined

63.3

4 Hygienic service

67.5

5 Plot accessible

40.0

6 Structurally safe

92.4

7 Not completely full

92.7

78.7

(18.0)

8 Half wall/door (8a Roof, full wall/door) Access, % properties

652

651

662

651

619

608

% Households

% Population

% Children ( 3 yrs)

12.4

N valid properties

13.5

10.5 608

Note: Improved technology and private (JMP access) ¼ 23% (148/652); Improved technology and shared ¼ 36% (237/652); Unimproved (private or shared) ¼ 41% (267/652); Open defecation ¼ 0% of residential properties. For households: improved private ¼ 14% (312/2250), improved shared ¼ 42% (943/2281), unimproved ¼ 44% (995/2250). For unplanned area population: improved private ¼ 20% (1315/6448), improved shared ¼ 42% (2713/6448), unimproved ¼ 38%. a

Only 16% of facilities would be considered ‘functional’ if a higher standard of superstructure functionality for indicator 8 were applied, requiring the cabin to have full height walls, a full

height door and a roof.

Access to safe emptying services

rate of those with safe and sustainable system access compared with those without, adjusting for income (adjusted

More than half of properties reported local availability of at

odds ratio: 4.78; 95% conﬁdence interval (CI): 2.22,

least one of the hygienic emptying services; however, phys-

10.36).

ical plot access by a small car or tanker vehicle was low (Table 3). Consequently, only 33% of the population in

Access to safe and sustainable functioning sanitation

study areas had access to safe emptying services.

systems

Access to safe and sustainable sanitation systems

About 7% of facilities had a partially or completely collapsed ﬂoor and a similar share had a completely full pit

Properties with both a safe and sustainable sanitation facil-

(Table 3). About 80% had walls and a door covering at

ity and access to a safe emptying service (hence, to a safe

least half-height (indicator 8), but just 18% had a complete

and sustainable sanitation system), made up just over 13%

superstructure (indicator 8a). Only 71% met all three basic

of properties and served 14% of the study population

functionality criteria of structural safety, remaining pit

(Table 3). To validate our indicators and classiﬁcation

capacity and minimal superstructure. Sanitation systems

method, we tested whether access to a safe and sustainable

which were functional and safe and sustainable rep-

sanitation system increased the odds of choosing a hygie-

resented just 11% of properties, serving a similar share

nic method to empty a full pit (among those emptied),

of the study population but a smaller share of young

and found a ﬁvefold increase in the hygienic emptying

children (8%).

138

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Shared versus single household access

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

properties in Dar do not appear to be disadvantaged in terms of the hygienic and sustainable standard of facilities

At 61% of residences, toilet facilities were shared by >1

or safe emptying service access, but rather encounter chal-

household, with very high rates of sharing at mixed land-

lenges with maintenance resulting in lower functionality.

lord–tenant

(92%) and tenant-only residences (90%)

compared with family-owned/occupied residences (30%).

Replacement facility and residency plan access

Sharing was positively associated with safe and sustainable facility access and with facility functionality (see Table A4,

Replacement latrines were more likely to have safe and sus-

which is available online in the Supplemental Material at

tainable designs (p ¼ 0.18) than original latrines, and were

http://www.iwaponline.com/washdev/004/180.pdf). Those

signiﬁcantly more likely to have safe emptying service

sharing their facility were about 40% more likely to have a

access (p < 0.001), resulting in a nearly 50% greater rate

functional, safe and sustainable sanitation system than

of safe and sustainable sanitation system access for

those using non-shared facilities (p ¼ 0.17). We hypothesize

replacement over original latrines (p ¼ 0.06). However,

economic factors may explain some of these observations:

replacement latrines were somewhat less likely to be func-

greater cash resources from multiple poor households are

tional mainly because they were more likely to be full than

available for building, maintaining and operating a shared

original latrines. Households may be motivated to replace

facility than from a single poor household for their own

and upgrade their facilities with safe and sustainable designs

facility, and landlords in poor, unplanned areas of Dar are

when safe emptying services are plot-accessible and locally

able to obtain higher rents when they offer better quality

available to empty them.

sanitation facilities to their tenants. The latter hypothesis is

Residency plans were associated with access in several

borne out by our data on reported rents among the small

ways (Table A4). Those uncertain about plans (36%) were

subset of tenant respondents (n ¼ 79), showing an average

least likely to have access to an improved or safe and sustain-

premium of TSH4,500 to 6,100 more in rent per month

able facility but had better access to safe emptying services,

paid by tenants with access to an improved sanitation facil-

compared with properties where the respondent had long-

ity and to a safe, sustainable and functional facility,

term plans; these differences were signiﬁcant. Landlords

respectively. Neither improved nor safe pit-emptying

uncertain about their residency plans may prefer not to

access was associated with sharing.

invest in better sanitation facilities at the property, while

A positive association between sharing and safe and sus-

uncertain renters may prefer not to pay the rental premium

tainable system access runs counter to the JMP assumption

for better sanitation facilities. Deeper dialogue with landlords

that facility sharing diminishes public health safety and ren-

of properties with poor quality facilities is needed to under-

ders sanitation ‘unimproved’. While shared facilities were

stand how landlords’ residency plans and security of tenure

used on average by signiﬁcantly more users (11) than

affect sanitation facility and ongoing maintenance invest-

single household facilities (8.4), the difference was relatively

ment decisions. A need to regulate the types of sanitation

small (2.6 people). Shared facilities were nearly all private,

facilities landlords build and their maintenance is one poten-

on-plot residential facilities (99%) shared among extended

tial implication. It is unclear why those with uncertain plans

family members (25% of sharing) or among non-related

have greater access to safe emptying services. This may have

households residing at the same property (75% of sharing).

to do with security of tenure or other attributes of where they

Tenant-only houses were signiﬁcantly more likely to have

live, given that hygienic emptying service access is largely

access to an improved facility compared with mixed land-

about location relative to main roads and municipal sludge

lord–tenant and family-only houses, but their facilities were

disposal sites. Those with temporary plans (stays <5 years;

signiﬁcantly less likely to be functional (Table A4). Tenant-

5%) had the lowest access to safe emptying services (18%)

only houses also had greater access to safe and sustainable

and to functional facilities (59%), with the result that none

facilities and safe emptying services than family-only proper-

of this group had access to a safe and sustainable functioning

ties, but these differences were not signiﬁcant. Tenant-only

system. Temporary and transient residents in Dar’s

139

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

unplanned communities appear to have the greatest exposure

services and to safe and sustainable facilities, but differences

to poor sanitation conditions.

were not signiﬁcant. Kikondoni MC had a signiﬁcantly lower rate of safe and sustainable functioning sanitation system

Household income and education effects on access

access (6.4%) than the other two MCs (11 and 14%), traceable directly to signiﬁcantly lower safe emptying service

Nearly all components of safe, sustainable and functional

access (23%) in Kikondoni compared with the other MCs

access showed signiﬁcant income effects (Table A4). Conse-

(40%). These results suggest signiﬁcant spatial disparities in

quently, access to a safe and functionally sustainable

the quality of sanitation services and conditions among

sanitation system was 2.6 times greater among Q5 house-

Dar’s low-income, unplanned sub-wards associated with dis-

holds (19%) compared with those in the two poorest (Q1

parities in access to hygienically safe and sustainable facilities

and 2) quintiles (7%). Improved facility access, facility func-

and hygienically safe emptying services, affected in part by

tionality, and safe and sustainable system access increased

geo-hydrologic, topographic and municipal council location.

signiﬁcantly with income quintile. Income effects were less clear on access to a safe and sustainable facility: Q1 and Q5

Trends in safe and sustainable functioning access

had similarly high safe and sustainable facility access as a result of higher rates of pit lining and absence of drain

The proportion of still-standing facilities built in each decade

pipes compared with the middle quintiles. Income had a lim-

with an unhygienic waste pit drain pipe for ﬂooding out rose

ited effect on safe emptying service access, plausible if safe

from 15% in the 1950/60s to 32% in the 2000s (linear trend

service access depends mainly on public investments in locat-

p ¼ 0.002), an alarming trend for environmental health and

ing and operating municipal fecal sludge waste disposal and

safety. At the same time, a signiﬁcant upward trend was

treatment facilities, and on upgrading policy actions to

observed of building facilities with hygienic below-ground

improve slum plot access and road networks.

lining (38% in 1950/60s, to 71% in 2000s, linear trend p ¼

Education was not associated with increasing access to

0.001). These two opposing trends have mixed implications

improved or safe and sustainable facilities nor with function-

for safe emptying and disposal of fecal waste in study com-

ality rates, but signiﬁcantly and strongly correlated with

munities. Year-by-year construction trends suggest a very

increasing access to safe emptying services when controlling

small increase in improved technology and safe and sustain-

for income, resulting in a small but signiﬁcant positive trend

able sanitation system access of 1.3% and 0.7% per year

in access to a safe and sustainable functioning sanitation

(R 2 ¼ 0.14 and 0.20). The slow, incremental improvement

system with education. Better educated households may

in the hygienic safety and sustainability of on-plot sanitation

have better information about safe emptying services, may

systems being built in low-income, unplanned settlements of

value more highly the beneﬁts of using safe emptying ser-

Dar is attributed mainly to greater private investment in

vices, or choose to live at plots with vehicle access or in

cement slabs and lined pit designs, but is greatly held back

neighborhoods where safe emptying services operate.

by increasing rates of installation of ﬂood out drain pipes and lack of improvement in access to safe emptying services.

Access by location

Very constrained gains are consistent with JMP-documented progress of about 1% increase per year in urban access to

Safe and sustainable sanitation system access was signiﬁ-

improved sanitation technology from 1990 to 2010 in Tanza-

cantly higher where ﬂooding was not frequent (30%), the

nia (WHO/UNICEF a).

water table was not high (41%), and the terrain was sandy (16%), at rates 1.6 to 2.1 times greater than areas with frequent ﬂooding, high water table or non-sandy terrain

CONCLUSIONS

(Table A4). Functionality was not associated with any tested location characteristics. Hilly areas (13%) were less

This study examined sanitation access and equity in Dar es

likely than non-hilly areas to have access to safe emptying

Salaam looking beyond the current MDG deﬁnition of

140

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

‘improved’ to assess the hygienic safety and sustainability of

those used by just one household in unplanned areas of

sanitation systems in the low-income, unplanned settle-

Dar were more likely to be hygienically safe and functionally

ments where 70–80% of Dar’s population resides. While it

sustainable, most likely because greater resources can be

might be argued that hygienic safety and sustainability of

mobilized from sharing households to invest in a higher

sanitation facilities and pit emptying are less important in

quality of sanitation service.

low density rural areas, they appear to be critical in low-

Lack of hygienic pit emptying services and plot inaccessi-

income, urban informal settlements where population den-

bility in low-income, unplanned areas of Dar leave many poor

sities are rapidly increasing.

households who invest in better quality latrine facilities with

We developed and applied a relatively straightforward

few good options for preserving the value of their investment,

household survey methodology to collect information on

leading an increasing number to install a ﬂood-out drain pipe

three critical factors: the sanitation facility design, its man-

and resort to unhygienic pit-emptying methods. On the other

agement

the

hand, where residents in unplanned poor areas have access to

information to assess ﬁve indicators of system hygienic

a hygienic emptying service they appear more willing to

safety and sustainability and three indicators of system func-

upgrade to a safe and sustainable sanitation facility design

tionality. The approach involved a small number of survey

that enables the use of these services. This suggests that

questions, accompanied by a systematic inspection of the

public investments to create widespread access to hygienic

sanitation facility and property, which together we estimate

emptying services, currently available to only 33% of poor

would add about 10 minutes per household on average to

residents in the study areas, could be an effective approach

the conventional set of JMP water, sanitation and hygiene

to stimulate greater private investment in and access to safe

survey questions. Surveyors would need to be trained to con-

and sustainable sanitation. As the current MDGs approach

duct the inspection accurately but would not require special

their expiry and new goals and targets are discussed

skills. In our study, we successfully trained consumer market

(WHO/UNICEF b), these ﬁndings may have important

and

functional

state.

then

used

research enumerators in the technical aspects of the

implications for deﬁning what constitutes ‘adequate’ sani-

inspection.

tation in urban areas, particularly for poor populations

As the sector looks to post-2015 goals and targets calling

living in high-density informal settlements.

for progressive improvement in the quality of sanitation services to address safe fecal sludge management (WHO/ UNICEF b), our methodology offers a starting point to

ACKNOWLEDGMENTS

consider in the post-2015 enhancement of the current JMP survey approach, going beyond a simple characterization

This research was made possible with UK aid from the

of technology type, to examine additional factors impacting

Department of International Development (DFID) as part

on the quality and viability of sanitation access.

of the SHARE research program (www.SHAREresearch.

Applying these new, more rigorous indicators, we found access to adequate sanitation facilities in Dar’s unplanned

org). However, the views expressed do not necessarily reﬂect the Department’s ofﬁcial policies.

areas to be far lower than current JMP estimates. We also found signiﬁcant income-associated, educational and spatial disparities in adequate access across low-income, unplanned settlements. Among Dar’s informal population, those in the two poorest quintiles, residents with less education, temporary and transient residents, those living in areas with difﬁcult hydro-geologic conditions, and in Kikondoni Municipal Council, were consistently the most vulnerable to poor sanitation conditions. On-plot sanitation facilities shared by multiple low-income households compared with

REFERENCES Barrenberg, E. & Stenström, T.  Sanitation Safety Plans (SSP): A Vehicle for Guideline Implementation, Concept Note. Water, Sanitation, Hygiene and Health Unit, World Health Organization, Geneva. Bartram, J. & Cairncross, S.  Forgotten foundations of health. PLoS Med. 7 (11), e1000367. Dar es Salaam City Council (DCC)  Dar es Salaam City Proﬁle. Document prepared with assistance of the Cities and

141

M. W. Jenkins et al.

Beyond ‘improved’ towards ‘safe and sustainable’ urban sanitation

Health Programme, WHO Centre for Development, Kobe, Japan, Nov. 2004. Available at www.dcc.go.tz (accessed 6 November 2013). Fry, S., Cousins, W. & Olivola, K.  Health of children living in urban slums in Asia and the Near East: Review of existing literature and data. Activity Report 109, Environmental Health Project. Report prepared for EHP Project 26568/ Other.ANE.Startup, Washington DC. Available at http://pdf. usaid.gov/pdf_docs/PNACQ101.pdf (accessed 30 September 2012). Hancioglu, A. & Arnold, F.  Measuring coverage in MNCH: tracking progress in health for women and children using DHS and MICS household surveys. PLoS Med. 10 (5), e1001391. Howard, G., Jahnel, J., Frimmel, F. H., McChesney, D., Reed, B., Schijven, J. & Braun-Howland, E.  Human excreta and sanitation: control and protection. In: Protecting Groundwater for Health: Managing the Quality of Drinking-water Sources (O. Schmoll, G. Howard, J. Chilton & I. Chorus, eds). World Health Organization, IWA Publishing, London. Koné, D. & Strauss, M.  Low-cost options for treating faecal sludges (FS) developing countries – challenges and performance. In: Proceedings 6th International Conference Waste Stabilization Ponds and 9th International Conference Wetland Systems (A. Liénard & H. Burnett eds), Avignon, France, 27 September–1 October 2004 pp. 213–219. McCubbin, C.  PhD Thesis, London School of Hygiene & Tropical Medicine. NBS  Household Budget Survey 2007, Final Report, 6 May 2009. National Bureau of Statistics, United Republic of Tanzania.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Penrose, K., Caldads de Castro, M., Werema, J. & Ryan, E. T.  Informal urban settlements and cholera risk in Dar es Salaam, Tanzania. PLoS Negl. Trop. Dis. 4 (3), e631. Thye, Y. P., Templeton, M. R. & Ali, M.  A critical review of technologies for pit latrine emptying in developing countries. Crit. Rev. Env. Sci. Tech. 41 (20), 1793–1819. UN-Habitat  The Sustainable Cities Programme in Tanzania 1992–2003. The SCP Documentation Series, Volume 2. UNHabitat, Nairobi, Kenya. UNICEF  Cities and Children: The Challenge of Urbanization in Tanzania. UNICEF, Tanzania. WHO/UNICEF a Estimates of the use of Improved Sanitation Facilities, United Republic of Tanzania. Joint Monitoring Programme for Water and Sanitation, Updated March 2012. World Health Organization, Geneva. WHO/UNICEF b Proposal for Consolidated Drinking Water, Sanitation and Hygiene Targets, Indicators, and Definitions. Joint Monitoring Program for Water and Sanitation, December 2012. World Health Organization, Geneva. Available at: http://www.wssinfo.org/post-2015monitoring/overview/ (accessed 23 May 2013). WHO/UNICEF  Types of Drinking-water Sources and Sanitation. Joint Monitoring Program for Water and Sanitation, World Health Organization, Geneva. Available at: http://www.wssinfo.org/definitions-methods/watsancategories/ (accessed 6 November 2013). WSP  Economics of Sanitation Initiative. Water and Sanitation Program, World Bank, Washington, DC. Available at: http://www.wsp.org/sites/wsp.org/files/publications/ WSP-ESI-Flier.pdf (accessed 6 November 2013).

First received 13 June 2013; accepted in revised form 18 October 2013. Available online 13 December 2013

142

04.1

2014

Planning for water and sanitation in the residential suburbs: the case of Aligarh City, India Abha Lakshmi Singh and Md. Zahir Abbas

ABSTRACT This paper examines the status of water and sanitation conditions in the residential suburbs of Aligarh city. The study is mainly based on primary sources of data collected through ﬁeld/household surveys with the help of questionnaire interviews. Data was collected from 996 households living in 66 developed residential colonies. The services were categorized into three levels (standards), level 1 (basic), level 2 (intermediate) and level 3 (higher), to assess their standard. The standard score

Abha Lakshmi Singh (corresponding author) Md. Zahir Abbas Department of Geography, Aligarh Muslim University, Aligarh, 202002, Uttar Pradesh, India E-mail: abhalakshmisingh@yahoo.com

additive model has been used to arrive at a Composite Mean Standard Index (CMSI). The average value of CMSI of water and sanitation were calculated and further categorized into four levels, high, moderate, low and very low, for the purpose of planning. Maps were prepared accordingly. Results have shown that colonies located along the main roads have a high level of water and sanitation facilities compared with the colonies located along the wedges of two roads. Key words

| levels, residential colonies, sanitation, suburbs, water

INTRODUCTION Managing urban growth has become one of the most impor-

of suburbs was the slow growth of cities. Any small increase

tant challenges of the 21st century (Cohen ). India is

in the population (the population of Aligarh city grew from

urbanizing rapidly and this trend is certain to gather pace

669,087 in 2001 to 872,500 in 2011; Census of India ,

in the next 20–30 years. It is anticipated that by 2030, 350

; Aligarh Nagar Nigam ) of a city was generally

million people will have moved from the countryside to

absorbed within the existing residential areas. It was only

urban centres and by 2050, 700 million people will have

with the ﬂow of new migrants into the city, that the city’s resi-

done so (Sabikhi ). The last two censuses (Census of

dential areas could no longer absorb the growth, and the city

India , ) show that there has been a steady rise in

began to expand physically, ﬁrst through the development of

the number of new census towns (towns that have a popu-

vacant land within the city itself, later by the slow encroach-

lation of over 5,000) over the years and in due course

ment of land in areas lying outside the city limits. The

many of these towns will grow in size and become cities.

uncontrolled and unplanned expansion in the spatial extent

There are no plans for these new settlements as well as for

of the cities creates residential hubs (concentration of resi-

the existing cities that are witnessing massive expansion.

dential areas) where many residential colonies develop

The population outside the cities has grown faster indicating

without any basic amenities and facilities. In India these

a strong tendency for outward expansion of urban areas over

are basically poor areas inhabited by poor, low skilled

their fringes, engulﬁng villages and farmlands and trans-

migrant workers, who cannot afford the higher cost of

forming them into dense commercial, industrial and

living in the city. Presently this trend is changing. Nowadays,

residential areas resulting in the development of suburbs

the land at edges not only provides housing but is also being

(Angel et al. ).

converted into property for commercial and industrial fac-

The occurrence of suburbs around Indian cities is a

tories and for middle and high income residential housing

rather recent phenomenon. The main reason for the absence

(Singh & Abbas ). Primary urban facilities, such as

doi: 10.2166/washdev.2013.013

143

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

water supply and sewerage, are for the most part not available

suggest suitable measures for providing a livable environ-

because the city provides these services to places within the

ment in the residential suburbs.

municipal limits; outside the municipal limits in the residential suburbs, lack of necessary administrative infrastructure and lack of ﬁnances mean that these services are not provided by the municipality (Ramachandran ). In order to

DATABASE AND METHODOLOGY

understand the needs of people and provide a livable environment in the residential suburbs, in-depth surveys have to be

The present study is mainly based on primary sources of

conducted to ﬁnd a solution.

data which has been collected through an in-depth survey

Keeping these aspects in mind, Aligarh (27 530 N latitude W

of the residential suburbs and households living in the devel-

and 78 4 E longitude), a medium-sized city located about

oped residential colonies. The following methods were used:

133 km away from the country’s capital, New Delhi, in the

1. The suburban residential colonies were identiﬁed with

western part of the state of the Uttar Pradesh, in the fertile

the help of satellite imagery IRS-P6, March 2007 and

Gangetic plain of North India, was selected as the study

extensive ﬁeld surveys which were conducted during

area (Figure 1). In this paper, an attempt has been made

2009 and 2010. Ninety-four colonies were identiﬁed, 66

to assess the state of water supply and sanitation and to

developed (colonies which were sold off by the

Figure 1

Aligarh city: Residential colonies in the suburbs (2009–2010). Source: IRS-P6, satellite imagery, March 2007; Based on ﬁeld survey, 2009–2010.

144

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

colonizers and were occupied by people) and 28 undeve-

(basic), level 2 (intermediate) and level 3 (higher) – to

loped (colonies in which only plots have been

show the present quality of the services. The choice of

demarcated, no or little construction work has been

the level of a particular service is inﬂuenced by affordabil-

started and no one lives there; Table 1 and Figure 1).

ity as well as need. The quality of service decreases from

2. To gather relevant information about the state of water

level – 1.

supply and sanitation in the residential colonies, a ques-

4. The standard score additive model has been used to

tionnaire was prepared with the help of questionnaires

arrive at a composite standard index (Smith ; Bishnoi

used in similar studies (MaGranahan ; Mitra ;

& Aneja ). Standard scores have been used to

Sharma & Sharma ; Rahman ; Singh &

develop a composite score availability of water supply

Rahman ; Khan ). For in-depth investigation,

and sanitation services in the residential suburbs. The

35 developed residential colonies (>50% of the devel-

standard score measures the departure of individual

oped residential colonies) were selected and from each

observations from the maximum and minimum values

colony up to 29 households were sampled. The total

of all observations in a comparable form.

sample size consisted of 996 households. 3. To examine the level (standard) of services, the data col-

Si ¼ ðXi minÞ=ðmax minÞ

(1)

lected regarding the state of water supply and sanitation has been categorized into three service levels – level 1

where Si ¼ standard index of the ith variable, Xi ¼ original value of the individual observation, min ¼ minimum value

Table 1

Overall levels of state of water supply in the residential suburbs of Aligarh city (2009–2010)

of the variable, max ¼ maximum value of the variable. The values of standard scores are added and their average is taken to obtain the composite mean standard index

Overall levels of water supply conditiona Name of the roads/bypasses

Level 1

Level 2

Level 3

and algebraically expressed as: CMSI ¼

(2)

Si=N

1. Anupshahar Road

–

16.67

83.33

(1). Anupshahar Road to Ramghat Road bypass

8.33

64.44

27.22

2. Ramghat Road

0.78

5.99

93.23

(2). Ramghat Road to GT Road (Kanpur) bypass

2.22

34.45

63.33

3. GT Road (Kanpur)

–

4.88

95.12

water supply and sanitation services in the residential

4. Agra Road (3). Mathura Road to Khair Road bypass

where CMSI ¼ composite mean standard index, ∑Si ¼ sum of standard index of variable j in observation i, N ¼ number of variables. 5. For demarcation and mapping of different levels of

3.33 13.33

24.17

72.50

colonies for the purpose of planning, a total of eight

21.67

indicators under two subheads were taken into

5. Khair Road

1.85

24.07

74.07

consideration: (i) For the state of water supply, two indicators (avail-

(4). Khair Road to GT Road (Delhi) bypass

3.33

37.78

58.89

ability of submersible/tubewells inside the premises,

6. GT Road (Delhi)

2.50

20.83

76.67

(ii) For the state of sanitation, six indicators (ﬂush

(5). GT Road (Delhi) to Anupshahar Road bypass

4.13

43.12

52.75

latrines inside the house; disposal of excreta from

Total sampled households 996

2.76

26.15

71.09

good quality of water supply).

Note: aAverages (in per cent) of two important factors considered for water supply condition under each level. Level 1 – roadside handpumps, poor quality of water supply. Level 2 – handpumps inside premises, average quality of water supply. Level 3 – submersible/tubewells inside premises, good quality of water supply. Source: Based on ﬁeld survey, 2009–2010.

ﬂush latrines into soak pits; pucca nali (cemented drains) in the premises; disposal of wastewater/ waterlogging in open plots/ﬁelds; disposal of household/commercial/industrial

waste

collection

points; and frequency of garbage collection by private agency twice a week) were considered.

145

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The average values of CMSI of water supply and sani-

and its periphery. It is a very difﬁcult task to improve the natu-

tation were calculated and further categorized into four

ral drainage. Water remains stagnant in the open drains

levels for the purpose of planning – high, moderate, low

because of the absence of an outlet. So the city and its periph-

and very low levels – with the help of mean and standard

eral areas not only become submerged during the rainy season

deviation methods (standard grading system) and were

but also remain submerged all year round because of poor

mapped with the help of GIS Arc view 3.1.

natural drainage. The waterlogged areas become breeding grounds for mosquitoes and other pests. Therefore, the city and its peripheral areas need proper laying and management

RESULTS AND DISCUSSION

of a drainage system to avoid overﬂowing, ﬂooding and waterlogging (Singh & Rahman ).

Background of the study area

Satellite imagery (IRS-P6) and ﬁeld surveys helped in identifying 94 suburban residential colonies. Of these, 66

Cities get their water supply from surface sources (rivers,

were developed and 28 undeveloped. Out of the 66 devel-

lakes, ponds) and from groundwater sources. Aligarh city

oped colonies, 39 (60%) were located along the main

depends on groundwater sources for its water. The water

roads and 27 (40%) along the wedges (connecting two

from the two rivers, Ganga (in the north-east) and Yamuna

main roads). Of the total developed colonies, most were

(in the north-west), which form the highland peripheries

located along Ramghat Road (13), GT Road (Kanpur, 11)

and ﬂow 50 60 km away from the city, cannot be tapped

and Khair Road (5). While along other roads such as GT

(owing to the saucepan-shaped topography of the city and

Road (Delhi, 4), Agra Road (4) and Anupshahar Road (2)

district). The other two rivers of some importance are the

there were only a few developed colonies. Along the

Kali and Sengar (downstream) but both are seasonal rivers

bypasses most of the developed colonies were located

and dry up during the summer season, and the city drains

along GT Road (Delhi) to Anupshahar Road (7), Anupsha-

(carrying wastewater) open into the Kali Nadi (river), so

har Road to Ramghat Road (6), Khair Road to GT Road

the water is not potable. Therefore, the city and its peripheral

(Delhi) (6) and Ramghat Road to GT Road (Kanpur) (6),

areas fully depend upon the groundwater. The alluvial terrain

while Mathura Road to Khair Road bypass had only 2 devel-

of the Ganga plain forms one of the richest water-bearing

oped colonies (Figure 1).

formations. The Pleistocene and recent deposits which are composed of sands and gravel constitute the aquifers. These

Status of water supply in the residential suburbs

plains are immense reservoirs of fresh, sweet water stored in more porous, coarser strata, beneath the level of satur-

During ﬁeld surveys it was observed that the suburbs were

ation, and they are easily accessible by means of ordinary

dotted by handpumps and submersible tubewells which

boring. The underground aquifers are supplemented by rain

were the main source of water supply in these areas. Hand-

water which sinks easily into the ground. The percolation

pumps were routinely installed, wherever they were

from major rivers and their tributaries also contributes to

required inside the homes of original dwellers in the core vil-

maintaining groundwater level. The physical setting there-

lages and by the migrants from the nearby villages who had

fore provides every facility for the boring of tubewells

settled in the poorer areas of the residential suburbs. So the

(Azam ).

original dwellers either manage to live without most services

Topographically, Aligarh district represents a shallow

or try to have the basic/lower order services. They obtain

trough, in a saucepan shape, with the city lying in the centre

water from ordinary wells and handpumps while the new

of the low-lying tract. Because of the saucepan-shaped topo-

suburban residents have better options, i.e. submersible/tube-

graphy, not only does the dirty water from the city collect

wells inside their premises, for obtaining water. Colonizers

here but so does water from peripheral areas, and there is

and private developers at times provide better water facilities

no outlet. The poor natural drainage is responsible for the

so that their ﬂats/apartments are sold off at better prices. The

peculiar drainage and sewerage problems faced by the city

submersible/tubewells were installed by the rich farmers in

146

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

the core villages and by the colonizers and ﬂat/apartment

soak pits, heaps of garbage, waterlogging and marshy con-

owners. Regarding the status of water supply, the residents

ditions so the seepage of contaminants takes place easily

reported that the supply was regular and they were getting

and hence the water becomes contaminated.

an adequate amount of water. This was because they were getting the supply either from handpumps or from submers-

Status of sanitation in the residential suburbs

ible/tubewells from which water is available 24 hours. The state of water supply in the residential suburbs was

During ﬁeld surveys it was observed that the sanitation con-

assessed and categorized according to its standard or levels.

dition in the residential suburbs was very poor. It was seen

Two important factors (i) source of water supply (level 1,

that people were defecating in the open and few had

roadside handpumps; level 2, handpumps inside premises;

either ﬂush or manual latrines in their courtyards or inside

and level 3, submersible/tubewells inside premises) and (ii)

their homes. The excreta was disposed either in drains/

quality of water supply (level 1, poor; level 2, average; and

along roads/in ﬁelds or in soak pits. It was observed that

level 3, good) were taken into consideration. Table 1

the residential suburbs had very poor drainage conditions.

shows the overall levels of state of water supply along the

Wastewater was disposed either in kutcha nali (uncemented

main roads and bypasses in the residential suburbs of Ali-

drains) or in a pucca nali which opened around the house or

garh city. The averages of the two important factors were

onto the roads/streets and/or in open plots/ﬁelds. House-

taken to assess the overall water supply at various levels in

hold/commercial/industrial solid waste was disposed of

the residential suburbs. Table 1 shows that in most of the

either along road/street sides or in open ﬁelds and at collec-

cases the residential suburbs of Aligarh city have a higher

tion

level of water supply (i.e. residents depended upon submers-

themselves for disposal of garbage. At most of the places gar-

ible/tubewells inside premises and reported good quality of

bage remained uncollected.

points.

The

residents

reported

hiring

persons

water) especially along the main routes (i.e. GT Road

The state of sanitation conditions in the residential sub-

(Kanpur), Ramghat Road, Anupshahar Road, GT Road

urbs has been assessed by examining the different levels of

(Delhi), Khair Road and Agra Road). Poor level of water

its availability. Six important factors (i) toilet facilities

supply (i.e. residents depended on handpumps as a source

(level 1 open defecation, level 2 ﬂush/manual latrines in

of water and reported poor quality of water) existed along

the courtyard and level 3 ﬂush latrines inside the house);

the Anupshahar Road to Ramghat Road bypass and

(ii) disposal of excreta (level 1 no disposal, level 2 disposal

Mathura Road to Khair Road bypass.

of excreta from manual latrines in drains/along roads/in

This area (located along the Anupshahar Road to Ram-

ﬁelds and level 3 disposal of excreta from ﬂush latrines

ghat Road bypass and Mathura Road to Khair Road bypass)

into soak pits); (iii) type of drainage (level 1 no drainage,

was dominated by poor Muslim migrants belonging to the

level 2 kutcha nali and level 3 pucca nali); (iv) disposal of

lower socio-economic strata; they were mostly involved in

wastewater/waterlogging (level 1 around the house, level 2

informal sectors as petty shopkeepers, petty shop owners,

on the roads/streets and level 3 in open plots/ﬁelds); (v) dis-

rickshaw pullers, labourers etc. and had large families (>6

posal of household/commercial/industrial waste (level 1

members) so they could only afford shallow bored hand-

disposal on road/street sides, level 2 disposal in open

pumps only 6–8 metres deep. The upper stratum of the

ﬁelds, level 3 disposal in collection points); and (vi) fre-

earth absorbs the wastewater from garbage dumps, collection

quency of garbage collection by private agencies (level 1

points and soak pits and thus pollutes the groundwater. Boil-

uncollected, level 2 collected fortnightly, level 3 collected

ing water for drinking purposes is not practised in these

twice a week) were taken into consideration. Table 2

households because of poverty, illiteracy and also because

shows the overall levels of sanitation along the main roads

of the popular perception that the overall quality of water is

and bypasses in the residential suburbs of Aligarh city. The

tolerable. Field observations have also revealed that the sur-

averages of the six important factors of sanitation were

roundings of these sources (handpumps) are generally very

taken to assess the overall sanitation at various levels in

unclean. Handpumps have been installed in areas with

the residential suburbs. Table 2 shows that in most of the

147

A. L. Singh & M. Z. Abbas

Table 2

Planning for water and sanitation in the residential suburbs

Overall levels of state of sanitation in the residential suburbs of Aligarh city (2009–2010)

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

wastewater/waterlogging around the house, disposal of waste on road/street sides, uncollected garbage) existed in

Overall levels of sanitation conditiona

the suburbs along Mathura Road to Khair Road bypass and Anupshahar Road to Ramghat Road bypass. These two bypasses are occupied by the Muslim migrants belong-

Name of the roads/bypasses

Level 1

Level 2

Level 3

1. Anupshahar Road

10.56

35.55

53.89

ing to low socio-economic strata with large family size (>6

(1). Anupshahar Road to Ramghat Road bypass

42.96

37.78

19.26

members) and are engaged in informal sectors characterized

2. Ramghat Road

14.84

25.35

59.81

(2). Ramghat Road to GT Road (Kanpur) bypass

23.52

26.48

3. GT Road (Kanpur)

16.36

24.29

59.35

4. Agra Road

16.67

73.33

(3). Mathura Road to Khair Road bypass

61.11

13.89

5. Khair Road

26.34

29.83

43.83

plan, for which (1) ﬁeld/household surveys have been con-

(4). Khair Road to GT Road (Delhi) bypass

32.22

30.37

37.41

ducted to collect information regarding the state of water

6. GT Road (Delhi)

16.11

15.83

68.06

(5). GT Road (Delhi) to Anupshahar Road bypass

32.57

27.83

39.60

Total sampled households 996

23.91

26.69

49.40

by insecurity of work, low wages, long working hours, etc.

Planning for water and sanitation in the residential suburbs The proper planning for water supply and sanitation in the residential suburbs requires the preparation of a detailed

Note: aAverages (in per cent) of six important factors considered for sanitation condition under each level. Level 1 – open defecation, no disposal of excreta, drainage, disposal of wastewater/waterlogging around the house, disposal of household/commercial/industrial waste on road/ street sides, garbage uncollected. Level 2 – ﬂush/manual latrines in the courtyard, disposal of excreta from manual latrines in drains/along roads/in ﬁelds, kutcha nali, disposal of wastewater/waterlogging on the

and sanitation (Tables 1 and 2); (2) on this basis, the residential suburbs could be demarcated and mapped showing the different areas with different levels of state of water and sanitation services; and (3) then plans could be prepared for different areas according to their requirements. The provision of services, especially water and sanitation, plays an important role in the processes of growth and development of the residential suburbs. The residential suburbs with better provision of water and sanitation will be more developed than

roads/streets, disposal of household/commercial/industrial waste in open ﬁelds, garbage

other areas. People will be attracted to settle in these suburbs

collected fortnightly. Level 3 – ﬂush latrines inside the house, disposal of excreta from ﬂush latrines into soak

and this area will be a hub of activities. An attempt has been

pits, pucca nali, disposal of wastewater/waterlogging in open plots/ﬁelds, disposal of

made to assess the levels of water and sanitation in the resi-

household/commercial/industrial waste in collection points, garbage collected twice a week. Source: Based on ﬁeld survey, 2009–2010.

dential suburbs with the help of eight indicators: 1. Two indicators of water supply condition: X1 – availability of submersible/tubewells inside the premises

cases the residential suburbs of Aligarh city have high to low levels of sanitation. A good (level 3) state of sanitation (i.e. ﬂush latrines inside the house, disposal of excreta from

X2 – good quality of water supply 2. Six indicators of sanitation condition: X3 – ﬂush latrines inside the house

ﬂush latrines into soak pits, pucca nali, disposal of waste-

X4 – disposal of excreta from ﬂush latrine into soak pits

water/waterlogging in open plots/ﬁelds, disposal of waste

X5 – pucca nali in the premises

in collection points, garbage collected by private agencies

X6 – disposal of wastewater/waterlogging in open plots/ﬁelds

fortnightly/twice a week) prevails in the residential suburbs

X7 – disposal of household/commercial/industrial waste

along the main roads, i.e. Agra Road and GT Road (Delhi), Ramghat Road, GT Road (Kanpur) and Anupshahar Road. A poor (level 1) state of sanitation (no toilet facility in the house/open defecation practised, no disposal of excreta, no

drainage

facilities

the

premises,

disposal

in collection points X8 – frequency of garbage collection by private agency twice a week. The development index (DI) of water supply and sanitation was calculated with the help of the standard index

148

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

method taken of the respective indicators. Finally, the com-

supply (with the help of two indicators) and sanitation

posite mean standard index (CMSI) was calculated through

(with the help of six indicators) were calculated and categorized into four levels – high, moderate, low and very low level

the standard index (SI)/aggregate score model. Road-wise levels of development of water supply and sani-

of development – with the help of mean and standard devi-

tation services in the residential suburbs of Aligarh city are

ation methods (standard grading system) (Tables 5 and 6)

presented in Tables 3, 4 and 5. A maximum value of CMSI of

and were mapped (Figure 2).

water supply was observed along GT Road (Kanpur) (0.993),

Field surveys have also revealed the same picture in all

followed by Ramghat Road (0.973), Anupshahar Road (0.870)

the residential colonies. Tables 5 and 6 and Figure 2 reveals

and GT Road (Delhi) (0.800), and minimum values were

that 18 residential colonies located along Agra Road (0.856),

observed along Mathura Road (0.000), Mathura Road to

GT Road (Delhi) (0.847), GT Road (Kanpur) (0.821), Ram-

Khair Road bypass (0.226) and Anupshahar Road to Ramghat

ghat Road (0.814) and Anupshahar Road (0.729) had high

Road bypass (0.284). A maximum value of CMSI of sanitation

levels of development of water supply and sanitation services

was observed along Agra Road (0.955) followed by GT Road

(>0.709). Whereas, 13 residential colonies located along

(Delhi) (0.893) and minimum values of CMSI of sanitation

Ramghat Road to GT Road (Kanpur) bypass (0.654), Khair

were observed along Mathura Road (0.000), Mathura Road to

Road (0.626), Khair Road to GT Road (Delhi) bypass

Khair Road bypass (0.158) and Anupshahar Road to Ramghat

(0.509) and GT Road (Delhi) to Anupshahar Road bypass

Road bypass (0.227). In each case of water supply and sani-

(0.492) had moderate levels of development of water

tation services the value of CMSI was at a minimum along

supply and sanitation services (0.709–0.424). Four residen-

Mathura Road because along this road there were no developed

tial colonies located along Anupshahar Road to Ramghat

residential colonies. Only demarcation of plots could be seen.

Road bypass (0.256) and Mathura Road to Khair Road

For the purpose of demarcating the residential suburban

bypass (0.192) had low levels of water supply and sanitation

areas on the basis of different levels of water supply and

services (0.424–0.139). A few residential colonies located

sanitation services, the average value of CMSI of water

along Ramghat Road (Lucky Compound, Railway and

Table 3

Road-wise levels of development of water supply in residential suburbs of Aligarh city (2009–2010) (Based on Standard Index)

Two indicators of water supply conditions Availability of submersible/

Good quality of

CMSI of water

Name of the roads/bypasses

tubewells inside the premises

water supply

supply conditions

1. Anupshahar Road

0.801

0.939

0.870

(1). Anupshahar Road to Ramghat Road bypass

0.000

0.568

0.284

2. Ramghat Road

0.947

1.000

0.973

(2). Ramghat Road to GT Road (Kanpur) bypass

0.441

0.881

0.661

3. GT Road (Kanpur)

1.000

0.986

0.993

4. Agra Road

0.644

0.870

0.757

(3). Mathura Road to Khair Road bypass

0.000

0.452

0.226

5. Khair Road

0.593

0.953

0.773

(4). Khair Road to GT Road (Delhi) bypass

0.302

0.927

0.615

6. GT Road (Delhi)

0.662

0.939

0.800

(5). GT Road (Delhi) to Anupshahar Road bypass

0.288

0.814

0.551

7. Mathura Road

0.000

Note: CMSI – Composite Mean Standard Index; R – Rank.

149

Table 4

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Road-wise levels of development of sanitation in residential suburbs of Aligarh city (2009–2010) (Based on Standard Index)

Six indicators of sanitation conditions Disposal of

Frequency of

Disposal of

household/

garbage

Pucca nali in the

wastewater/ waterlogging in

commercial/ industrial waste in

collection by private agency

CMSI of sanitation

into soak pits

premises

open plots/ﬁelds

collection points

twice a week

conditions

0.904

0.944

0.830

0.851

0.000

0.588

0.186

0.127

0.865

0.000

0.227

2. Ramghat Road

1.000

1.010

0.935

0.977

0.000

0.654

(2). Ramghat Road to GT Road (Kanpur) bypass

0.626

0.653

0.623

0.653

0.667

0.648

3. GT Road (Kanpur)

0.948

0.972

1.000

0.973

0.000

0.649

Flush

Disposal of

latrines inside the

excreta from ﬂush latrine

Name of the roads/bypasses

house

1. Anupshahar Road (1). Anupshahar Road to Ramghat Road bypass

4. Agra Road

0.870

0.909

0.951

1.000

0.955

(3). Mathura Road to Khair Road bypass

0.174

0.175

0.173

0.426

0.000

0.158

5. Khair Road

0.657

0.699

0.807

0.709

0.000

0.479

(4). Khair Road to GT Road (Delhi) bypass

0.580

0.583

0.819

0.440

0.000

0.404

6. GT Road (Delhi)

0.835

0.839

0.917

0.766

1.000

0.893

(5). GT Road (Delhi) to Anupshahar Road bypass

0.651

0.654

0.638

0.656

0.000

0.433

7. Mathura Road

0.000

CMSI – Composite Mean Standard Index; R – Rank. Source: Based on ﬁeld survey, 2009–2010.

Table 5

Road-wise overall levels of development of water supply and sanitation services in residential suburbs of Aligarh city (2009–2010) (Based on Standard Index)

CMSI of water supply and R

Name of the roads/bypasses

Water supply conditions

Sanitation conditions

sanitation services

1. Anupshahar Road

0.870

0.588

0.729

(1). Anupshahar Road to Ramghat Road bypass

0.284

0.227

0.256

2. Ramghat Road

0.973

0.654

0.814

(2). Ramghat Road to GT Road (Kanpur) bypass

0.661

0.648

0.654

3. GT Road (Kanpur)

0.993

0.649

0.821

4. Agra Road

0.757

0.955

0.856

(3). Mathura Road to Khair Road bypass

0.226

0.158

0.192

5. Khair Road

0.773

0.479

0.626

(4). Khair Road to GT Road (Delhi) bypass

0.615

0.404

0.509

6. GT Road (Delhi)

0.800

0.893

0.847

(5). GT Road (Delhi) to Anupshahar Road bypass

0.551

0.433

0.492

7. Mathura Road

0.000

CMSI – Composite Mean Standard Index; R – Rank. Source: Based on ﬁeld survey, 2009–2010.

150

Table 6

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Assessment of levels of development of water supply and sanitation services in residential suburbs of Aligarh city (2009–2010) for the purpose of future planning

Overall levels of urban basic services

Location and names of the colonies located in the residential suburbs

1. High level of development (>0.709) 18 residential colonies (51.43%)

1. Anupshahar Road (1 residential colony) (Sagar Housing Complex) 2. Ramghat Road (7 residential colonies) (Raj Vihar, Chaudhary Vihar, Lucky Compound, Shankar Vihar, Mukund Vihar, Railway, Harizon Basti colony) 3. GT Road (Kanpur) (6 residential colonies) (Vishnu Dham, Basant Vihar, Prabhat Nagar, Shiva, Keshav Vihar, Hydel colony) 4. Agra Road (2 residential colonies) (Damodar Apartment, Neem Karoli colony) 5. GT Road (Delhi) (2 residential colonies) (Nai, Guru Ramdas Nagar)

2. Moderate level of development (0.709–0.424) 13 residential colonies (37.14%)

1. Ramghat Road to GT Road (Kanpur) bypass (3 residential colonies) (Engineer’s, Kuldeep Vihar, Brij Vihar colony) 2. Khair Road (3 residential colonies) (Shiv Dham, Nada Bajitpur Nayaabadi, Singh Vihar) 3. Khair Road to GT Road (Delhi) bypass (3 residential colonies) (Ashok Vihar, Baghel, Bank colony) 4. GT Road (Delhi) to Anupshahar Road bypass (4 residential colonies) (Dev Nagar, Ganesh Nagar, Riyaz, Islam Nagar colony)

3. Low level of development (0.424–0.139) 4 residential colonies (11.43%)

1. Anupshahar Road to Ramghat Road bypass (3 residential colonies) (Nishad Bagh, Bihari, Raza Nagar) 2. Mathura Road to Khair Road bypass (1 residential colony) (Sadbhavana Nagar)

4. Very low level of development (<0.139) No developed residential colony

1. Mathura Road

Source: Based on ﬁeld survey, 2009–2010.

Shankar Vihar), along GT Road (Delhi) (Guru Ramdas

It was observed that the level of development of water

Nagar, Sai Vihar), along Anupshahar Road (Sagar Housing

supply and sanitation was good in the suburban residential

Complex), along GT Road (Kanpur) (Vishnu Dham, Basant

colonies developed along the main roads such as Agra

Vihar, Prabhat Nagar, Vidyut Nagar), and along Agra Road

Road, GT Road (Delhi), GT Road (Kanpur), Ramghat

(Damodar Apartment, Neem Karoli) have well developed

Road and Anupshahar Road. This is because the land

water supply and sanitation either provided by the colonizers

prices are much higher here and only the bigger colonizers

or arranged by the residents themselves. Residents living in

purchase these lands, sell it off at higher prices and provide

these colonies belong to the high income strata so they

all the urban basic services. People who purchase these

have equipped their homes with all the modern urban ser-

lands are economically prosperous, so they also develop

vices. Colonies located along the city margins such as

basic services according to their needs. Living near the

along Anupshahar Road to Ramghat Road bypass are an

main roads means an easy access to transport, marketing,

extension of the Jeewangarh colony which is mainly occu-

health education facilities and so on. Availability of urban

pied by poor Muslim migrants who are engaged in

basic services in the developed colonies located along the

informal sectors, have large family size (>6 members) and

wedges of the main roads was low to very low because

belong to lower socio-economic strata. The same situation

these wedges have been garbage and wastewater dumping

prevails in the extended colonies located along the GT

grounds. So this area has not developed. Another reason

Road (Delhi) to Anupshahar Road bypass and Mathura

which emerged during the ﬁeld investigation was that

Road to Khair Road bypass. So the residents living in these

these were poor areas, occupied by people belonging to

colonies could not cope with these problems by themselves.

lower socio-economic strata. The colonizers did not provide

151

Figure 2

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Aligarh city: Assessment of levels of water supply and sanitation services in residential suburbs for the purpose of planning (2009–2010). Source: Based on ﬁeld survey, 2009– 2010.

the basic facilities and the residents did not have sufﬁcient

in population and land area) and sell it off at higher prices

money to develop the facilities by themselves.

to the colony developers; they in turn will or will not provide the urban basic services. So there is a need for better urban governance. The municipal authority should pay attention

CONCLUSIONS

and consider the suburbs as being within city limits and in turn should collect taxes for their convenience to provide

The main conclusion which derived from the present study is

all the basic services. So the residents living in the residential

that colonies located along the main roads have a high level

suburbs get equal advantages of all services.

of water and sanitation facilities compared with the colonies

A number of action programmes have been suggested

located along the wedges of two roads, and the services are

which should be implemented in a comprehensive manner:

mostly arranged by the residents themselves. But city populations will keep on growing, the urban expansion will keep on increasing and the suburban residential colonies will

•

Managing residential suburbs requires local solutions rather than state, national ones. There is an urgent need

also keep on developing. So what can be done? The coloni-

to build and support the capacity of local/municipal

zers will purchase the agricultural lands at low prices from

bodies to manage the problems of rapidly growing resi-

the farmers in anticipation of the city’s growth (i.e. increase

dential suburbs.

152

•

A. L. Singh & M. Z. Abbas

Planning for water and sanitation in the residential suburbs

Local municipal authorities need to be empowered with ﬁnancial and human resources to develop the required infrastructure and deliver the services in the residential

•

suburbs. Large investment is needed for the creation, expansion and maintenance of water supplies, systems for collection and disposal of waste, both garbage and wastewater, elec-

•

tricity, roads, houses etc. People living in the residential suburbs should pay taxes so that they are also provided with these services. These services must reach all sections of the people living in

•

the residential suburbs. Policies should be framed so that the colonizers and land developers develop the area properly and provide all the urban basic services in the individual houses and apart-

•

ments before selling them for residential purposes. Every city should have a master plan, not only for the city but also for the suburban areas, and planning should not be entrusted to the state town planning department but to the city town planning department; plans should be implemented properly.

REFERENCES Aligarh Nagar Nigam (Aligarh Municipal Corporation)  Sewa Bhawan, Aligarh. Angel, S., Sheppard, S. C. & Civco, D. L.  The Dynamics of Global Urban Expansion. Transport and Urban Development, World Bank, Washington, DC.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Azam, S. F.  Irrigation and Agricultural Development. B.R. Publishing Corporation, New Delhi. Bishnoi, N. K. & Aneja, R.  Regional variation of socioeconomic development in Haryana: a district level analysis. Indian Journal of Regional Science 40 (2), 26–40. Census of India  Registrar General and Census Commissioner, New Delhi, India. Census of India  Registrar General and Census Commissioner, New Delhi, India. Cohen, B.  Urbanization in Developing Countries: Current Trends, Future Projections and Key Challenges for Sustainability. Technology in Society 28 (1–2), 63–80. Khan, N.  Water Supply and Sanitation Conditions in Aligarh City. PhD Thesis, AMU, Aligarh. MaGranahan, G.  Household environmental problems in low income cities: an overviews of problems and prospects for improvement. Habitat International 7 (2), 105–121. Mitra, A.  Status of basic services in Indian cities. In: Urbanisation in Developing Countries (B. Mohanty, ed.). Concept Publishing Company, New Delhi. Rahman, A.  Household Environment and Health. B.R. Publishing Corporation, New Delhi. Ramachandran, R.  Urbanization and Urban System in India. Oxford University Press, New Delhi. Sabikhi, R.  Designing the future cities. Hindustan Times, 5 September 2012, New Delhi, p. 8. Sharma, P. M. & Sharma, V. K.  Urban water supply. In: Water Supply Problems of Alwar City (N. L. Gupta, ed.). Rawat Publications, Jaipur and New Delhi. Singh, A. L. & Abbas, M. Z.  Planning for urban basic services in the residential suburbs of Aligarh City. Transactions IIG 33 (2), 199–210. Singh, A. L. & Rahman, A.  Drainage problems in Aligarh City. The Indian Geographical Journal 73 (1), 35–40. Smith, D. M.  The Geography of Social Well-Being in the United States. McgrawHill Book Company, New York, pp. 85–90.

First received 18 November 2012; accepted in revised form 14 September 2013. Available online 12 December 2013

153

Short Communication

04.1

2014

Short Communication Prefabricated biogas reactor-based systems for community wastewater and organic waste treatment in developing regions S. Cheng, Z. Li, H.-P. Mang, X. Liu and F. Yin

ABSTRACT Population densities in cities of developing countries are much higher than in other parts of the world, and the predominance of poverty in urban slums is ubiquitous throughout the region. In many urban areas, the lack of wastewater and waste management continues to be a huge challenge for environment and health protection. Decentralized approaches are proposed to provide practical, alternative options for sustainable urban wastewater and waste management in urban conditions. Conventionally, on-site constructed brick/concrete biogas reactors are the most used models. However, long construction periods, quality issues and leakage of biogas are often the disadvantages

Z. Li (corresponding author) S. Cheng H.-P. Mang X. Liu F. Yin School of Civil & Environmental Engineering, University of Science and Technology Beijing, Xueyuan Road 30, Haidian District, 100083, Beijing, China E-mail: zifulee@aliyun.com

of construction design. In contrast to these systems, prefabricated biogas reactors can be produced off-site from different kinds of material. In this paper, prefabricated biogas reactor and treatment systems will be discussed, which could be applied in different developing countries. Meanwhile, some existing cases in China, Indonesia and South Africa are presented to show clear scenarios. Key words

| DEWATS, prefabricated biogas reactor, slum, wastewater and waste management

INTRODUCTION At the turn of the century, the Millennium Development

urbanization is attributed to natural increase and rural–

Goals (MDGs) unveiled a special horizon – one that the

urban migration. The rate of urbanization often exceeds

entire developing world has been tasked to arrive at by

the rate of economic growth. As a result, infrastructure

2015. The MDGs become the parade-horse of the inter-

development lags behind the population growth, particu-

national community with respect to poverty eradication

larly in the low-income areas (Konteh ). Population

(ADB/UNDP/UNESCAP/WHO ). MDG 7 requires

increase in most cities is projected to be highest in slums

achieving a signiﬁcant improvement in the lives of at least

(Paterson et al. ). A third of the urban population in

100 million slum dwellers by 2020 by reducing poverty

the developing countries is estimated to be living in slums,

and improving water and sanitation. Up to now, although

with the highest number in sub-Saharan Africa, where

the world has met the MDG drinking water target, ﬁve

62% of the urban population lives in slums (UN-Habitat

years ahead of schedule, there are no reasons to relax. Pro-

, ).

jections indicate that in 2015 more than 600 million people

Providing reliable and affordable wastewater and

worldwide will still be using unimproved water sources (UN

waste treatment systems is a challenge in many parts of

).

the world. Poverty, high density, the absence of utilities

Rapid urbanization in developing countries creates

and lack of infrastructure are common features of

massive demand for basic infrastructure in cities. The

urban slums. The environment of slums is extremely

doi: 10.2166/washdev.2013.135

154

S. Cheng et al.

Prefabricated systems for waste and wastewater treatment

unhygienic as they are located at neglected sites such as solid waste dumps, open drains and sewers, low

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

APPROPRIATE TECHNOLOGIES FOR DECENTRALIZED SYSTEMS

land, embankments, and often along the rail lines (UN-Habitat ). Due to the high population density

Slums create their own simple wastewater management tech-

and overcrowding in slums, the potential spread of

niques to improve their living conditions. Examples of simple

diseases is usually much greater than in less crowded

techniques are grease traps, ﬁltering systems, vegetative sys-

neighborhoods.

tems, trash removal and natural circulation systems.

To develop a sustainable urban wastewater and waste

DEWATS are ideal for treating domestic wastewater of

management platform for this situation, a decentralized

small cities, towns and townships as well as suburbs of big

approach for waste and wastewater treatment for biogas

cities without sewage pipeline networks. Selection of appro-

production and resource recovery is proposed as a practi-

priate technologies depends on a number of factors: for

cal option. Centralized wastewater treatment systems

example, population density in the targeted area; economic

depend on sophisticated technologies and skilled man-

level and activities; work performance of local authorities

power for their operation and maintenance (O&M). They

in township, village or community; terrain; geology features;

require huge investment capital and recovery mechanisms

climate; and last but not least the wastewater discharge

of signiﬁcant O&M costs (Giovanni et al. ). Therefore,

requirements. DEWATS are based on a modular technical

decentralized wastewater treatment systems (DEWATS)

conﬁguration concept. Appropriate combinations of treat-

could be beneﬁcial in developing countries, allowing

ment modules can be selected, depending on the required

locals to improve their situation when there is a lack of

treatment efﬁciency, costs, land availability and other rel-

action or capacity of the governing body (Green & Ho

evant issues for speciﬁc locations. Typical technologies for

).

DEWATS are presented in Table 1.

Table 1

Typical technologies for DEWATS

Technology

Description

Biogas septic tank (BST)

The simple septic tank system is the most commonly known primary treatment method for on-site wastewater treatment. Septic tanks remove most settleable solids and – if gas-tight sealed – function as an anaerobic bioreactor that promotes partial digestion of organic matter and provides energy in the form of biogas. The Imhoff tank model is another primary treatment method that can accommodate higher ﬂow rates than the septic tank, but it is less common (May et al. ).

Anaerobic bafﬂed reactors (ABR)

ABR is modified from the simple septic tank system and could – if gas-tight sealed – be used for biogas recovery. In 1981, the ABR was developed by McCarty to treat wastewater with extremely high organic loads (William & David ). A model developed in Nepal consists of ABR and hybrid constructed wetland (HCW). ABR has good potential for use as primary treatment. It is very effective in the removal of organic parameters (Shirish et al. ) and could be filled with carrier materials similar to an anaerobic filter (AF) to provide highly efficient contaminant removal (Feng et al. ).

Membrane biological reactor (MBR)

MBR combines the features of membrane separation and biological treatment technology. It can be regarded as an advanced treatment technology. MBR is not an easy-to-maintain technology and is not ideal for most rural, peri-urban and slum conditions. MBR mainly suits areas where a high efﬂuent standard is required (Adriano et al. ).

Constructed wetland (CW)

CW makes full use of a triple synergy among chemical, physical and biological effects of natural ecosystems. Wastewater is puriﬁed when passing through a constructed basin ﬁlled with coarse sand and gravel, and planted with water-tolerant plants like reeds, cattails and other hygrophilous plants, thus imitating natural marshland conditions. CW does not treat very polluted wastewater. It normally functions as a post-treatment unit.

Sand ﬁltration system

Known as natural treatment systems, sand ﬁltration (mostly buried) systems are regarded as an environmentally friendly ﬁnal treatment step, which is cost effective, and free of problems from solid waste generation – compared with conventional systems (Kemal & Bilal ).

155

S. Cheng et al.

Prefabricated systems for waste and wastewater treatment

JUSTFICATIONS FOR PREFABRICATED SYSTEMS AND COMPONENTS

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

construction cost or extended program overheads due

•

to repeatedly required training sessions. Where residential areas are modiﬁed and rebuilt as a

Traditional wastewater and waste treatment systems are

result of rural reconstruction and land reform measures,

usually built with bricks and reinforced concrete, materials

or inheritance, which affects the permanent siting of con-

which are widely available and low cost. Movable and cor-

ventional reactors.

roding parts should be avoided, if possible by gravity flow. Low investment and operation costs, owner involvement for cost saving, convenient maintenance, biogas production and sludge recycling are the main features of the traditional process. However, long construction time, quality issues and leakage of biogas are often the disadvantages of the construction design. Prefabricated modules were introduced in recent years. Prefabricated systems have some obvious advantages over the traditional ones: (1) the final quality is better due to industrial style quality control; (2) they bear sufficient mechanical strengths with good air tightness and a long service-life; (3) under normal conditions a good insulation maintains a relatively stable temperature inside the reactor; (4) the light weight facilitates easy transport; and (5) as the systems and/or modules are manufactured offsite, the installation period is short. Prefabricated systems and components are suitable for the following situations:

•

• •

CASE 1: CHINA China plays a leading role in the sector of low-cost biogas reactors as well as the prefabricated biogas reactor industry. In China, prefabricated reactors are also called commercialized reactors. The initial development of prefabricated reactors aimed at treating animal or human manure. Since the 1980s, China has developed many kinds of commercialized or semi-commercialized household biogas reactors, such as the PE (polyethylene) reactor, semi-hard plastic reactor, iron reactor, GRP or FRP (glass-ﬁber reinforced plastics) reactor, GRC (glass-ﬁber reinforced composite) reactor, ABS (acrylonitrile butadiene styrene) reactor, ferro-/bamboo cement reactor, and the wire mesh cement prefabricated reactor (Zhao ; Liang et al. ). Since 2000, prefabricated reactors have entered the real commer-

High ground water levels: in coastal areas where it is dif-

cial stage, and a number of manufacturers have started

ﬁcult to construct brick, stone, concrete or molded

businesses. Currently, prefabricated reactors are available

reactors; where the quality of reactor construction

as FRP reactors, soft material reactors and plastic material

cannot be controlled but gas- and water-tightness

reactors (CAREI ). Typical prefabricated reactors are

should be absolutely guaranteed.

presented in Figure 1 and Table 2.

Locations in remote/mountainous areas: where it is difﬁ-

As for wastewater treatment, China launched a series of

cult to provide and transport conventional construction

plans and regulations to control wastewater discharge, one

materials.

of which explicitly stipulated that dispersed settlements construction

without access to municipal sewage collection systems

materials and specialized labor force: to avoid increased

should adopt an on-site wastewater treatment approach to

Site

Figure 1

with

shortages

conventional

Typical prefabricated reactors in China (from left to right: FRP, soft material and plastic material).

156

Table 2

S. Cheng et al.

Prefabricated systems for waste and wastewater treatment

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Classiﬁcation of prefabricated reactors by China Association of Rural Energy Industry (CAREI)

Type

Chosen material

Description

FRP reactor

FRP (glass-ﬁber reinforced plastics)

Representative for prefabricated reactors

Soft reactor

For example: PVC (polyvinyl chloride), PAMM (polymethyl methacrylate), LDPE (low-density polyethylene), PE (polyethylene)

Also called bag reactors in other countries, because it looks like a big soft bag

Plastic reactor

For example: hard PVC, ABS (acrylonitrile butadiene styrene), PP (polypropylene), HDPE (high-density polyethylene), LLDPE (linear low density polyethylene), DCPD (dicyclopentadiene)

A hard reactor in contrast to soft reactor

meet the national discharge standard. Hence, prefabricated

until 2014 (Ekasanti ; Miller ). Prefabricated

wastewater treatment facilities entered the market (see

DEWATS in Indonesia is shown in Figure 3 and convention-

Figure 2). For such a system (Figure 2 left), micro-power is

al and prefabricated wastewater treatment facilities are

installed between 0.04 and 0.075 kW, serving a population

compared in Table 3.

of less than 30, while for larger populations up to 1,500, installed power is 2.88 kW (Figure 2 right). Removal efﬁciency of COD, SS and TN of such a system can reach 90,

CASE 3: SOUTH AFRICA

85 and 60%, respectively. On 10 August 2011 the ﬁrst two prefabricated DEWATS modules were installed in the Frasers informal settlement

CASE 2: INDONESIA

of eThekwini Municipality, South Africa. They are treating domestic wastewater from a community ablution block ser-

Indonesia is one of the leading countries with a large

ving 75 households. On average each household consists of

number of DEWATS. In 2010 the Indonesian government

5 to 6 people. As most of the existing scattered pit latrines do

launched a special budget line from the national budget

not work anymore and the public sewer line ends 2 km away

through the Settlement Sanitation Development Accelera-

from the settlement, the local responsible department pro-

tion Program (PPSP), which encourages local governments

vided residents with communal ablutions blocks, i.e.

to implement a community-based sanitation (CBS) like ser-

standard container with toilets, urinals, showers and

vice package in a larger scale of up to 5000 units in over

basins on ﬁve different sites. The wastewater is treated in

300 cities until 2014. The population served by DEWATS

decentralized septic tanks. No efﬂuent goes to any water

systems is already larger than the population served by cen-

body. The treated wastewater ﬂows into evapotranspiration

tralized systems in Indonesia and will increase exponentially

areas. Three of the ﬁve treatment systems are designed

Figure 2

Low-power prefabricated wastewater treatment equipment in China.

157

S. Cheng et al.

Figure 3

Table 3

Prefabricated systems for waste and wastewater treatment

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Prefabricated DEWATS in Indonesia.

Comparison of implementing wastewater treatment facilities between conventional and prefabricated with FRP in Indonesia (Ekasanti 2012)

Conventional

Site selection

Prefabricated

Fixed building

• •

Can be added/removed; easy to meet changing demand and user numbers High mobility, ﬂexible, can be relocated

Site condition

Complicated construction method when meeting difﬁcult soil conditions

Easy to install in any soil condition

Quality control

Need high skills to supervise the implementation on site to keep the quality required

Quality control can be done off site in the workshop during production to meet the standard

Construction period

70–90 days construction

3–7 days installation

Maintenance

Difﬁcult to ﬁnd the leakages and costly to repair

Easy to repair

according to BORDA’s DEWAT systems, while for the ﬁrst

In addition to the improvements in the informal settle-

time two different prefabricated concrete modules were

ment, the old and unhygienic toilets at the local primary

installed,

local

school with 330 learners will be replaced by robust prefabri-

company. Transportation and installation of prefabricated

cated toilets. The wastewater treatment system will include a

packages are shown in Figure 4.

biogas reactor, which provides the school kitchen with

Figure 4

manufactured

and

assembled

Transportation and installation of prefabricated packages in South Africa.

158

S. Cheng et al.

Prefabricated systems for waste and wastewater treatment

energy for cooking. This helps to partly substitute the currently used liqueﬁed petroleum gas (LPG).

CONCLUSIONS The concept of decentralized technologies for wastewater and waste management aims at the development of wastewater systems that are ﬁnancially more affordable, more socially responsible and more environmentally benign than conventional centralized systems. Decentralization is a promising approach for wastewater and waste management in urban slums. The technology of DEWATS integrates a package of anaerobic treatment and natural treatment systems. This paper discusses prefabricated biogas reactor-based systems for wastewater and waste treatment, which could replace traditional, on-site constructed brick and concrete biogas reactors. Prefabricated biogas reactors possess obvious advantages over traditional reactors, such as easy maintenance, short construction period, good air tightness and easy transportation. China leads the world in prefabricated biogas reactors, for which available materials are FRP, soft material and plastic material; prefabricated DEWAT systems were tested and disseminated by BORDA in several countries such as South Africa and Indonesia; they have proved to show great potential in developing regions.

REFERENCES ADB, UNDP, UNESCAP and WHO  Asia water watch: are countries in Asia on track to meet target 10 of the millennium development goals? 2015. Adriano, B., Michele, S., Martin, A., Soren, N. B., Lorenzen, J., Avi, S., Anders, D., Anita, F., Secondo, G., Wolfram, K., Finn, P. & Mathias, N. A.  Decentralised water and wastewater treatment technologies to produce functional water for irrigation. Agric. Water Manage. 98, 385–402.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

China Association of Rural Energy Industry (CAREI)  Development Report of Chinese Commercial Household Biogas Digester. (in Chinese). Ekasanti, P.  Prefabrication of DEWATS in Indonesia with RFP. IWA Conference on DEWATS, Nagpur, India, 2012. Feng, H., Hua, L., Qaisar, M., Qiu, C., Fang, C. & Shen, D.  Anaerobic domestic wastewater treatment with bamboo carrier anaerobic baffled reactor. Int. Biodeter. Biodegr. 62, 232–238. Giovanni, L., Annamaria, V. G. & Francesco, A.  To centralise or to decentralise: An overview of the most recent trends in wastewater treatment management. J. Environ. Manage. 94, 61–68. Green, W. & Ho, G.  Small scale sanitation technologies. Water. Sci. Technol. 51 (10), 29–38. Kemal, G. & Bilal, T.  A serially connected sand filtration and constructed wetland system for small community wastewater treatment. Ecol. Eng. 35, 1208–1215. Konteh, F. H.  Urban sanitation and health in the developing world: reminiscing the nineteenth century industrial nations. Health & Place 15 (1), 69–78. Liang, Z., Guo, J., Liang, L. & Tao, B.  Application and discussion of red mud low-pressure biogas reactor (In Chinese). China Biogas. 15 (3), 26–27. May, A. M., Akram, T. & Joumana, A. N.  Decentralized approaches to wastewater treatment and management: Applicability in developing countries. J. Environ. Manage. 90, 652–659. Miller, A.  Sustainable wastewater management operation and maintenance management models. IWA World Water Congress, Busan, Korea, 2012. Paterson, C., Mara, D. & Curtis, T.  Pro-poor sanitation technologies. Geoforum. 38, 901–907. Shirish, S., Raimund, H., Otto, M., Roshan, R. S., Prajwal, S. & Rajendra, S.  Performance of an anaerobic baffled reactor and hybrid constructed wetland treating high-strength wastewater in Nepal: a model for DEWATS. Ecol. Eng. 35, 654–660. UN  The Millennium Development Goals Report. 2012. UN-Habitat  The Challenges of Slums: Global Report on Human Settlements. Earthscan, London. UN-Habitat  Planning Sustainable Cities: Global Report on Human Settlements. Earthscan, London. UN-Habitat  State of the World’s Cities 2010/2011: Bridging the Urban Divide. Earthscan, London. William, P. B. & David, C. S.  The use of the anaerobic baffled reactor (ABR) for wastewater treatment: a review. Water Res. 33 (7), 1559–1578. Zhao, Y.  Development and prospect of rural household reactor industrialization (In Chinese). Energy of China. 6, 46–48.

First received 13 February 2013; accepted in revised form 22 July 2013. Available online 21 October 2013

159

Short Communication

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Short Communication Development of Haiti’s rural water, sanitation and hygiene workforce Brian Hubbard, Gabriella Lockhart, Richard J. Gelting and Fabienne Bertrand

ABSTRACT In 2009 the Haitian Directorate of Potable Water and Sanitation (DINEPA) identiﬁed an inadequately trained and under-staffed rural workforce as one of their main institutional challenges. Plans to address this challenge were impacted by the devastating earthquake of January 12, 2010 and the cholera outbreak of October 2010, both of which further complicated Haiti’s already poor water and sanitation conditions. Recognizing the importance of DINEPA’s institutional priorities, donor and technical assistance groups provided needed support to improve the country’s conditions and build the rural water and sanitation workforce. This report describes how DINEPA and the US Centers for Disease Control and Prevention (CDC) collaborated to design and implement a training program for 264 potable water and sanitation technicians for rural areas. The paper also describes the initial ﬁeld activities of the newly trained technicians and the immediate impact of their work in the rural water, sanitation and hygiene sector. Key words

| cholera, Haiti, sanitation and hygiene, water quality, water, workforce development

Brian Hubbard (corresponding Author) Environmental Health Services Branch, National Center for Environmental Health, United States Centers for Disease Control and Prevention, Department of Health and Human Services, 4770 Buford Highway, NE., MS F-58, Atlanta, GA 30341, USA E-mail: bnh5@cdc.gov Gabriella Lockhart Environmental Health Services Branch, National Center for Environmental Health, United States Centers for Disease Control and Prevention, Department of Health and Human Services, 33 Pond Avenue, Unit 711, Brookline, MA 02445, USA Richard J. Gelting Health Systems Reconstruction Team, Emergency Response and Recovery Branch, Division for Global Health Protection, Center for Global Health, United States Centers for Disease Control and Prevention, Department of Health and Human Services, 1600 Clifton Rd NE, MS A-05, Atlanta, GA 30329, USA Fabienne Bertrand National Directorate of Potable Water and Sanitation, 4, Angle Rue Métellus et Route Ibo Lélé, Pétion-Ville, Haiti

INTRODUCTION: WATER, SANITATION AND HYGIENE CHALLENGES IN HAITI In 2009 the Haitian Directorate of Potable Water and Sani-

). Achieving this goal was complicated by the devastating

tation

Haitian

earthquake of January 12, 2010 and the cholera epidemic that

Government to reform, regulate and develop the country’s

(DINEPA)

was

established

the

began in October 2010. Prior to the earthquake Haiti already

water, sanitation and hygiene (WASH) sector (DINEPA

had the lowest water and sanitation coverage in the Western Hemisphere and was the only country in the world to have experienced lower sanitation coverage between 1995 and

2010 for reasons other than population decline (WHO/ UNICEF ). By September 8, 2013, there were 675,216 cases and 8,261 deaths (MSPP ) in Haiti related to

160

B. Hubbard et al.

Haiti’s WASH workforce

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

cholera, making it the largest cholera epidemic for any country on record (PAHO ). DINEPA worked to address the critical water and sani-

Box 1 |

The Role of DINEPA’s Potable Water and Sanitation Technicians for the

Communes

tation situation throughout the country, but did not have the human resource capacity to do this effectively (DINEPA

To support the regional and departmental water and

). On average, there were only seven DINEPA staff

sanitation ofﬁces, the TEPACs’ activities include:

members stationed in each of the country’s 10 departmental rural units (URDs), and they were located in each department’s urban center (DINEPA ). DINEPA needed a trained workforce to be present in every commune in the rural areas to ensure the safety of drinking water, increase access to improved sanitation facilities, and promote

•

Working to inventory and assess rural drinking

•

Monitoring the risks that affect those drinking water

•

hygiene (communes are third-level administrative subdivisions in Haiti; the 10 departments have 42 districts that are further divided into 140 communes). To address their

•

institutional challenges and respond to the on-going cholera epidemic, DINEPA requested technical assistance from the US Centers for Disease Control and Prevention (CDC) to

•

build their rural water and sanitation workforce.

In 2011 DINEPA and CDC began developing the Potable

systems. Improving management of systems and sources, and strengthening communication and reporting of results. Identifying key WASH actors in the communes, and building collaborative relationships with these individuals and organizations. Mobilizing local communities and promoting communication

between

community

water

and

sanitation committees and DINEPA’s Departmental

• MATERIALS AND METHODS: ESTABLISHING A RURAL WATER, SANITATION AND HYGIENE WORKFORCE

water systems.

Rural Units (URDs). Educating community members in the principles and techniques of household water treatment and safe storage, safe sanitation, and proper hygiene

•

practices. Serving as DINEPA’s point of contact in the communes during emergency response events.

Water and Sanitation Technicians for the Communes (TEPAC) Program, and created its vision: a workforce that would dramatically expand DINEPA’s scope and effective-

Each module contained PowerPoint presentations, stu-

ness nationally (the TEPAC program was funded by the

dent and instructor guides, and suggested examination

Inter-American Development Bank, the Spanish Agency for

questions to assess participants’ understanding of the

International Development Cooperation, the World Bank,

material. The training materials also included lab exercises,

and the Centers for Disease Control and Prevention). Placing

activities for ﬁeld visits to community water systems, and a

two TEPACs in each of the country’s 133 rural communes

ﬁnal exam (which was not standardized but developed

would improve DINEPA’s access to local water committees

speciﬁcally for each department based on the trainer’s judg-

and water sources nationally, particularly the more remote

ment). The materials were then presented in a Train-the-

rural communities. Box 1 outlines the TEPACs’ key roles

Trainers session to approximately 20 of DINEPA’s WASH

and responsibilities in the communes.

specialists from the URDs who had been selected to

In 2011 and 2012, CDC developed a 2-week long train-

implement the training in their departments. All materials

ing course for the TEPAC program, and met with DINEPA

produced for the trainings were made available in English

representatives to review the content and verify that it was

and Haitian Creole. In the summer and fall of 2012,

consistent with existing DINEPA guidelines and ﬁeld

DINEPA conducted the TEPAC trainings in each of

practices (see Table 1).

Haiti’s 10 departments.

161

Table 1

B. Hubbard et al.

Haiti’s WASH workforce

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Potable Water and Sanitation Technicians for the Communes (TEPAC) training program modules

Module topic

Description

DINEPA institutional framework

Provides key information about employment with DINEPA

Principles of water quality

Highlights the WHO framework for drinking water and explains the role of the TEPACs

Water associated diseases

Lists speciﬁc diseases classiﬁed as water-borne, water-washed, water-based and water-related

Groundwater ﬂow

Emphasizes the concept of groundwater and promotes understanding of factors affecting storage and ﬂow of groundwater

Environmental sampling methods

Presents the role of environmental sampling, lists essential equipment, describes typical water quality tests and sampling protocols

Principles of water supply infrastructure

Provides a basic understanding of water supply infrastructure and function

Sanitary inspection

Explains the value of sanitary inspection and provides speciﬁc guidance on conducting sanitary assessment

Household water treatment and safe storage

Describes the concept of safe water and introduces several methods and disinfecting agents for household drinking water

Water quality testing I: measuring free chlorine residual

Offers step by step, visual instruction for measuring free chlorine residual using a color disc test kit with reagent

Water quality testing lab and practice

Suggests ideas for setting up ﬁeld experiences to promote competence of technicians measuring free chlorine residual

Disinfection of small distribution systems

Provides information on disinfecting agents, explains how to calculate disinfectant dosage, and suggests methods for disinfection of community systems

Chlorine demand and jar test

Describes the jar test methodology and how to determine proper dosage to obtain expected free chlorine residual

Formation of local water committees

Presents guidance related to the DINEPA requirements for establishing a local water and sanitation committee

Sanitation

Lists different sanitation technologies and the suitability for each technology based on environmental conditions

Hygiene

Emphasizes importance of water disinfection, handwashing and practicing safe sanitation

Emergency response

Relates all the potential roles and responsibilities of an environmental health specialist during an emergency or disaster event related to water and sanitation

Cholera

Emphasizes lessons learned from early efforts to mitigate the spread of cholera in Haiti

TRAINING RESULTS

were particularly interested in understanding where TEPACs had performed well, and where they had scored

The DINEPA TEPAC coordinator and trainers decided to

the lowest to indicate how additional training resources

only administer quizzes after the delivery of each training

should be directed. Evaluation of lowest scoring examin-

module (listed in Table 1) and conduct a comprehensive

ation results from the training indicated that many

ﬁnal exam after all training modules had been delivered.

TEPACs would beneﬁt from additional training on

Trainers did not administer a pre-test exam. The results of

ground water ﬂow, disinfection of small water systems,

the quiz scores and ﬁnal comprehensive exams were aver-

and sanitary inspections. However, ﬁeld visits to small

aged across all departments and training modules. The

water supply systems conducted after the delivery of the

ﬁnal, country-wide results and analyses were shared with

abovementioned training modules demonstrated that

all trainers after the departmental trainings were com-

TEPACs’ understanding of these topics would improve

pleted. The DINEPA TEPAC Coordinator and trainers

with continued practice.

162

B. Hubbard et al.

Haiti’s WASH workforce

RESULTS FROM INITIAL FIELD ACTIVITIES OF DEPLOYED TEPACS

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

regional and departmental level, and are simultaneously being integrated into their local municipalities, thus advancing DINEPA’s long-term goal of decentralization. Future

By March 2013, 264 TEPACs had been trained and

assessments of the TEPAC program should consider how

deployed to 133 communes throughout Haiti. Their ﬁrst

data collected by the TEPACs are used to improve rural

task was to conduct a baseline inventory of their commune,

water sources, and whether the use of smartphones for

including the water infrastructure, community water com-

reporting data from rural locations directly into a central

mittees and community health workers present in the

database has been feasible. To understand decentralization

area. This work was essential in helping DINEPA identify

efforts, assessments should also describe how TEPACs,

critical needs and set key priorities in the rural WASH

with the continued support from DINEPA regional ofﬁces,

sector. The DINEPA Rural Water and Sanitation Depart-

have been integrated into local municipal governments.

ment expanded stafﬁng to include a monitoring and evaluation specialist to support the TEPAC Coordinator. The addition of monitoring and evaluation support allowed

CONCLUSION

DINEPA to begin designing an online monitoring and evaluation system to document, analyze and report results from

The TEPAC program was essential in providing Haiti with a

all ﬁeld data collected by the TEPACs. DINEPA also

national water and sanitation workforce, one that did not

made preparations to train the TEPACs on how to report

previously exist at the commune level. The deployment of

their data electronically via smartphone and GPS devices.

the TEPACs throughout Haiti expanded DINEPA’s inﬂu-

Based on feedback from the TEPAC ﬁeld experiences,

ence and presence in the rural areas, and was a critical

DINEPA has also identiﬁed the need for more targeted

step in ensuring a standardized strategy for safe water and

trainings on issues such as sanitation, household water treat-

sanitation provision throughout the country. Coordinating

ment and safe storage, and emergency response.

activities in each commune will be a key beneﬁt for all future WASH work managed by the Government of Haiti as well as both local and international organizations.

DISCUSSION Since being deployed in late 2012 and early 2013, the

ACKNOWLEDGMENTS

TEPACs have become an integral part of DINEPA’s national water and sanitation strategy, and are now recog-

The authors would like to acknowledge the contributions of

nized across the country as being essential to DINEPA’s

the Inter-American Development Bank (IADB), and the

plan to eliminate cholera (Republic of Haiti a, b). In

World Bank for their ﬁnancial and technical support of

addition to providing DINEPA with a baseline understand-

the TEPAC program. Additionally, the authors would

ing of the WASH conditions in the rural areas, the

like to thank Dr Thomas Handzel for his leadership

TEPACs have started dialogue with other WASH actors in

and direction of WASH activities in Haiti, and Kitty

their communes as a way to promote collaboration and

Middleton, Molly Patrick and Martin Fogl for their

transparency. Finally, the TEPACs are the driving force

technical, management and communication support with

behind DINEPA’s process of decentralizing the water and

stakeholders in the water and sanitation sector in Haiti.

sanitation sector. The TEPACs are located in the communes, therefore very close to the communities and the local authorities. The agency’s long-term goal is to have the sector

DISCLAIMER

managed by the local governments in each commune with oversight from DINEPA’s four regional ofﬁces. The

The ﬁndings and conclusions in this report are those of the

TEPACs receive technical support from DINEPA at the

authors and do not necessarily represent the views of CDC.

163

B. Hubbard et al.

Haiti’s WASH workforce

REFERENCES Ministry of Public Health and Population (MSPP)  Cholera Case Report. http://www.mspp.gouv.ht/site/index.php? option=com_content&view=article&id=120&Itemid=1 (accessed 12 September 2013). Pan American Health Organization (PAHO)  PAHO Director Welcomes UN Support for Cholera Elimination in Hispaniola. http://www.paho.org/coleracoalicion/?p=295 (accessed 5 December 2013). Republic of Haiti: Ministry of Public Health and Population, National Directorate for Water Supply and Sanitation a National Plan for the Elimination of Cholera in Haiti, 2013– 2022. Port-au-Prince: Republic of Haiti. Available at: http:// new.paho.org/hq/index.php?option=com_docman& task=doc_download&gid=20326&Itemid=270&lang=eng (accessed 12 September 2013). Republic of Haiti: Ministry of Public Health and Population, National Directorate for Water Supply and Sanitation b National Plan for the Elimination of Cholera in Haiti,

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

2013–2022, Short Term Plan 2013–2015. Port-au-Prince: Republic of Haiti. Available at: http://new.paho.org/hq/index. php?option=com_docman&task=doc_view&gid=20578& Itemid=270&lang=en (accessed 12 September 2013). Republic of Haiti: National Directorate for Potable Water and Sanitation (DINEPA)  Rapport Annuel 2009–2010. http:// www.pseau.org/haiti-eau/images/documents/ dinepa_annual2009_2010ﬁnal.pdf (accessed 5 December 2013). Republic of Haiti: National Directorate of Potable Water and Sanitation (DINEPA)  Plan Opérationnel Général 2012– 2014: Reforme et investissements dans le secteur de l’eau potable et assainissement en Haïti. March 2012. Spanish Agency for International Development Cooperation. http:// www.fondodelagua.aecid.es/galerias/fcas/descargas/ proyectos/pog/POG_HTI-003-B.pdf (accessed 8 July 2013). WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation (JMP)  Progress on Drinking Water and Sanitation: 2012 Update. http://www.wssinfo.org/ﬁleadmin/ user_upload/resources/JMP-report-2012-en.pdf (accessed 19 March 2013).

First received 16 July 2013; accepted in revised form 13 September 2013. Available online 22 October 2013

164

Practical Paper

04.1

2014

Practical Paper Developing behaviour change communication for improving faecal sludge management in Bhutan G. Halcrow, T. Yetsho, N. Nguyen and G. Tshering

ABSTRACT Approaches to improve faecal sludge management practices and on-site sanitation services can better be supported through an understanding of the behavioural determinants and consumer preferences for the services. This practice paper describes a process in Bhutan in which stakeholders collaborated to carry out formative research to identify motivations and barriers to improved septic tank management by building owners in Thimphu City. Using a sanitation behaviour change framework, the ﬁndings focused on prioritised behavioural determinants of knowledge of septic tank management and services, willingness to pay and the use of enforcements. The research ﬁndings were then used to develop a practical yet evidence-based behaviour change communication strategy for the City in 2012 that aimed to improve safe and timely emptying practices, improve services and

G. Halcrow (corresponding author) T. Yetsho N. Nguyen G. Tshering SNV Netherlands Development Organisation, Langjophaka, PO Box 815, Thimphu, Bhutan E-mail: ghalcrow@snvworld.org and Ministry of Works and Human Settlements, PO Box 791, Thimphu, Bhutan

increase demand from households, business and institutions. Key words

| behaviour change, faecal sludge, urban sanitation

INTRODUCTION In Bhutan, urban growth is accelerating at an average of 7%

undertaken of the legal and institutional arrangements guid-

per year. By 2020 over half of the population will reside in

ing urban sanitation and hygiene nationally identiﬁed

urban areas (World Bank ). The rapid urbanisation is

several gaps and recommended institutional capacity

increasing pressure on the environment and the provision

strengthening and resource allocation for sludge manage-

of on-site sanitation services, highlighting the need for

ment, wastewater management and hygiene promotion

improved management of faecal sludge. Approximately

(Carrard & Omu ). In parallel in 2012, national regu-

70% of the estimated 120,000 residents of Bhutan’s capital

lations were introduced under the Waste Prevention and

city of Thimphu rely on on-site sanitation (Dorjee ).

Management Act 2009 that provided a basis to apply sanc-

Beyond the capital, regional towns also continue to promote

tions and enforcements.

on-site sanitation but without a corresponding focus on

Since the partnership began in 2011, formative research

sludge management services. The growing environmental

has been conducted to improve understanding of consumer

health issues are difﬁcult to ignore with leaking septic

needs and preferences for de-sludging services and of behav-

tanks and infrequent and unsafe emptying practices.

ioural

determinants

that

can

motivate

and

enable

The Ministry of Works and Human Settlements

householders to access and pay for these services – ﬁrstly

(MoWHS) as the lead line agency is working in partnership

in three small regional towns in Chukha District (Pedi

with SNV (Netherlands Development Agency) and Thim-

) and then in Thimphu City (Yetsho & Nguyen ).

phu City Corporation (the ‘Çity’) to develop approaches to

The consumer research study from Chukha District pro-

ultimately improve access to environmentally safe sanitation

vided initial insights into key barriers to proper sludge

and hygiene practices in urban settlements. A review

management which included access to services and

doi: 10.2166/washdev.2013.131

165

G. Halcrow et al.

Behaviour change communication to improve faecal sludge management

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

knowledge. They also indicated a demand for and potential

willingness to pay; and the effect of enforcements. The

willingness to pay for de-sludging services within towns that

research was facilitated as a capacity-building process with

are planning to provide future services. The study included

key government agency and practitioners actively engaged

ﬁndings that were explored in further depth and validated

in lead roles. It informed the development of a planned be-

in Thimphu where, in the absence of private sector oper-

haviour change communications strategy for the City.

ators, the City currently provides sludge removal services but with limited uptake and challenges in service quality. For Thimphu City’s estimated 84,000 residents, typically

METHODS

living in rented apartment buildings and relying on septic tanks, there are a range of service options provided. Timely

The formative research focused on tenants’ and building

de-sludging is recommended every two years on an on-call

owners’ knowledge and understanding of on-site sanitation

basis. Building inspectors regulate the provision and design

and related operational and maintenance needs, existing

of the septic tank as part of the occupancy permit process

practices related to septic tank maintenance, awareness of

in line with the Building Code but guidance on operation

existing service options, preferences for service options

and maintenance is not speciﬁcally addressed. For residents

and reported willingness to pay. Formative research was

connected to the water metre, 50% is allocated to sanitation

understood to be research conducted during the develop-

services and entitles the resident to ‘three free trips’ per year. This service was recently expanded to ‘six free trips’. The actual number of trips required though depends on the size of the septic tanks and the volume of the tankers at 3 cubic metres. Larger buildings and institutions or communal

ment of a programme to assist in planning and designing interventions (Gittelsohn et al. ). The study used the SaniFOAM (sanitation Focus – Opportunity – Ability – Motivation) behaviour change framework to identify the key barriers and motivations for

tank owners may need to pay for additional trips at a cost

the behaviour leading to development of the research

equivalent to US$19 if the tankers’ capacity is not large

hypothesis by stakeholders (Devine ). SaniFOAM is a

enough to empty the tanks within the allocated number of

framework for behavioural determinants developed in

trips. Residents without a water metre can access the cess-

rural sanitation to understand opportunities, abilities and

pool services for this same fee per trip plus additional

motivations for the behaviour of interest among the target

transportation costs for households located outside of the

population. The SaniFOAM framework illustrated in

Thimphu core area. There are only limited records main-

Figure 1 is structured in terms of:

tained, the total number of septic tanks and the actual

•

Desired behaviour. The behaviour that needs to be chan-

•

Target population. The group that in most cases needs to

frequency of emptying is unknown. It is considered too low based on the number of overﬂowing or failing tanks visible or reported as complaints. Records indicate on average 54 trips are made by the three tankers per month with sludge disposed of to the treatment plant. Those not accessing the City service resort to manual unregulated emptying, potentially unsafe and illegal disposal or allow tanks to overﬂow. This practice paper describes the use of formative research to inform strategies to improve management of onsite sanitation systems in Thimphu undertaken collabora-

• • •

ged, e.g. timely de-sludging of septic tanks. be targeted to adopt the behaviour. Opportunity. Institutional or structural factors that influence an individual’s chance to perform the behaviour. Ability. An individual’s skills and proficiency to perform the behaviour. Motivation. Drives, wishes, urge or desire that influence an individual to perform behaviour.

tively by stakeholders in 2012. The objective of the research

Whilst no available literature was found on consumers’

was to identify motivations and barriers to improved septic

views on faecal sludge management, the process built on the

tank management by building owners in Thimphu City with

preliminary consumer research in Chukha and available base-

a speciﬁc focus on the prioritised behavioural determinants

line data. Based on a household survey of a 10% stratiﬁed

of: knowledge of septic tank management and services;

random sample, the baseline reported that, whilst most

166

Figure 1

G. Halcrow et al.

Behaviour change communication to improve faecal sludge management

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

The SaniFOAM framework and classiﬁcation of behavioural determinants. Source: Devine 2009, p 5.

households knew where the septic tank was located, 60%

their water bills and those who had used the recently avail-

reported that they did not know how often their septic tank

able ‘six free trips’ extended services provided within the last

needed to be emptied and only 6% of respondents felt it was

few months (service users). The qualitative study involved

necessary to empty their septic tank regularly (Penjor et al. ).

in-depth interviews with a simple random selection of build-

Hypotheses were developed in a participatory workshop

ing owners and tenants from the City’s record of users who

process with key stakeholders and practitioners to target the

had called for de-sludging services from January 2011 till the

key behavioural determinants for exploration in the second

beginning of 2012, as summarised below.

study. Rather than focus on all of the behavioural determi-

•

nants in the framework, this step sought to prioritise and reﬁne the scope based on the existing experience and knowl-

spective including volume of services, operation and

most value to explore in the given context for the strategy iour determinants were developed:

sewerage division and cesspool tanker workers were interviewed to better understand the service provider per-

edge of the stakeholders and the area perceived of being of development. The following hypotheses grouped by behav-

Key informant in-depth interviews: The head of the City

•

maintenance and scope for improvement, etc. In-depth interviews with users and non-users of the services: A total of 43 interviews were conducted, 38 with

•

If building owners knew about the de-sludging services

building owners (15 M, 23 F) and 7 with tenants (3 M,

•

available, they would empty their septic tanks (knowledge).

4 F), to better understand the demand for de-sludging ser-

If building owners knew how to check if their tanks are

vices. Of these, 28% reported as non-users of the service.

• •

full, they would empty their septic tanks (knowledge). If building owners felt that the cost of de-sludging was

Four interviewers led by a lead researcher from SNV

good value for money, they would pay for the additional

and the MoWHS in close coordination with the City

cost for more than ‘three free trips’ (willingness to pay).

during April 2012 undertook the data collection. The

If building owners knew that they could be ﬁned they would empty their septic tanks (sanctions).

research team members received one day’s training on the data collection methods, tools and process and were mentored through the process subsequently. Question guides

Research tools included in-depth interviews with Thim-

were developed and tested. Data processing and analysis fol-

phu City Corporation staff and building owners including

lowed manual transcriptions, collation, summarisation and

those who had never called for de-sludging services (non-ser-

analysis and was undertaken by the team identifying key

vice users), those who called for services provided as part of

themes under the determinant classiﬁcations.

167

G. Halcrow et al.

Behaviour change communication to improve faecal sludge management

RESULTS

Table 2

The ﬁndings are presented according to the prioritised behavioural determinants of knowledge as it relates to ability, sanctions as it relates to opportunity and willingness to pay in terms of motivations using the SaniFOAM framework. Knowledge As Table 1 illustrates, all building owners were aware of the location of their septic tank and water metre systems and most reported they knew how to look inside the tank.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Knowledge of sludge management among non-service users

Enablers (þ)

Inhibitors ( )

þ All building owners were aware of the location of their septic tank and water metre system

Building owners did not know how to determine if their septic tanks are ‘full’

þ Most building owners looked inside the inspection chamber to determine if it needed emptying

Building owners did not know that they need to check the sludge level to determine if it needed emptying

þ Building owners knew that grey water should not go into the toilet

Most building owners were not clear on the services provided by the City and the process of de-sludging services

Through experience, they were aware of the existing services including the ‘three free trips’ offered by the City as well as the process of de-sludging.

using the service. The main difference was that most non-

Knowledge gaps did exist. Although most respondents

service users had less understanding of the existing services

reported looking inside the inspection chamber this will

provided by the City and the process of de-sludging. Fewer

only indicate the liquid depth. The MoWHS recommends

building owners reported looking inside the inspection

that the tank be emptied when the sludge occupies approxi-

chambers of their septic tank than users did. Like users,

mately two-thirds of the total depth of the septic tank, which

non-users did not know how to determine correctly whether

would require sludge depth to be checked. With the recent

their tanks were ‘full’.

addition of the ‘six free trips’ by the City, most service users were not yet aware that they were now entitled to an additional three trips per year. Table 2 represents knowledge among the non-service user group who had not had their septic tanks de-sludged

Beliefs and attitudes More than half of the building owners did not feel they needed to empty their septic tanks until they overﬂowed. For example ‘I take my own initiative to empty the septic

Table 1

Knowledge of sludge management among service users

tank when it is full and starts to overﬂow. I also do not know how to check my septic tank unless it overﬂows’

Enablers (þ)

Inhibitors ( )

(building owner, Thimphu, 26 April 2012). The majority of

þ All building owners were aware of the location of their septic tank and water metre system

Most building owners did not know how to determine if their septic tanks were ‘full’ and therefore when to call for services

building owners were satisﬁed with the process of accessing

þ Most building owners did report looking inside the inspection chamber to determine if it needed emptying

Many building owners did not know that they need to check the sludge level to determine if it needed emptying

time consuming for ofﬁce people and waste of resources

þ Most building owners knew about existing services and the process of de-sludging

Most building owners did not know about the new extended services

þ Building owners knew that grey water should not go into the toilet

the services; however, a few institutions felt the process was time-consuming and could be improved. ‘The process of getting the services from Thimphu Thromde (the City) is really such as ﬁlling in the hard copy form’ (institution, Thimphu, 26 April 2012).

Willingness to pay The majority of building owners were willing to pay the price charged by the City for de-sludging services. Those who were not connected to the water metre though found

168

G. Halcrow et al.

Behaviour change communication to improve faecal sludge management

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

it expensive owing to the cost of de-sludging coupled with

In the Bhutan context the use of formative research in

additional transportation costs if outside the core town area.

the development of behaviour change communication and improving services was new. Typically communication

Sanctions

materials or messages are developed based on perceived knowledge gaps with a strong health message focus. In

Most of the respondents felt that imposing ﬁnes would

this capacity-development process, the ﬁndings of the for-

increase the demand for de-sludging services and force

mative research in comparison with the hypotheses

building owners to become more aware of the services pro-

developed by the stakeholders highlighted the need for

vided by the City. However, some informants felt that if

the concerned agencies to undertake consumer research.

services could be improved, then ﬁnes would be unnecess-

In understanding more about consumer needs and

ary. ‘If there are improvements in services from Thimphu

preferences they either validated or debunked existing

Thromde, then imposing ﬁne is not required or relevant’

assumptions before developing communications as part of

(ofﬁce, Thimphu, 26 April 2012).

broader programme development. For example, in terms

Consumers reported access to services as adequate but sat-

of the hypotheses that most building owners will not pay

isfaction with the quality recorded was low, including response

for services beyond the free trips included in their water

time, the process of accessing the services and administrative

bills or that they feel that the cost of de-sludging their

processes. There is an indication of the presence of physical

septic tanks is not worth it both proved to be incorrect.

and social pressures, which are potential motivators. Some

The research instead found that, for example, they are will-

building owners were encouraged to call for services after

ing to pay, but a variety of factors related to knowledge and

foul odours emitted from their septic tanks and/or upon com-

attitudes may play a larger role in convincing owners to call

plaints from their neighbours. Institutions in turn were

for services.

motivated to call for de-sludging services owing to regulations

The outcomes from the research were used to develop

requiring public places such as schools and ofﬁces to be clean

an evidence-based behaviour change communication strat-

and hygienic. ‘The school requires regular emptying of septic

egy for Thimphu City and for the urban settlements of

tanks as it is the school space where there are lots of children

Chukha District in 2012 in a participatory workshop pro-

and it’s very important to maintain hygiene at all times’ (tea-

cess with stakeholders. In addition to improving timely

cher, middle secondary school, Thimphu, 27 April 2012).

and safe de-sludging, the strategies also target hand-washing practices at critical times and improving sanitation in the informal settlements based on parallel formative research

DISCUSSION

processes. The strategies for timely and safe de-sludging included communication objectives focused on the determi-

Users of the services were more knowledgeable about the need

nants of knowledge and attitude and used a range of

to empty their septic tanks than non-service users. With a ser-

communication channels to reach the target audience. A

vice model based on on-call services, a key area to address was

speciﬁc objective was included to raise awareness of the

the belief that services were only needed when tanks failed or

new waste management regulations to ensure that Thimphu

overﬂowed owing to the related environmental health risks

City residents are conscious of the ﬁnes and fees should they

and reduced effectiveness of the tank over time. The knowl-

be imposed at a later date.

edge gaps related to how to check sludge depths can be

Informed by the research, stakeholders identiﬁed a

addressed though communications but this needs to be moni-

number of communication activities and channels to reach

tored. Knowledge alone may not be a sufﬁcient enabler and

building owners:

scheduled servicing may be a more realistic solution in the longer term. Customers reported a willingness to pay for extra services if they were improved although quantifying this would require more in-depth research.

•

City ofﬁce. Visual materials to be displayed so that customers coming to enrol or pay for services are exposed to messages.

169

• •

G. Halcrow et al.

Behaviour change communication to improve faecal sludge management

Water bill user meetings. During the year, meetings are held with households connected to the water metre, which can be used to communicate the objectives. Toll-free hotline. Existing service for residents seeking

Journal of Water, Sanitation and Hygiene for Development

Table 3

Building

inspectors

play

Communications

Strategy targeting building owners’ knowledge and attitudes to: Know why and how to check the sludge depth of their septic tanks Know when to call for de-sludging services Know the range of services offered and their entitlements Feel that it is important to empty their septic tanks before they overﬂow Know about the new regulation on waste management

• • • •

crucial role throughout the construction process and could communicate in relation to operation and

•

maintenance.

•

Occupancy permit. The required meeting to issue the permit offers another opportunity to ensure that building owners know how to manage their septic tanks and when to call for services.

Service delivery level improvements

• •

Water bills. Messages and reminders related to septic tank management and how to access services, including the details of the toll-free hotline and the extended ser-

• •

vices can be printed on all water bills. The strategies summarised in Table 3 are now being implemented and monitored combined with longer-term

2014

Actions

de-sludging. inspectors.

Level

advice or to lodge a complaint can also serve as a

Building

04.1

Strategies to improve communications and service delivery improvements

channel to provide information on timely and proper

•

Institutional level

improvements to service delivery and institutional arrange-

• •

ments as endorsed by the Mayor in 2012.

•

CONCLUSION The SaniFOAM framework has historically been used in

Greater responsiveness to requests for services, particularly emergency requests Provide customers with advice on septic tank monitoring and management following de-sludging of their septic tanks Improve the toll-free service hotline Provide equal attention to customers not connected to a water metre Develop interim measures to enable households without access to roads to safely empty tanks Develop long-term plans to enable more households to connect to sewerage treatment options Fines can be considered but must be coupled with a strong communications campaign on the beneﬁts of compliance and the disadvantages of non-compliance

designing rural sanitation programmes but was applied to the urban sanitation service delivery setting in this process. As a tool for understanding the determinants of sanitation

hypothesis in a participatory process using the framework

behaviours it proved a useful framework when applied in

the research was able to focus on a deﬁned target group,

the urban context of faecal sludge management. It ensured

the building owners, and the prioritised determinants,

a better understanding of issues from the perspective of con-

which was manageable from a human resources and

sumers and end-users that can enable the government and

capacity perspective. Following these types of processes

service providers to effectively target resources towards sani-

in broader programme designs may also assist in providing

tation services that people want, understand and are willing

practitioners with clarity in terms of integrating and target-

to pay for.

ing communications.

The framing in terms of behavioural determinants

The process overall contributed to developing a more

allowed the stakeholders to break down the larger question

appropriate and targeted consumer awareness, education

of ‘why don’t people empty their septic tanks’ in a struc-

and behavioural change communication campaigns and par-

tured process concerning determinants of knowledge,

allel consumer support service improvements. Undertaking

attitudes and willingness to pay. Through deﬁning the

formative

research

capacity-building

approach

170

G. Halcrow et al.

Behaviour change communication to improve faecal sludge management

embedded within a broader programme contributed substantially to the ownership of the overall process and the sustainability.

REFERENCES Carrard, N. & Omu, D.  Review of Policy, Legal and Institutional Arrangements for Urban Sanitation and Hygiene in Bhutan. SNV and Ministry of Works and Human Settlement, Bhutan. Devine, J.  Introducing SaniFOAM: A Framework to Analyse Sanitation Behaviours to Design Effective Sanitation Programs. Water and Sanitation Programme, Washington DC. Dorjee, T.  Population Projections 2005–2050 Report. National Statistics Bureau, Thimphu.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Gittelsohn, J., Steckler, A., Johnson, C., Pratt, C., Grieser, M., Pickrel, J., Stone, E. J., Conway, T., Coombs, D. & Staten, L. K.  Formative research in school and communitybased health programs and studies: ‘State of the Art’ and the TAAG approach. Health Educ Behav. February 33 (1), 25–39. Pedi, D.  Consumer Research Report. Chukha District, Ministry of Works and Human Settlements and SNV, Bhutan. Penjor, K., Yetsho, T. & Bond, M.  Sustainable Sanitation and Hygiene for All in Small Towns. Chukha Baseline Assessment Report, SNV Bhutan and Ministry of Works and Human Settlements, Bhutan. World Bank  Bhutan Urbanization Strategy. Infrastructure Professionals Enterprise, (P) Ltd, India & in association with Gyaltshen Consultancy Services, Bhutan. Yetsho, T. & Nguyen, N.  Formative Research of Sludge Management Related Behaviours in Thimphu. Ministry of Works and Human Settlements, Thimphu City Corporation and SNV, Bhutan.

First received 8 February 2013; accepted in revised form 6 October 2013. Available online 13 December 2013

171

Practical Paper

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Practical Paper Development of indicators for measuring outcomes of water safety plans Gabriella Lockhart, William E. Oswald, Brian Hubbard, Elizabeth Medlin and Richard J. Gelting

ABSTRACT Water safety plans (WSPs) are endorsed by the World Health Organization as the most effective method of protecting a water supply. With the increase in WSPs worldwide, several valuable resources have been developed to assist practitioners in the implementation of WSPs, yet there is still a need for a practical and standardized method of evaluating WSP effectiveness. In 2012, the Centers for Disease Control and Prevention (CDC) published a conceptual framework for the evaluation of WSPs, presenting four key outcomes of the WSP process: institutional, operational, financial and policy change. In this paper, we seek to operationalize this conceptual framework by providing a set of simple and practical indicators for assessing WSP outcomes. Using CDC’s WSP framework as a foundation and incorporating various existing performance monitoring indicators for water utilities, we developed a set of approximately 25 indicators of institutional, operational, financial and policy change within the WSP context. These outcome indicators hold great potential for the continued implementation and expansion of WSPs worldwide. Having a defined framework for evaluating a WSP’s effectiveness, along with a set of measurable indicators by which to carry out

Gabriella Lockhart William E. Oswald Brian Hubbard Elizabeth Medlin Richard J. Gelting (corresponding author) Water, Sanitation, and Hygiene Team, Environmental Health Services Branch, National Center for Environmental Health, United States Centers for Disease Control and Prevention, Department of Health and Human Services, Atlanta, GA, USA E-mail: rgelting@cdc.gov Richard J. Gelting 4770 Buford Highway NE, MS F-57, Atlanta, GA 30341, USA

that evaluation, will help implementers assess key WSP outcomes internally, as well as benchmark their progress against other WSPs in their region and globally. Key words

| drinking water, evaluation, indicators, outcomes, water safety plans

INTRODUCTION The World Health Organization (WHO) deﬁnes water safety

means of evaluating WSP implementation and measuring

plans (WSPs) as the ‘use of a comprehensive risk assessment

their effectiveness.

and risk management approach that encompasses all steps in

The US Centers for Disease Control and Prevention (CDC)

water supply from catchment to consumer’, and promotes

published a conceptual framework for the evaluation of WSPs

them as ‘the most effective means of consistently ensuring

as a means of standardizing the way that they are evaluated

the safety of a drinking-water supply’ (WHO ). WSPs

worldwide (see Figure 1; Gelting et al. ). The framework

are currently being implemented in contexts as diverse as

focuses on a WSP’s outcomes and overall impacts, and is

Australia, Uganda, Canada and Jamaica; with such wide-

designed speciﬁcally to support existing WSP evaluation

spread adoption, there exists a need for a straightforward

tools, such as the Water Safety Plan Quality Assurance Tool developed by WHO and the International Water Association

(IWA) (WHO/IWA ). The framework presents four principal categories of outcome that represent the changes that occur from WSP implementation: institutional, operational, ﬁnancial

172

G. Lockhart et al.

Figure 1

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Conceptual framework for the evaluation of water safety plans (Gelting et al. 2012).

and policy changes. In order to apply this evaluation frame-

at www.ib-net.org). While all of these resources are valuable

work, there is a need for a set of simple and measurable

for water utilities in various contexts, there is no existing set

indicators that speciﬁcally assess outcomes in the WSP process.

of standardized indicators to evaluate water safety plans in

For some of the outcome areas – particularly the oper-

particular. There is still the need for a list of simple and mea-

ational and ﬁnancial changes – there are already many

surable indicators that apply speciﬁcally to WSP outcomes.

standardized and validated performance indicators that

Impacts from WSPs, such as improved water quality and

exist to monitor water utilities worldwide. A particularly

health, have been documented in a limited number of cases

comprehensive resource for monitoring the performance of

(e.g. Dyck et al. ; Gunnarsdottir et al. a); however,

water utilities is the International Water Association’s Per-

these impacts often take a long time to become apparent

formance Indicators for Water Supply Services (Alegre et al.

and can be difﬁcult to measure. In addition, impact evalu-

). There are also various country- and region-speciﬁc

ation typically requires surveillance data or expensive

methodologies for performance monitoring and benchmark-

studies that may not be feasible in many locations.

International

Therefore, we instead focus on evaluating outcomes

Benchmarking Network for Water and Sanitation Utilities

because they are a necessary ‘intermediate change’ that

(IBNET) is a part of the World Bank’s Water and Sanitation

then results in impacts such as water supply and health

Program and presents indicators, comparison data and tools

improvements. Outcomes from WSP implementation have

to facilitate benchmarking among water and sanitation sys-

also been more widely documented (Gunnarsdottir et al.

tems worldwide (van den Berg & Danilenko ). The

b, Gelting et al. ). Evaluating the proposed outcomes

network’s extensive list of regional partners includes the

will help to demonstrate the positive effects from WSPs

Association of Potable Water and Sanitation Regulators for

without requiring the extensive surveillance, studies and

Latin America (ADERASA) and the South East Asia Water

time necessary to show impacts such as improved water

Utilities Network (SEAWUN). These partners conduct

quality and health (Gelting et al. ).

ing

that

have

been

developed.

The

benchmarking activities among regional water utilities, and

In this paper, we aim to operationalize CDC’s conceptual

their data contribute to the IBNET global database (available

framework for evaluating WSP outcomes by presenting a set

173

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

of speciﬁc, measurable and standardized indicators that are applicable to various WSP contexts worldwide. These indicators were selected as practical, straightforward tools that WSP teams, water utilities and regulatory agencies of any size or stage of development can use to evaluate the outcomes of the WSP process. The indicators are designed to be clear and simple enough to be easily measured using data that the water utility or other participating organizations involved in WSPs may already collect for their own records. These indicators would not only facilitate the evaluation of individual WSP programs, but they would also permit benchmarking between multiple WSP initiatives, further incentivizing and increasing the effectiveness of implementing organizations (Alegre et al. ; Vieira ). The purpose of this paper is not to provide an exhaustive listing of all possible indicators that could be used to measure WSP effectiveness; rather, it is to offer a core list of indicators that are relevant and applicable to the multiple contexts in which WSPs are implemented. Chosen speciﬁ-

Figure 2

Water safety plan conceptual framework outcomes.

cally to evaluate WSPs rather than water utilities in general, these indicators will measure the four outcome

The indicators for each outcome area are also listed in

areas in CDC’s conceptual framework for WSP evaluation.

Tables 1–4. Some of the indicators for institutional, operational and ﬁnancial changes include an IWA and/or IBNET code, showing that the indicator was either taken

METHODS

directly from or adapted from IWA’s or IBNET’s lists of performance indicators for water utilities (Alegre et al. ;

CDC’s conceptual framework for WSP evaluation identiﬁed

van den Berg & Danilenko ).

four categories of outcomes; the changes that are expected to occur within these four outcome areas are listed in

Institutional outcomes

Figure 2. Indicators have been selected in order to operationalize the framework and measure changes in the outcomes.

Increased communication and collaboration

For practicality, various indicators already developed by IWA and IBNET were incorporated into this list and are

The formation of a complete WSP team itself is a vital ﬁrst

referenced accordingly. The indicators presented in this

step to the WSP process because it brings together relevant

paper are examples that may be relevant to any utility.

stakeholders to discuss, devise and implement steps to

WSP teams and other evaluators can choose and adapt

ensure the safety of a water supply. As with any group, com-

those indicators that will be most beneﬁcial to them in

munication and collaboration might not occur immediately

their speciﬁc contexts.

but instead develop over time. The following areas of focus for evaluating increased communication and collaboration were loosely developed, in part, from the network level of

RESULTS AND DISCUSSION

effectiveness criteria presented by Provan & Milward (). However, as discussed above, these indicators are

The proposed indicators are presented in bold lettering and

designed to be practical rather than comprehensive, so

explained in terms of the relevant outcomes they will measure.

many of them are simpliﬁed and process-oriented.

174

Table 1

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indicators of institutional outcomes

Outcome area

Indicator

Code

Measurement

Data source

Increased communication and collaboration among stakeholders

Number of meetings per year

N/A

# of meetings

Number of participating entities on WSP team

N/A

WSP team attendance logs WSP team attendance logs

Existence of inter-institutional agreements or scopes of work Documented WSP team work plan

N/A

(Number of registered stakeholders represented at WSP team activities / number of registered stakeholders) × 100 (in %) Yes/No

N/A

Yes/No

WSP team documents

Improved knowledge and attitudes related to the drinking water system among water utility staff and other stakeholders

Existence of a comprehensive description of the water supply system and the identiﬁcation of actual and potential hazards Existence of a revised draft of the document each year

N/A

Yes/No

WSP team documents

N/A

Yes/No

WSP team documents

Improved knowledge and attitudes related to employee satisfaction among water utility staff

Employee satisfaction

N/A

Measurement may vary based on the utility’s existing human resources data collection methods; staff turnover ratio (# of employees who left over the past year/total # of employees) if nothing else available

HR documents

Increased training

Total training

IWA Pe19

Hours / employee / year

Training attendance logs

WSP team documents

The number of WSP team meetings per year is an

WSP team to carry out its activities. Second, these inter-

indicator that reﬂects the level of communication within

institutional agreements formalize and solidify the relation-

the team by noting the potential opportunities to bring

ships between WSP team members or other entities

together team members to develop and implement the

involved in activities related to water supply. Relationships

WSP process. Tracking the number of participating enti-

among various participating entities may be tenuous at ﬁrst

ties on the WSP team can identify the individuals and

and take time to build trust; inter-institutional agreements

institutions (e.g. the water utility, ministry of health,

help to build and strengthen the team and the WSP

etc.) that make up the WSP team and measure how

process.

these stakeholders and their active participation may

The production of a WSP team work plan is indicative

change over time. A broad WSP team is more likely to

of greater collaboration and is useful as a baseline record by

have the experience and expertise necessary to under-

which to evaluate the team’s progress as it moves towards

stand all facets of the water supply system (Bartram

meeting its goals. The work plan provides the structural

et al. ).

mechanisms for the coordination and administration of

One aspect of a team’s development is the institutiona-

the WSP team – for example, decision-making processes

lization of the WSP team and process. The existence of

rules and regulations, management structures and resource

inter-institutional agreements or scopes of work related

allocation. The work plan can provide a clear measure not

to drinking water are instrumental in two important

only of the team’s effectiveness, but also of its operation

ways. First, these documents reﬂect the legitimacy of the

and development.

175

Table 2

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indicators of operational outcomes

Outcome area

Indicators

Code

Measurement

Data source

Improvements to system infrastructure

Source protection

N/A

Water utility records

Water treatment: chlorine residual

IBNET 15.4

Storage tank cleaning

Op2

Distribution: Mains rehabilitated, renovated, and replaced Mains added Water losses per system input volume Alternative: nonrevenue water by volume

IWA Op 16–18

Yes/No; Specify type (fencing, wellhead protection, legal protection, etc.) Number of tests of treated water for residual chlorine/required number of tests of treated water for residual chlorine ×100 (in %) Volume of storage tanks cleaned during the assessment period/total volume of storage tanks %/year

Implementation of improved procedures

N/A IWA variables A15/A3 (see text) IWA Fi46 IBNET 6.1

Customer metering level/density

IWA Ph11 IBNET 7.1

Development and implementation of standard operating procedures

IWA Qp1–4, 6 IWA Op7–11 N/A IWA Op40–44, IBNET 15.3

Increased documentation of standard operating procedures Customer complaints

N/A

IWA Qs26 IWA Qs34 IBNET 16.1*

%/year Water losses during the assessment period/ system input volume ×100 (in %) Non-revenue water/system input volume ×100 (%) (Volume of water produced minus volume of water sold) / volume of water produced ×100 (%) Number of meters/service connection Total number of connections with operating meter/total number of connections ×100 (in %) Inspection and maintenance of physical assets ( /year) Calibration of equipment ( /year) Number of sanitary inspections conducted ( /year) Water quality monitoring tests carried out / water quality tests required by applicable standards during assessment period (%) Periodic updates of changes made to standard operating procedures documented: Yes/No

Service complaints per connection (number of complaints/1000 connections/year) Response to written complaints (%) Complaints about water services (total number of water complaints/year, expressed as a % of the total number of water connections)

Water utility records

*Note: IBNET Indicator 16.1 represents complaints about both water and wastewater services, and is deﬁned as ‘total number of water and waste-water complaints per year expressed as a percentage of the total number of water and waste-water connections’ (van den Berg & Danilenko 2011). For the purposes of this paper, which presents outcome indicators solely applicable for water supply services within the WSP context, this indicator is modiﬁed to exclude measurements of wastewater complaints

Improved knowledge and attitudes

potential hazards represent the compilation and documentation

dispersed

institutional

knowledge.

This

The existence of a comprehensive description of the water

knowledge is then centralized within the WSP team or the

supply system and the identiﬁcation of actual and

water utility. Periodic review of the WSP document,

176

Table 3

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Indicators of ﬁnancial outcomes

Outcome area

Indicators

Code

Measurement

Data source

Cost savings

Unit total costs ($/m3)

IWA Fi4, IBNET 11.1

(Running costs þ capital costs)/authorized consumption (including exported water), during the assessment period

Water utility records

Cost recovery

Operating cost coverage ratio Collection ratio

IWA Fi31, IBNET 24.1 IBNET 23.2

Total revenues/running costs, during the assessment period Total operating revenue / total operational expenses ×100

Water utility records

Increased donor support and/or increased investment

Unit investment ($/m3)

IWA Fi25

Cost of investments (expenditures for plant and equipment) / authorized consumption (including exported water), during the assessment period

Water utility records

Table 4

Indicators of policy outcomes

Outcome area

Indicators

WSP knowledge sharing and promotion

Increased awareness of WSPs

WSP as norms of practice

Formal regulatory requirements for WSPs

Code

Measurement

Data source

N/A

Number of water utilities with knowledge of WSPs/number of water utilities, measured per year

Analysis of communications such as emails, letters and conversations, other written records Surveys or questionnaires

Number of pilot WSP projects implemented

N/A

Number implemented

Existence of WSPs being incorporated into non-regulatory guidance documents (for use by WSP implementers – utilities, NGOs, stakeholders)

N/A

Yes/No

Surveys or questionnaires by water utility associations or regulators Uniform guidelines for countries or regions Lessons learned documents

WSP preventive concepts incorporated into drinking water regulations WSP policy adoption in drinking water regulations

N/A

Yes/No

N/A

Yes/No

National drinking water regulations National drinking water regulations

resulting in a revised draft of the WSP document each

is already being collected. Human resource associations such

year, indicates the iterative nature of this process. However,

as the Society for Human Resource Management (www.

centralized knowledge will only bring about subsequent

shrm.org) produce surveys and other measures of employee

improvements to the water supply system if it is shared

satisfaction that may be helpful for WSP evaluators. If no

amongst the system’s operators, through additional training

other measures are implemented, staff turnover ratio (deﬁned

and understanding of the changes in the water system.

here as the ratio of employees who left over the past year to

Improved attitudes regarding WSP activities or, more fundamentally, regarding the provision of safe drinking

the total number of employees) is a crude measurement that can be utilized to assess employee satisfaction.

water, can be reﬂected in the level of employee satisfaction among the water utility staff. Standard human resource

Increased training

measures of satisfaction, such as job satisfaction and employee engagement surveys, could be employed to

Increased training is the most tangible concept within the

measure this within the utility, especially if such information

expected institutional changes, and its measurement is

177

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

relatively straightforward. All types of training can be

measurement and the cleaning of water tanks, respectively.

encompassed in a single indicator by measuring the

Examples of distribution indicators that demonstrate system

amount of time allotted to training. As noted in the WSP

improvements include the percentage of mains that are

evaluation framework, there are many examples of WSPs

rehabilitated, renovated, or replaced each year, as well as

leading to an increase in training on a variety of issues

the percentage of new mains that are added each year.

related to the water system (Gelting et al. ). Training

Tracking the percentage of water losses per system

staff members in areas of weakness identiﬁed through the

input volume is also an effective indicator of infrastructural

WSP process indicates efforts to improve system perform-

improvements in the system. Reducing water losses – which

ance. In addition, the ongoing education of operators and

include leaks as well as illegal connections to the system –

other relevant staff regarding the principles and ﬁndings of

can often be a low-cost, effective method by which a utility

the WSP process itself indicates increased knowledge and

can increase its system capacity. Thus, diminishing these

understanding of the drinking water system.

losses over time can indicate that the system has undergone preventive measures and infrastructural improvements. There is no speciﬁc IWA indicator for measuring total

Operational outcomes

water losses as a percentage; however, the IWA variables Service-level improvements that are visible to customers –

A15 (Water losses) and A3 (System input volume) can be

namely, improvements in water cost, quality, quantity, conti-

used to simply calculate this indicator. Alternatively, IWA

nuity and coverage – often take considerable time to occur;

and IBNET both include indicators for non-revenue water

CDC’s conceptual framework for evaluation considers these

by volume that could be measured (see Table 2). Although

improvements as impacts rather than outcomes. The WSP

these are ﬁnancial (vs. operational) indicators, they provide

team can more easily assess the effectiveness of the WSP

easy and straightforward ways to measure water loss.

with outcome indicators of direct actions resulting from

In some contexts, water metering is not well established.

the WSP process leading to these impacts (e.g. the

In these cases, an additional sign that a water utility has

implementation of control measures, rehabilitation and

undergone improvements and is moving towards sustain-

improved management of infrastructure, and implemen-

ability is the installation of water meters. Water meter

tation

These

installation can be used as a better alternative to ﬁxed or

operational indicators are primarily taken from the lists of

tiered tariffs for water services. Therefore, tracking custo-

performance indicators put forth by IWA and IBNET

mer metering level/density indicates improvements to the

(Alegre et al. ; van den Berg & Danilenko ) and

water system.

improved

operating

procedures).

are described in the following sections. Improved system infrastructure

Implementation of improved procedures

There are various points in a water system where a water uti-

In response to operational and infrastructural weaknesses

lity may identify needed infrastructure improvements,

identiﬁed throughout the water system through the WSP

ranging from water sources, storage, treatment to distri-

process, the water utility should develop and implement

bution. The indicators discussed here were selected as

more streamlined methods of operation and risk reduction

examples because they can be measured simply, often with

(Gunnarsdottir et al. b). Therefore, the development

information that a water utility may already be collecting

and implementation of standard operating procedures is

for its own records. They also clearly demonstrate improve-

an indicator of positive operational change and marks the

ments in system infrastructure. Examples of water source

increased effectiveness of the WSP (Gelting et al. ).

protection include fencing, wellhead protection or legal pro-

This process includes both establishing clear standard oper-

tection of watersheds or recharge zones. Simple indicators

ating procedures at the outset of the WSP process if they do

for water treatment and storage are chlorine residual

not already exist, as well as updating and redeﬁning these

178

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

operating procedures over time, to reﬂect needs identiﬁed

satisfaction. Indicators related to complaints should be

during continued monitoring.

taken in context, however. As Alegre et al. () explain,

One way to measure this indicator is to track the

the absence of customer complaints may not always indicate

implementation and frequency of key operations within

that the utility is functioning well; in some cases, customers

the system, such as the number of inspections of physical

whose complaints are habitually ignored or unaddressed by

assets that are conducted, the number of required mainten-

the water utility may simply stop complaining because they

ance activities that take place, and the number of times

believe they hold no inﬂuence. Then, as a utility improves its

the system’s equipment is calibrated. It is possible to calcu-

operations, complaints may initially increase as customers

late these measures on a yearly basis, but this is ﬂexible

feel the utility is more responsive. Therefore, using customer

based on the capacities of the water utility staff.

complaints as an indicator of improved operating pro-

For some water utilities, an appropriate measure may be

cedures is only recommended if the WSP’s context is

the number of sanitary inspections that are conducted to

taken into consideration. It is important to analyze trends

examine water infrastructure, as this is a key activity for

in this indicator over time rather than at any discrete

reducing contaminant loads in water before treatment. In

point, so as to fully understand the pattern of customer com-

addition, the water utility can measure improved oper-

plaints within the context of that water utility (Alegre et al.

ational monitoring of water quality before treatment, at the

).

treatment works, and within the distribution system. Common operational monitoring parameters that are mea-

Financial outcomes

surable and simple include chlorine residuals, pH and turbidity. While using an array of microbiological, physical,

A WSP’s effectiveness is partly assessed by the ﬁnancial

chemical and radiological water quality tests is the ideal for

changes that occur within the water utility. Expected ﬁnan-

veriﬁcation of ﬁnished water quality, it may be that a utility

cial outcomes discussed in the CDC’s evaluation framework

has limited ﬁnancial and technical capacity to carry out

include reductions in costs, increased cost recovery due to

more complex monitoring exercises repeatedly. The WSP

clients’ greater willingness to pay for improved services,

team and water utility should explore the various options

and an increase in either local investments and subsidies

for testing water quality recommended by such institutions

or external donor support (Gelting et al. ). IWA and

as the WHO and IWA (Alegre et al. ; WHO ).

IBNET have compiled many standardized performance

The implementation of improved operations is also demonstrated by the increased documentation of stan-

measurement indicators related to ﬁnances. A selection are presented and discussed below.

dard operating procedures. Documenting changes to the operating procedures of the water utility not only increases

Cost savings

institutional knowledge among the water utility staff and other stakeholders, but it also provides a record that

An increase in cost savings is identiﬁed by the unit total

helps track the water utility’s operational progress over

costs, which indicates the costs of the water utility during

time.

a set period of time. Tracking this indicator during the

Improved operating procedures can be assessed by the number of customer complaints that the water utility

course of WSP implementation may help to identify costs savings resulting from the WSP process.

receives during a set period of time. This is measured, for example, by the number of service complaints received per

Cost recovery

connection in the water supply network. An increase or decrease in complaints can reﬂect the effectiveness of its

Tracking the operating cost coverage ratio allows a utility to

operations. Similarly, the percentage of written complaints

determine if their revenues cover operating costs, and track

that the water utility addresses can indicate an active com-

changes in cost recovery over time. As Gelting et al. ()

mitment to improving procedures and increasing customer

discuss in their conceptual framework for WSP evaluation,

179

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

cost recovery can also be inﬂuenced by a rise in customer

of best practice and begin to incorporate the WSP process

satisfaction due to an improved and more reliable water

into their guidelines and methodology.

supply system (Bhandari & Grant ; Gelting et al. ). An increase in the amount or the consistency with

WSP knowledge sharing and promotion

which customers pay for water services reﬂects this satisfaction. Similarly, payments from consumers could reﬂect a

WSP knowledge is shared among water utilities and other

level of dissatisfaction with the services provided by the

institutions as the WSP model gains recognition as a

water utility. Tracking the collection ratio over time pro-

viable option for high-quality water service delivery. This

vides insight into customer satisfaction and, subsequently,

process leads to what DiMaggio & Powell () refer to

cost recovery. These indicators are practical and feasible

as ‘mimetic isomorphism’, which includes the concept that

for evaluators to use because information about costs and

‘organizations tend to model themselves after similar organ-

revenues is most likely already collected by the water utility.

izations in their ﬁeld that they perceive to be more legitimate or successful’ (DiMaggio & Powell ). This acquired

Increased donor support and/or increased investment

understanding of WSP principles can be identiﬁed by tracking the increased awareness of WSPs among water utilities

Finally, unit investment is used to measure an increase in

in a country or speciﬁc region within a country. It should be

support of and investment in the water utility. Utilities in

noted that it can be difﬁcult for many utilities to assess WSP

Iceland reported improved success in obtaining resources

knowledge sharing and promotion, but this indicator is an

to improve water systems after implementing WSPs (Gun-

important measure of a shift towards more formalized

narsdottir et al. b). As previously mentioned, this

policy change. Evaluators can gather this information in a

support can include local investments and subsidies as

variety of ways, ranging from analyzing informal communi-

well as donations or loans from external sources.

cation such as email, letters and conversations to written

It should be noted that the outcomes discussed here refer only to ﬁnancial changes that occur within the water utility

records, media communication such as press releases, or through surveys or questionnaires.

and affect its ﬁnancial status. Financial changes that the utility’s clients experience as a result of the WSP are better

WSPs as norms of practice

classiﬁed as impacts, because they reﬂect socioeconomic effects on consumers. For example, access to an improved

Progress in policy change continues when more and more

and affordable water supply system can reduce consumers’

institutions begin to incorporate WSPs as a standard of

opportunity cost for such behaviors as buying bottled water

best practice. This change can be understood within the fra-

or traveling long distances to collect water from a safe

mework of Tolbert & Zucker’s () three-stage process for

source. As with other changes that directly affect consumers’

implementing a new practice; in this process, a practice

health or socioeconomic status, these ﬁnancial changes

moves from ‘pre-institutionalization’ to ‘semi-institutionali-

should be considered as overall impacts of the WSP rather

zation’

than outcomes, and are beyond the scope of this paper.

Summerill et al. () applied this process directly to the

and

eventually

‘full

institutionalization’.

WSP context by describing how the WSP model moves Policy outcomes

from pre-institutionalization, in which there is low awareness of WSP practice, to semi-institutionalization, in which

Policy outcomes are often the last to become apparent,

WSPs are relatively new but have become widely recognized

because policy change often happens gradually and in

and adopted by some – yet not all – water supply institutions.

stages. This process of policy change begins when WSP

This evolution of the WSP model into best practice – a move

knowledge is shared and promoted informally among

from pre-institutionalization to semi-institutionalization –

water utilities and other stakeholders, and continues when

can be measured by the number of WSP pilot projects

these institutions consider the WSP model as a standard

that are implemented, as well as by the existence of

180

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

WSPs being incorporated into non-regulatory guidance

of WSPs may occur, identifying them is beyond the scope of

documents for use by WSP implementers such as utilities,

measurement for most institutions implementing WSPs, and

non-governmental organizations and other stakeholders.

they are therefore not addressed in this paper. Future work could develop standard approaches to evaluate WSP impacts.

WSP formal regulatory requirements A more advanced stage of policy change becomes apparent

CONCLUSION

when formal regulatory requirements are developed for WSPs. Summerill et al. () suggest that buy-in for WSPs

With the increase in WSPs worldwide, there have been a

among water utilities is currently in Tolbert & Zucker’s

number of valuable manuals, guidelines and case studies

() semi-institutionalization stage at the global level,

developed to assist practitioners in the coordination and

and argue that the model has not yet moved to the ﬁnal

implementation of WSPs. Together, these resources make

stage of ‘full institutionalization’ in which the WSP practice

up a toolkit that service providers and WSP team members

is fully accepted and required in the water supply sector

can use in virtually every step of the process. For example,

worldwide (Summerill et al. ). Within individual

the WHO’s Water Safety Plan Manual guides water utilities

countries, however, it is possible for WSPs to move through

through organizing a WSP team and developing a WSP plan

all three stages of institutionalization. The incorporation of

(Bartram et al. ). Similarly, WHO’s Water Safety Plan-

WSP preventive concepts into national regulations indi-

ning for Small Community Water Supplies provides

cates that the concept of active risk prevention at every

guidance in WSP initiation and implementation speciﬁcally

stage within a water delivery network is included in a coun-

for community-managed small water systems (WHO ).

try’s national guidelines and regulations, even though

Some tools such as the WHO/IWA Water Safety Plan Qual-

utilities are not speciﬁcally required to develop a WSP.

ity Assurance Tool (WHO/IWA ) focus on WSP

Subsequently, the most advanced policy outcome is full

performance assessment and are designed speciﬁcally to

inclusion of the WSP model in the national drinking water

help water utilities track their progress and quality of

guidelines, at which point WSPs and their principles of

service.

risk prevention are required of all water utilities in the

The set of indicators proposed in this paper is by no

country. WSP policy adoption in drinking water regu-

means intended to replace these valuable tools. Rather

lations easily identiﬁes this ﬁnal level of policy change.

than suggesting another way to monitor performance, the set of indicators proposed in this paper contributes to the

Limitations

WSP toolkit by offering a way to evaluate WSP outcomes. In other words, the Quality Assurance Tool and other exist-

When considering these proposed indicators, it is important to

ing assessment instruments are designed for the water utility

take into account their limitations. As previously noted, the

to monitor its progress and ensure that it is consistently

indicators were designed to be applicable to any context in

meeting its quality standards, while the proposed indicators

which WSPs are implemented. This ﬂexibility allows the indi-

in this paper were compiled to help the water utility identify

cators to be adapted to ﬁt a particular WSP. While the

and measure the changes that should occur as a result of this

simplicity and adaptability of these indicators is necessary in

progress.

order to be relevant to any individual WSP, it could present a

These indicators that measure outcomes hold great

challenge for standardized benchmarking if the indicators

potential for the continued implementation and expansion

become signiﬁcantly altered in each context. In addition,

of WSPs worldwide. Having a deﬁned framework for evalu-

changes that occur and are tracked by these indicators may

ating the effectiveness of a WSP, along with a set of speciﬁc

not always be direct results of WSP implementation. An

and measurable indicators by which to carry out that evalu-

additional limitation is that, as previously discussed, these indi-

ation, will help implementers assess key WSP outcomes

cators were only selected to evaluate outcomes. While impacts

internally, as well as benchmark their progress against

181

G. Lockhart et al.

Development of indicators for measuring outcomes of water safety plans

other WSPs in their region and globally. It is helpful to note that collaborations among WSPs already exist in various contexts worldwide; for example, there are WSP networks located in Latin America, Africa and the Asia-Paciﬁc region. These networks exist to provide support, share innovations and facilitate cooperation among WSP implementers. They could also become a platform for scaling up internal WSP evaluation activities and promoting increased accountability and benchmarking among WSPs worldwide. It is our hope that these proposed outcome indicators will be a helpful tool in facilitating the evaluation of WSPs worldwide.

REFERENCES Alegre, H., Baptista, J. M., Cabrera Jr., E., Cubillo, F., Duarte, P., Hirner, W., Merkel, W. & Parena, R.  Performance Indicators for Water Supply Services. 2nd edn., IWA Publishing, London, UK. Bartram, J., Corrales, L., Davidson, A., Deere, D., Gordon, B., Howard, G., Rinehold, A. & Stevens, M.  Water Safety Plan Manual: Step-by-Step Risk Management for DrinkingWater Suppliers. World Health Organization, Geneva. Bhandari, B. & Grant, M.  User satisfaction and sustainability of drinking water schemes in rural communities of Nepal. Sustainability: Science, Practice & Policy 3 (1), 12–20. DiMaggio, P. & Powell, W.  The iron cage revisited: institutional isomorphism and collective rationality in organizational ﬁelds. American Sociological Review 48 (2), 147–160. Dyck, A., Exner, M. & Kramer, A.  Experimental based experiences with the introduction of a water safety plan for a multi-located university clinic and its efﬁcacy according to WHO recommendations. BMC Public Health 7 (34). Available from: http://www.biomedcentral.com/1471–2458/7/34.

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

Gelting, R. J., Delea, K. & Medlin, E.  A conceptual framework to evaluate the outcomes and impacts of water safety plans. Journal of Water, Sanitation and Hygiene for Development 2 (2), 103–111. Gunnarsdottir, M. J., Gardarsson, S. M., Elliot, M., Sigmundsdottir, G. & Bartram, J. a Beneﬁts of water safety plans: microbiology, compliance, and public health. Environmental Science and Technology 46, 7782–7789. Gunnarsdottir, M. J., Gardarsson, S. M. & Bartram, J. b Icelandic experience with water safety plans. Water Science and Technology 65.2, 277–288. Provan, K. G. & Milward, H. B.  Do networks really work? A framework for evaluating public-sector organizational networks. Public Administration Review 61 (4), 414–423. Summerill, C., Smith, J., Webster, J. & Pollard, S.  An international review of the challenges associated with securing ‘buy-in’ for water safety plans within providers of drinking water supplies. Journal of Water and Health 8 (2), 387–398. Tolbert, P. S. & Zucker, L. G.  The institutionalization of institutional theory [Electronic version]. In: Handbook of Organization Studies (S. Clegg, C. Hardy & W. Nord, eds). SAGE, London, pp. 175–190. van den Berg, C. & Danilenko, A.  The IBNET Water Supply and Sanitation Performance Blue Book: The International Benchmarking Network for Water and Sanitation Utilities Databook. The World Bank, Washington, DC. Vieira, J.  A strategic approach for water safety plans implementation in Portugal. Journal of Water and Health 9 (1), 107–116. World Health Organization  Guidelines for Drinking-Water Quality, 4th edn., World Health Organization, Geneva. World Health Organization  Water Safety Planning for Small Community Water Supplies: Step-by-Step Risk Management Guidance for Drinking-Water Supplies in Small Communities. World Health Organization, Geneva. World Health Organization/International Water Association  User Manual: Water Safety Plan Quality Assurance Tool. World Health Organization, Geneva.

First received 20 March 2013; accepted in revised form 2 August 2013. Available online 21 October 2013

182

Practical Paper

04.1

2014

Practical Paper Rights-based indicators regarding non-discrimination and equity in access to water and sanitation Urooj Quezon Amjad, Georgia Kayser and Benjamin Mason Meier

ABSTRACT The declaration of the Human Right to Water and Sanitation by the UN General Assembly in 2010 is a significant achievement. Successfully implementing, enforcing and monitoring this right in various countries, contexts and scales will be an even greater achievement. Facilitating accountability for water and sanitation as a human right, The Water Institute at the University of North Carolina at Chapel Hill and the United Nations Office of the High Commissioner for Human Rights hosted a workshop to examine the translation of public health data into human rights indicators. Experts in water and sanitation policy and human rights law discussed and debated the challenges and ways forward in developing indicators for equity and non-discrimination that would assist with meaningful practice of the Human Right to Water and Sanitation. The discussion was framed by three main questions: what should be measured, where to find this information, and what additional information is needed. The workshop was convened at The Water Institute’s annual conference, Water and Health: Where Science Meets Policy on October 4, 2011. Key words

| equity, human right to water and sanitation, indicators, public health, water quality

Urooj Quezon Amjad (corresponding author) Post-doctoral Research Associate, The Water Institute, Gillings School of Global Public Health, and Research Fellow, Global Research Institute, University of North Carolina, Chapel Hill, Rosenau Hall, CB #7431, 135 Dauer Dr. Chapel Hill, North Carolina 27599-7431, USA E-mail: amjad@unc.edu Georgia Kayser Post-doctoral Research Associate, Department of Public Policy and The Water Institute at UNC, CB#7431, University of North Carolina, Chapel Hill North Carolina USA Benjamin Mason Meier Assistant Professor of Global Health Policy, Department of Public Policy, University of North Carolina, Chapel Hill, 218 Abernethy Hall, CB #3435, Chapel Hill, North Carolina 27599-3435, USA

INTRODUCTION With the human right to water and sanitation gaining rapid

obligations, thereby holding governments accountable for

acceptance under international law and framing global gov-

realizing the right to water and sanitation. Central to legal

ernance efforts to improve water and sanitation, the

obligations for the human right to water and sanitation,

implementation of this right will require corresponding

this indicator development has begun with a focus on

accountability frameworks to assess national human rights

equity and non-discrimination.

realization. Facilitating this implementation, the United

Attended by approximately 50 human rights, public

Nations has begun a process to identify speciﬁc indicators

policy, and water and sanitation professionals, this three-

reﬂective of the progressive realization of human rights

hour workshop on ‘Human Rights-Based Indicators: Non--

norms, distinguishing human rights indicators from develop-

Discrimination and Equity in Access to Water and

ment goals and laying the groundwork for global water

Sanitation’ encouraged participants to examine and assess

governance through international treaty monitoring. By

existing data and data collection methods, that would be

drawing on the process of developing indicators for other

amenable to application as rights-based indicators speciﬁc

human rights, this indicator development project seeks to

to equity and non-discrimination. To provide background

create a means of monitoring and evaluating rights-based

doi: 10.2166/washdev.2013.041

rights-based

principles

and

prepare

workshop

183

U. Q. Amjad et al.

Human rights-based indicators in water and sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

participants for interdisciplinary group discussions, the

identiﬁed in Berlin included exploring water supply and sani-

workshop began with an opening panel presentation to:

tation indicators, proposing indicators for equity, exploring

(1) introduce the human right to water and sanitation; (2)

standards for rural and urban areas, and reﬁning and improv-

deﬁne rights-based principles of equity, equality and non-dis-

ing the current system of monitoring (WHO and UNICEF

crimination; (3) outline human rights indicators and the

).

indicator development process; and (4) review indicator development efforts for water and sanitation. Thereafter working in independent groups – with representation in

EQUITY, NON-DISCRIMINATION AND EQUALITY

each group from service operators, policy makers, civil society, and water professionals – ﬁve groups examined

As a basis to propose indicators for equity, the principles of

the data reﬂective of the implementation of the right to

‘non-discrimination and equality are linked: under human

water and sanitation before presenting their individual

rights law, states must ensure that individuals do not suffer

group conclusions for comment by the full workshop.

discrimination, and that they can enjoy equality’. Where human equality is the bedrock principle of human rights law,

THE INDICATOR DEVELOPMENT PROJECT FOR WATER AND SANITATION

‘the principle of equality before the law does not mean … absolute equality, namely the equal treatment of men without regard to individual, concrete circumstances,

The human right to water and sanitation is derived under

but it means … relative equality, namely the principle to

international law from a 2003 General Comment on the

treat equally what are equal and unequally what are

International Covenant on Economic, Social and Cultural

unequal … To treat unequal matters differently according

Rights and the 2010 UN General Assembly Resolution on

to their inequality is not only permitted but required

the Human Right to Water and Sanitation. Incorporating

(Satterthwaite )’.

human rights in the development agenda reafﬁrmed that a debate on appropriate goals, targets and indicators is under

Useful to developing indicators for the implementation of

way in preparing for the 2015 assessment of the Millennium

the human right to water and sanitation is the concept of

Development Goals, and that there will be opportunities to

‘substantive equality’, which moves beyond formal equality

integrate the human right to water and sanitation in the

to require afﬁrmative action to rectify inequity under

Post-2015 Agenda. From an initial May 2011 Consultation

human rights law. In contrast to equality, the deﬁnition of

on Post-2015 Monitoring of Drinking-Water and Sanitation

equity usually refers to economic barriers, in which monitor-

in Berlin, the WHO/UNICEF Joint Monitoring Programme

ing helps us to understand who is left behind, focusing on

for Water Supply and Sanitation (JMP) provided a framework

the most disadvantaged members of society. Viewed

for future efforts to implement the human right to water and

through a human rights lens, equity can be fruitfully paired

sanitation. In carrying out these next steps, the consultation

with equality, creating a legally binding obligation to

stakeholders concluded that: (1) global goals, targets and

ensure that everyone – regardless of status, race, sex, class,

indicators are necessary for human rights implementation;

caste or any other factor – has access to safe water and sani-

(2) more attention should be directed to ensuring that water

tation (WHO and UNICEF ).

and sanitation systems respond to information needs of decision-makers at national, regional and international levels; and (3) global and national monitoring systems

MEASURES VS. INDICATORS

should feed into national decisions on overall resource allocations, targeting of services and selection of interventions.

Clarifying issues of equity through indicators, indicators for

Providing an imperative to bring together the water and sani-

non-discrimination and equality could be linked speciﬁcally

tation community around human rights norms, the next steps

to affordability, as it is one of the key barriers to achieving

184

U. Q. Amjad et al.

Human rights-based indicators in water and sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

equality and equity. With the water sector examining and

right to water and sanitation: access, quality, quantity, avail-

categorizing indicators based on a framework of structure,

ability, priority of use and user friendliness. The question of

process and outcome, advocates have focused on separate

what should be measured in developing appropriate and

qualitative and quantitative indicators to assess state com-

practical indicators for implementing the Human Right to

mitments, efforts and results (Roaf et al. ). In

Water and Sanitation is important because these character-

examining these indicators under the human right to water

istics will drive policy and practice efforts. Suggested areas

and sanitation, the preferences and needs of vulnerable

to measure regarding water and sanitation were discussed

groups must be directly linked to speciﬁc indicators for the

as well as areas for further consideration (see Box 1).

human right to water and sanitation. However, the needs

Many of the indicators for water and sanitation – water qual-

of vulnerable groups are not adequately highlighted and dis-

ity, water quantity, water availability and sanitation safety

aggregated data are not consistently collected. For example,

and waste treatment – impact health and could be compared

regulators need to understand the role of affordability, but

across groups of varying class, race, ethnicity, gender, dis-

they tend to gather information only for those customers

ability, sexual orientation, social, economic and political

who have a connection. Indicators used in monitoring will

opportunity, reﬂecting different levels of wealth, acceptance

need to be tailored to local circumstances and feasible in

and/or prestige. Some of the indicators recommended are

national level monitoring efforts, with non-governmental

direct measures of discrimination. This demonstrates how

organizations bearing a key role in rights-based monitoring.

health determinants can be translated into human rights indicators. Not all possible indicators were mentioned, but many factors that affect the development of a list of indi-

SMALL GROUP DISCUSSIONS ON THE DEVELOPMENT OF RIGHTS-BASED INDICATORS

cators for measuring equity and non-discrimination were discussed. These factors included: data disaggregated by geographic area, gender, vulnerable group, caste or racial group

As a participatory workshop incorporating the water sector in

and health status, among others.

indicator development efforts, ﬁve groups of participants discussed and shared their perspectives on key questions in the implementation of the human right to sanitation and water. The members of each group sought to develop answers to

WHAT MEASUREMENTS ARE AVAILABLE TO MONITOR EQUITY AND DISCRIMINATION?

three key questions: (1) What should be measured for nondiscrimination and equity? (2) What measurements are avail-

Existing datasets could be harnessed to develop indicators

able to monitor equity and discrimination? and (3) What

and sources for monitoring instead of ‘re-creating the

additional information is necessary for rights-based indi-

wheel’ of newer datasets. Identifying where and what is

cators? Working groups reported their results to all

readily available for the complicated task of developing prac-

participants (summarized in Boxes 1, 2 and 3, respectively),

tical indicators would be more efﬁcient and strategic. The

with the synthesis of these ideas facilitating the development

water and sanitation datasets shown in Box 2, suggested by

and implementation of indicators for measuring, monitoring

workshop participants, could be matched to human rights

and enforcing the human right to water and sanitation.

indicators for achieving non-discrimination and equity in the context of the human right to water and sanitation.

WHAT SHOULD BE MEASURED FOR NONDISCRIMINATION AND EQUITY?

WHAT ADDITIONAL INFORMATION IS NECESSARY FOR RIGHTS-BASED INDICATORS?

The following themes were suggested by workshop particinon-

Identifying what should be measured and where to ﬁnd such

discrimination and equity in the context of the human

existing information is not sufﬁcient in an effective indicator

pants

concrete

indicators

for

assessing

185

U. Q. Amjad et al.

Box 1 |

Human rights-based indicators in water and sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

What should be measured for non-discrimination and equity?

Water

•

Access – population ratio of access to use ○ Time to source. ○ Distance to source.

• • • • • •

○ Access by socioeconomic group. Quantity (liters/person/per day). Quality – Microbiological and chemical components of water quality: at the source, after ﬁltration, at consumption. Availability (h/day/week). Priority of use (discrimination at point of source and use). User friendliness (disabilities, children, pregnant women). Participation in decision making.

Sanitation

• • • • • •

Distance from home. Privacy. Safety. Community sanitation facilities. Disposal/treatment.

User friendliness. For consideration

• • • • • • • • • • • • • •

Definitions: Before indicators may be defined, definitions are needed for non-discrimination – process-oriented, and equality vs. equity and they need to be outcome-oriented. Target audience: Need to differentiate between country and regional-level and global-level monitoring. Imposing a strict global template for country and regional-level surveys prohibits useful outside-the-box data from being utilized. Geography: Urban/rural and disaggregate peri-urban. Affordability: Percentage of income spent on water by wealth quintiles; existence of social tariff; breakdown of cut off rates. Rules of law: Are fair laws on the books? Are the rules respected and enforced? Unit of measurement: per capita might be better than by household. Single indicator: e.g. maternal education, age Vulnerable groups: How vulnerable households are defined when considering members who are chronically ill, disabled, or non-biological children via adoption or foster care. Access to water outside and inside the home: Difficulty of getting to water source; access to water within the household between women and men. Burden of access: Access to safe water in relation to other responsibilities. Socioeconomic status: Identifying who will get a filter and who will use it according to socioeconomics. Epidemiological data stratified by caste: not only quantity but quality. Health status: Data for access by health status (e.g. individuals with HIV/AIDS have different water needs). Accuracy of measurements and testing: Where water quality testing takes place affects accuracy of test results.

Disease outbreaks: number of diarrheal disease outbreaks in a year, and similar water and sanitation related diseases.* * Note: This comment did not directly emerge from the workshop, although relevant, and was suggested by one of the reviewers of the manuscript.

186

U. Q. Amjad et al.

Box 2 |

• • • • • • • • • • • • • •

Human rights-based indicators in water and sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

What measurements are available to monitor equity and discrimination?

Demographic and health surveys: Urban/rural; wealth quintiles; religion; gender of household head; age. Multiplied indicator and cluster survey. Living standards measurement study: World Bank data on household income, expenses, and facilities can be found. Core welfare indicator questionnaire. Access to improved and unimproved water sources & JMP National water quality data. Water safety plans. National water and sanitation policies and legislation. Children’s rights – UNICEF and project level. Human rights to education and health. Laws and policies (instruments on legislations, strategies, regulations). Tariff structures and levels. National water and sanitation plans. World Bank asset index measured by socioeconomic quintile.

development process. New sources of data and themes are

the national level and compared globally, isolating water

also needed. In facilitating rights-based accountability, it is

quality measures that can be collected in a range of settings,

necessary to develop data sources that reﬂect human

locating measures for affordability that incorporate the

rights realization, with targeted exploration of water and

complexities of household income and unpaid work, identi-

sanitation to understand what cannot be assessed through

fying how to measure reliability and sustainability at scale,

current measures. Box 3 lists questions suggested by work-

and ﬁnding ways to measure the progress of the most discri-

shop participants that will be necessary to identify new

minated in each society, speciﬁcally, illegal inhabitants.

sources of data for assessing non-discrimination and equity

Looking toward the future of global water and sanitation

in the context of the human right to water and sanitation.

policy and how the UN General Assembly’s ofﬁcial recognition of the right to water and sanitation will inﬂuence the discussions surrounding the Post-2015 Agenda, one of

NEXT STEPS IN IDENTIFYING INDICATORS TO FULFILL HUMAN RIGHTS OBLIGATIONS

the remaining questions is ‘What should the post 2015 development agenda take account of – what should be our new goal with regard to water and sanitation?’ Further-

The future development of indicators and implications of

more, practical goals matched with practical indicators that

the new UN General Assembly Resolution on the human

assist with effective implementation of the Human Right to

right to water and sanitation ‘require that these [rights]

Water and Sanitation would be key. In assisting policy

are accessible, available, affordable, acceptable and safe.

makers to understand the data that is reﬂective of human

Indicators in accordance with this understanding will

rights norms, the thinking developed in Chapel Hill will

bring us towards a more holistic understanding [of water

be instrumental to developing indicators for the human

and sanitation policy]’ (de Albuquerque ). The difﬁcult

right to water and sanitation, facilitating accountability for

issues that need to be addressed in indicator development

the realization of human rights through the water and sani-

include: identifying indicators that can be aggregated to

tation sectors.

187

U. Q. Amjad et al.

Box 3 |

• • • • • • • • • • • • • •

Human rights-based indicators in water and sanitation

Journal of Water, Sanitation and Hygiene for Development

04.1

2014

What additional information is necessary for rights-based indicators?

Who are the duty bearers? Whether rights-based considerations are included in the national legislations regarding water and sanitation plans. Percentage of budget and of GDP currently devoted to water and sanitation infrastructure. More extensive information about community stand-pump use (size of community; social and cultural context to access). Regions prone to natural disasters; regional seasonality fluctuation information (create proactive solutions). Most disadvantaged communities and how they were disproportionately affected and take measures to protect those communities in the future. Develop definitions of who is most at risk and disaggregate data accordingly. In relation to three types of indicators in mind (structural, process and output), the JMP could be used for output indicators. Gendered roles and ability to participate in policy change. Disaggregated data by sex, and process indicators. Much of the data is tied to households, so alternatives need to be developed for those who do not have secure tenure, such as slum dwellers. Clarifying how access to water can be achieved by slum dwellers is important. How to coordinate local and global indicators to make sure they are compatible. Context is hard to measure. Currently we are not able to develop context-based indicators, and no meta-data are currently available. Global datasets have become so expensive to build, maintain and use. The availability of opportunities for participation is difficult to measure. Ability of marginalized groups to impact decision making is even more difficult to measure.

ACKNOWLEDGMENTS Organized by the UK ofﬁce of the High Commissioner for Human Rights and the Water Institute at the University of North Carolina at Chapel Hill.

REFERENCES de Albuquerque, C.  Special UN Rapporteur on Human Right to Water and Sanitation, Public Presentation.

Roaf, V., Khalfan, A. & Langford, M.  Monitoring Implementation of the Right to Water: A Framework for Developing Indicators. Satterthwaite, M.  Public presentation referencing Judge Tanaka (dissenting), South West Africa case, International Court of Justice. WHO and UNICEF  First Consultation on Post-2015 Monitoring of Drinking-Water and Sanitation. 3–5 May, WHO and UNICEF Joint Monitoring Programme, Berlin. WHO and UNICEF  JMP Working Group on Equity and NonDiscrimination. Final Report, August 2, available at: http:// www.wssinfo.org/ﬁleadmin/user_upload/resources/JMPEND-WG-Final-Report-20120821.pdf.

First received 7 December 2011; accepted in revised form 1 August 2013. Available online 21 October 2013