Health Dialogue - Inside the frontline of the mental health crisis by AssociationofSalariedMedicalSpecialists

Inside the frontline of the mental health crisis

HEALTH DIALOGUE ISSUE 18 | NOVEMBER 2021

Dr Charlotte Chambers, ASMS Director of Policy and Research Queries about this research can be sent to charlotte.chambers@asms.org.nz

Contents Foreword

By the numbers

Introduction

Issues facing psychiatry

The psychiatry workforce

Methods

Outcome measures

Predictor variables

Analytical procedures

Quantitative analysis

Regression analysis

Qualitative analysis

Results

Participants

International medical graduates

Main outcome measures

Burnout

Stress at work

Levels of workplace demands and peer and managerial support

Intentions to leave

Job satisfaction

Time, resourcing, and workloads

Perceptions of workload change over time

Correlations between variables and main outcome measures

Demographic variables

Regression analysis

Discussion

Indicators of wellbeing

Risk factors

Changes since 2018

Job satisfaction and intentions to leave

Conclusion

Recommendations in summary

References

Appendix 1

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 1

2 HEALTH DIALOGUE NOVEMBER 2021

Foreword Aotearoa New Zealand is in the midst of a mental health crisis. Numerous reports have detailed the challenges facing the mental health system for both those seeking care and those providing services. In our recently released report What Price Mental Health? The Crisis and the Cure, ASMS details the story of rising demand, reduced bed capacity and clinically stressed services. What is less well known is how those who provide these crucial mental health services are faring. We know that along with the majority of the public health workforce, psychiatrists continue to put their own personal wellbeing on the line to ensure high quality mental health services are provided to those who need it. But they are struggling. Anecdotally, we have heard that psychiatrists feel overburdened, stressed, and stretched, and in some instances, unsafe in the course of their work. To date, however, there is little hard data as to the specific nature of this stress, how it relates to their workloads, and how these workloads have changed over time. Moreover, there is need for more research on the consequences of a workforce facing high patient demand and stressful working conditions.

In 2018, New Zealand had the lowest number of practising psychiatrists per capita of a group of 11 countries, including Australia, the UK and Canada. Our system is heavily reliant on international medical graduates and their retention rates are poor. This research sets out in detail the measures of wellbeing of our psychiatry workforce, the degree to which psychiatrists feel supported at work, the demands they are facing in terms of workload pressures and acuity, and the possible consequences in terms of their intentions to leave. The research uses qualitative data throughout to illustrate what these indicators mean in terms of individual impact. Overall, this research paints a picture of a critical workforce facing significant stress, swelling demand, and expressing significant concern at its capacity to keep going. Despite this, the research highlights the commitment and professionalism of this vital workforce and the determination and commitment of psychiatrists to keep providing high quality care for New Zealanders in need.

Given the tsunami of unmet need facing mental health, we cannot afford to lose more doctors from psychiatry.

Sarah Dalton Executive Director Association of Salaried Medical Specialists

“We often feel like patients are being discharged to the community to fail. This failure takes the form of suicide, homicide, estrangement & homelessness”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 3

By the numbers This report describes and discusses the findings of a survey of psychiatry members of the Association of Salaried Medical Specialists Toi Mata Hauora. Seventy percent (368/526) responded to the survey.

Who are our psychiatrists?

Key findings

4 HEALTH DIALOGUE NOVEMBER 2021

50%

65%

44%

Male

Aged over 50

Pākehā

37%

60%

Working in adult mental health

International medical graduates, of which 52% have been in New Zealand for 15 years or more

35%

45%

87%

Report high levels of burnout

Strongly agree or agree that they would leave their current job if able

Disagree or strongly disagree that they are working in a well-resourced mental health service

64%

41%

Rarely or are never able to access the recommended non-clinical time

Always or usually end up covering other colleagues’ caseloads

In the past three years

95%

86%

77%

76%

Report a significant increase or an increase in demand for specialist mental health services

Report a significant increase or an increase in complexity of caseload

Report a significant increase or an increase in demands of on-call workloads

Report a significant increase or an increase in size of caseload

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 5

Introduction Three years ago, the New Zealand Mental Health and Addiction Inquiry released its report into the state of mental health services in New Zealand. Somewhat optimistically, the report stated that “… demand for specialist services [is expected to] reduce as issues are dealt with earlier, before they escalate.” Fast forward to 2021, however, and national statistics suggest demand for mental health and addiction services (MHA) has not reduced. The 2019/20 New Zealand Health Survey reported that the prevalence of very high or high psychological distress has increased to 7.4% of adults since 2011/12 when it was 4.5% (MoH 2021). In contrast, however, numbers of staff, acute beds and availability of respite care are not increasing concurrently. For the medical specialists at the coalface of this mental health crisis, these factors combined represent stressful and challenging conditions of work with no signs of positive change. This research explores how our psychiatrists are faring. It focuses on core dimensions of their psychosocial wellbeing and attempts to quantify their work demands. It complements an earlier ASMS publication which presented the figures behind the mental health crisis while proposing potential solutions. As this previous research substantiated, mental health service provision in New Zealand is in crisis with demand for mental health services estimated to be higher than the growth in DHB-employed psychiatrists. Furthermore, despite the number of mental health and addiction patients growing well above the rate of population growth in New Zealand, the number of available inpatient beds per population has fallen by nearly 10% in the last five years (ASMS 2021). Perhaps most concerning is that funding for mental health in New Zealand remains set for an estimated 3% of the population having a severe need for mental health services each year while the evidence indicates that closer to 5% of the working-age population in New Zealand has a severe mental health condition. These issues are neither particular to New Zealand, nor are they new. A recent study focusing on

6 HEALTH DIALOGUE NOVEMBER 2021

mental health services in Australia noted that “the spectres of inadequate staffing, facilities and resources lour over everyday practice and contribute to an enduring sense of futility of effort. It is often simply not possible to provide optimal psychiatric care within existing public mental health services. Problems abound at all levels” (Looi and Maguire 2019 p635). Research on New Zealand psychiatrists from 2004 highlighted issues with the differing priorities of administrators and psychiatrists working towards mental health provision. These tensions between financial prerogatives and patient-centred care were postulated as contributing to issues with the retention of psychiatrists, particularly those who were trained overseas (Snyder and Kumar 2004). Subsequent studies on New Zealand psychiatrists throughout the mid-2000s documented issues with burnout (Kumar, Fischer et al. 2007) and prefaced issues with growing patient demand, reduced administrative support and burgeoning administrative protocols (Fischer, Kumar et al. 2007). As summarised by Kumar, Hatcher et al. (2010), psychiatrists are prone to experiencing burnout, stress and suicidal ideation. They note four key factors associated with psychiatrist burnout: too much work, working long hours, an aggressive administrative environment, and a dearth of support from management. A follow-up study into rates of burnout suggested rates of depersonalisation (a key facet of burnout as measured by the Maslach Burnout Inventory (MBI) were on the increase (Kumar, Sinha et al. 2012). Despite this, the authors noted burnout overall could be mitigated by degree of job satisfaction, particularly the dimensions of task significance and receiving positive feedback.

In 2015 psychiatrists were found to have the second highest rate of patient-related burnout in a nationwide study conducted by ASMS (Chambers, Frampton et al. 2016). In a follow-up study in 2020 psychiatrists were found to have rates above the survey average for all three measures of burnout, but no significant changes to the mean rates of burnout when compared to the 2015 study (Chambers 2021). To the best of the author’s knowledge, there have been no further recent studies explicitly focused on the degree of burnout, stress and workplace demands experienced by psychiatrists in New Zealand. There have also been no recent studies combining quantitative indicators of stress with qualitative illustrations of the lived experiences of working in a stretched and strained mental health service. This research seeks to fill this gap.

Issues facing psychiatry Psychiatrists serve as cornerstones of the teams responsible for the delivery of mental health services. High quality mental health care in turn depends on healthy, well supported mental health workforces. Psychiatry requires therapeutic relationships based on the establishment of highquality interpersonal relationships and rapport (Priebe and McCabe 2008). It requires clinicians to interface with patients with affective and cognitive empathy and understanding (Gunasekara, Patterson et al. 2017, Singh, Karanika-Murray et al. 2020). For psychiatrists to deliver such highquality patient-centred care, having adequate time for meaningful engagement with both patients and their families is vital (Fukui, Salyers et al. 2021). As articulated by Rotstein, Hudaib et al. (2019), “intense interactions with patients … are not simply a hazardous component of psychiatric

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 7

practice … [they are] both a diagnostic tool and a therapeutic intervention” (p252-253) (see also Priebe and McCabe 2008). Because of this requisite working relationship and the types of patients that psychiatrists treat, psychiatry is known to be a stressful and demanding specialty. As articulated by Maslach and Leiter (2016), psychiatrists deal with extremely challenging circumstances including patients who may be violent, and/or experiencing mental distress and trauma. The risks and stressors facing psychiatrists can include physical assault, aggression, and violence from patients (Ezeobele, Mock et al. 2020), all of which can have a negative impact on the wellbeing of mental health providers (Kavanagh and Watters 2010). The challenging demands posed by psychiatric patients can result in psychiatrists having a greater propensity for developing burnout, particularly due to the emotionally draining nature of encounters (Kumar 2007, Maslach and Leiter 2016). Research into rates of compassion fatigue experienced by mental health professionals emphasises that high workloads, challenging clinical settings and direct exposure to trauma are all likely to elevate psychiatrists’ propensity for burnout (Singh, Karanika-Murray et al. 2020). Other research emphasises unrealistic demands from patients and their families as having the capacity to result in elevated levels of exhaustion and cynicism (Maslach and Leiter 2016). In turn, research by Johnson, Hall et al. (2018) emphasises the risks to the quality of patient care should those with responsibility be themselves suffering mental distress. Research to date has noted that psychiatrists suffering from burnout are more likely to have negative feelings regarding their patients, poorer communication, and higher levels of avoidant behaviour (Holmqvist and Armelius 2006, Fischer, Kumar et al. 2007).

The psychiatry workforce Official statistics provided by the Health Workforce Information Programme (HWIP) at Technical Advisory Services Limited (TAS) estimate that the DHB-employed workforce increased by 22%

8 HEALTH DIALOGUE NOVEMBER 2021

between 2011 and 2020 (TAS unpublished data). Over the same period, however, MHA patients increased by an estimated 28%. Moreover in 2019, New Zealand had the lowest number of practising psychiatrists per capita when matched with 10 other comparable OECD countries. New Zealand also has a high rate of reliance on international medical graduates (IMGs) to staff specialist mental health services. The current data suggests that in the five years from 2015 to 2019, 78% of new psychiatrist vocational registrations were doctors from overseas. The heavy reliance on IMGs is a key feature of the New Zealand medical workforce, and psychiatry has the highest proportion of IMGs of all the medical specialties. Latest statistics from the Medical Council of New Zealand estimate that the overall proportion of IMGs in psychiatry sits at around 60%. This trend is consistent over time; nearly twenty years ago Snyder and Kumar (2004) noted that understanding the experiences of IMGs in psychiatry was key to reducing the heavy reliance on locum providers in psychiatry as well as the tendency for IMGs to leave New Zealand after a relatively short period of time. ASMS has previously warned of the risks of an overall medical workforce with a high reliance on IMGs, particularly given that competition for IMGs is likely to escalate as countries struggle to fill vacancies (ASMS 2017). Given the growing demand for mental health services outlined by ASMS in the accompanying research brief, it is imperative that there is a good level of insight into the experiences of IMG psychiatrists in New Zealand particularly in terms of what factors attract them and how they may better be retained. International research suggests that the number of trainees choosing to enter into psychiatry is declining; UK research from 2017 describes an apparent recruitment crisis in psychiatry due to stigma towards psychiatry as a specialty, perceived levels of bureaucracy in the profession and poor morale (Choudry and Farooq 2017). The World Psychiatry Association in the same year also notes that there is a recruitment crisis where vacancy

rates in training in many countries are over 10% (Shields, Ng et al. 2017). More recent research has reiterated concerns with the growing demand for psychiatric care, and the challenges for trainees posed by COVID-19-related lockdowns and associated changes to training programmes (Greig 2020). While rates of recruitment into psychiatry in the UK remain stable at around 4%–5%, the demand for mental health services is continuing to increase worldwide (Goldacre, Turner et al. 2018). Moreover, the psychiatry workforce is also ageing, with fewer registrars to replace those reaching retirement. Latest statistics from the University of Otago Medical School report on medical school graduate intentions finds psychiatry 8th overall of 10 broad specialty choices (MSOD 2021). Other research notes the risk of losing existing psychiatrists due to poor satisfaction with working conditions and growing rates of burnout (Broderick, Vaughan et al. 2021). For psychiatry specifically, Morgan, Finn et al. (2021) note that in the UK, high levels of vacancies for psychiatry posts have led to a high reliance on locum roles, which can have significant negative consequences for the quality of patient care and safety. In summary, not only is psychiatry understaffed, facing challenges with recruitment, and facing a growing demand for service, but issues with

burnout pose risks for attrition from the profession as well as having the capacity to negatively affect the quality of mental health care. Given the national and international context, this Health Dialogue seeks to address the following research objectives: • investigate the psychosocial wellbeing of psychiatrists working in the public health system in New Zealand • identify the main risk factors for work-related stress • gauge perceptions of how workload (quantity and quality) has changed over time, and • assess degree of job satisfaction and whether individuals intend/desire to leave their work.

“We do not have enough staff or resources to retain staff, the staff around me are burnt out, unmotivated and it is painfully obvious”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 9

Methods All 526 members of the ASMS with psychiatry and psychiatry sub-specialties specified as their medical field were invited to participate in the study in May 2021. No distinction was made between the participants who were specialists or medical officers. There was representation from all DHBs and all sub-specialties of psychiatry specified. Data was collected using an online survey hosted by SurveyMonkey. The full survey is included in Appendix 1. This survey was open for three weeks (2–23 May 2021) and weekly reminders were sent to encourage participation. No incentives for participation were provided. It is important to note the timing of this research; the survey was distributed in May 2021, a year after the first nationwide lockdown in New Zealand, but prior to the current spread of the Delta variant and the lockdowns ongoing at the time of writing. The impact of COVID-19 was not a key focus of this research, but it is worth considering how recent developments – the spread of COVID-19 and repeated lockdowns – may be contributing to the demand for mental health services even more than is captured by this research.

Outcome measures The main outcome measures were degree of burnout and degree of stress experienced at work. Burnout was measured using the single item measure of emotional exhaustion from the Maslach Burnout inventory, namely the question: “I feel burned out from my work” (7-point Likert scale ranging from Never to Daily). High levels of burnout were defined as occurring at least once a week or more, or a score of 5 or higher on the Likert scale (West, Dyrbye et al. 2012). The single item non-proprietary method of assessing burnout without needing to utilise either the full MBI or the CBI is recommended in the literature when accompanied by other measures of workforce stress and workplace demands (see for example West, Dyrbye et al. 2009, Dolan, Mohr et al. 2015).

10 HEALTH DIALOGUE NOVEMBER 2021

Stress at work was also measured using a single item question, specifically “In general, how do you find your job?” (5-point Likert scale ranging from not at all stressful to extremely stressful). This measure of job-related stress relies upon respondents’ understandings of the term stress and, while limited, again provides a validated overall indication of job stressfulness (Houdmont, Jachens et al. 2019). Supplementary outcome measures were taken from the UK Health and Safety Executive Management Standards (HSE) survey with the specific measures of perceived level of workplace demands, and levels of peer and managerial support included (HSE 2012). The HSE provides a validated assessment of issues such as workload, work patterns and the work environment, and support, which addresses the encouragement, resources and support provided by employers and colleagues (Cousins, MacKay et al. 2004). Average HSE scores were benchmarked against averages determined in a UK study (Edwards, Webster et al. 2008) and a previous survey on ASMS members (Chambers, Frampton et al. 2018). The study also assessed degree of job satisfaction, measured with a single item question: “Considering everything, how satisfied are you with your job?” (5-point Likert scale ranging from 1 = very dissatisfied to 5 = very satisfied) and two questions probing turnover intentions: “If I were able, I would leave my current job” and “I plan to leave my job within the next 6 months” (5-point Likert scale, 1 = strongly disagree and 5 = strongly disagree). The single item measure of job satisfaction provides another pre-validated, rigorous yet simple method of assessing this important measure of workplace satisfaction (Dolbier, Webster et al. 2005).

Predictor variables To gauge perceptions of how workload and work demands had changed over time, the survey asked respondents who had been working in

New Zealand when the 2018 Mental Health and Addiction Inquiry was released to recall their experiences of work at this time. That year was chosen since the inquiry had key recommendations focused on improving prevention and early intervention in the community, and its report stated that “we expect demand for specialist services will reduce as issues are dealt with earlier, before they escalate …” (He Ara Oranga 2018, p110). This statement was provided for participants, and respondents were asked to recall and compare current complexity, demand for specialist services, size of caseload and on-call demands with their workloads in 2018. The survey also asked participants to subjectively assess whether they feel they are currently provided with enough administrative support, whether they feel their mental health service is adequately resourced and whether they feel able to provide their preferred level of patient-centred care in the time that they have available to spend with patients. These were all assessed according to a 5-point Likert scale (strongly agree to strongly disagree). The survey finally asked respondents how often they end up covering their colleagues’ caseloads, how often they can access the recommended levels of non-clinical time for their service and how often they are able to see patients for follow-up appointments within clinically appropriate timeframes. These were also assessed according to a 5-point Likert scale (strongly agree to strongly disagree). Demographic information was sought on gender (three categories), age according to ten-year increments, place of work, sub-specialty, ethnicity, and country of primary medical qualification. For those participants who did not receive their primary medical qualification in New Zealand, a separate set of questions was asked pertaining to length of time in New Zealand, motivation for moving to New Zealand and first medical role upon arrival. Opportunities for comment were provided at the end of each block of questions and a final comments box was provided at the end of the survey for general commentary.

Analytical procedures Quantitative analysis Raw data was summarized in Excel with Likertscale responses translated into numeric form and basic descriptive statistics were produced. Further statistical analysis was undertaken by Prof. Frampton with correlations assessed using nonparametric Spearman’s rank correlation coefficients and Kruskal–Wallis tests as appropriate. The two main outcome measures were assessed against the main variables outlined above as well as the key demographic variables. HSE scores were also assessed against each other for degree of association. A two-tailed p-value <.01 was taken to indicate statistical significance.

Regression analysis The associations between outcome variables and workplace perception measures defined above were further explored using logistic regression analyses. These analyses enabled the estimation of the likelihood of experiencing burnout or stress at different levels of the workplace variables.

Qualitative analysis Qualitative data from the free text sections of the survey were read in conjunction with the associated quantitative questions and presented alongside the data. Initial analysis of these comments was based on an iterative process where the patterns arising from the quantitative data led to further examination of the qualitative data. Many of the comments alluded to more than one theme simultaneously, which allowed for an exploration of how the issues of concern were related to each other as well as the complexity and tensions of participant’s views.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 11

Results Participants Of the 526 selected ASMS members, 368 individuals responded (70% response rate) and 147 (30%) left comments for qualitative analysis. Analysis was undertaken on the highest number of complete responses for each question with n specified where relevant. The demographic profile of respondents is detailed in Table 1. Respondents were mainly international medical graduates (60%), aged in their 50s (37%), Pākehā/ NZ European (44%), and commonly working in adult mental health services (37%).

TABLE 1: SAMPLE PROFILE CHARACTERISTICS

VARIABLE

FREQUENCY PERCENTAGE

GENDER

Total n=350

Male

174

49.7%

Female

159

45.4%

Prefer not to answer

2.6%

Gender diverse

2.3%

AGE GROUP

Total n=348

30-39

9.2%

40-49

24.1%

50-59

129

37.1%

60-69

24.7%

70 and over

2.9%

2.0%

Prefer not to answer ETHNICITY Pākehā/NZ European

Total n=340 151

44.4%

European

24.4%

Indian

8.8%

South African

7.1%

Other Asian

5.3%

NZ Māori

2.6%

Chinese

2.1%

Southeast Asian

1.8%

Middle Eastern

1.2%

Tongan

0.6%

Latin American

0.6%

Samoan

0.3%

Other Pacific Island

0.3%

Other African

0.3%

Cook Islands Māori

0.3%

12 HEALTH DIALOGUE NOVEMBER 2021

VARIABLE

FREQUENCY PERCENTAGE

COUNTRY OF PRIMARY MEDICAL QUALIFICATION

Total n=341

New Zealand

136

39.9%

United Kingdom of Great Britain and Northern Ireland

18.5%

South Africa

10.0%

India

6.5%

United States of America

5.6%

Netherlands

2.6%

Germany

2.3%

Australia

2.1%

12.6%

Other AREA OF MENTAL HEALTH

Total n=347

Adult mental health

128

36.9%

Child and adolescent

14.4%

Community mental health

12.7%

Other

12.1%

Older persons/psychogeriatrics

11.8%

Forensic psychiatry

6.9%

Alcohol and drug/addiction services

5.2%

International medical graduates Analysis of the places of work for the 208 IMGs who responded to the survey shows their geographic spread. Only Canterbury, Auckland and the West Coast had less than 50% of their staffing made up of IMGs (Figure 1).

Wairarapa Lakes Hutt Waikato Taranaki Hawkes Bay Nelson/Marl Counties Whanganui MidCentral Bay of Plenty Southern Northland Waitematā Capital and Coast Tairāwhiti South Canterbury Canterbury Auckland West Coast 0%

10%

20%

30%

40%

50% IMG

60%

70%

80%

90%

100%

FIGURE 1: DISTRIBUTION OF IMGS VS NEW ZEALAND-TRAINED PSYCHIATRISTS BY DHB

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 13

Over half (52.2%) had been in New Zealand for 15 years or more and the majority worked as senior medical officers {SMO) upon their arrival (54%). The main motivating factor for moving to New Zealand was the quality of life (56%) followed by family reasons (Table 2). Many noted that they had initially intended just to stay for a year but ended up staying much longer. Other comments referenced adventure, safety, and desire to leave country of origin due to war and security concerns.

TABLE 2: CHARACTERISTICS OF IMGS WORKING IN NZ PSYCHIATRY

Less than a year

6.4%

1–2 years

4.9%

2–4 years

5.4%

4–10 years

18.7%

10–14 years

12.3%

15 years or more

52.2%

106

Resident Medical Officer (RMO)

37.3%

Senior Medical Officer (SMO)

53.8%

114

9.0%

55.7%

118

Better working conditions

6.1%

Family reasons

17%

Job opportunities

8.0%

HOW LONG HAVE YOU BEEN IN NEW ZEALAND?

WHAT WAS YOUR FIRST MEDICAL JOB IN NEW ZEALAND?

Other (please specify) WHAT WAS YOUR PRIMARY MOTIVATION FOR MOVING TO NEW ZEALAND? Quality of life

“Very distressing to see very unwell patients who are unable to be admitted due to lack of beds”

14 HEALTH DIALOGUE NOVEMBER 2021

Main outcome measures Burnout Analysis of the single item measure burnout scores found 34.6% with high levels of burnout using the cutpoint of feeling burnt out once a week or more. Fifty eight percent of the population scored as experiencing moderate to high levels of burnout defined as feeling burnt out once a month or more (Figure 2).

Never A few times a year or less Once a month or less A few times a month Once a week A few times a week Every day 0%

10%

15%

20%

25%

30%

FIGURE 2: FREQUENCY OF FEELING BURNT OUT FROM WORK

Comments left in this section of the survey included reference to factors driving burnout as well as previous experiences of burnout. Others described strategies to mitigate against burnout, including taking time off work, changing their place of work, or decreasing their hours of work (FTE) to make their work more manageable. Comments included the following: • “Very distressing to see very unwell patients who are unable to be admitted due to lack of beds”. • “I frequently work overtime at the end of the day to feel that I’m on top of my work. If urgent things happen this may not be possible, and that is when I’m likely to feel burnt out. This is on top of the stress of working long hours several days in succession”. • “I’ve good managers and they try their best, but everything is tricky - the computer notes system doesn’t work after upgrades - it’s so hard to book a car for a home visit and I need to do that every day to do my work - can’t they work with us?”

Stress at work Levels of workplace stress were similar to those of the study population experiencing burnout. The single item measure of stress related to work revealed just over half (52%) finding their work moderately stressful and over a third (35.3%) reporting work as either very stressful or extremely stressful (Figure 3). Only three individuals did not find their work stressful at all.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 15

Not at all stressful

Mildly stressful

Moderately stressful

Very stressful

Extremely stressful 0%

10%

20%

30%

40%

50%

60%

FIGURE 3: IN GENERAL, HOW DO YOU FIND YOUR JOB?

Comments left for this section referenced many common issues including the impact of after-hours and call work and the impact of high and growing caseloads and inadequate or mismatched resourcing: • “On-call shifts are busy and I struggle to work 24 hours on-call, be woken overnight and then go to work the next day. I now take annual leave to manage my on-call shifts. I don’t work before the shift or the day afterwards. In my day job we deal with a lot of risk which is stressful”. • “Increasing demands especially with paperwork/forms that serve no useful purpose, resource limitations mean [I’m] unable to provide appropriate treatment”. • “Increasing gap between demand and resource; forced into situations to make decisions not at best interest of patient e.g., risk; staff overburdened and stressed”. • “The patient’s troubles in many areas including with family stressors, drug use, major psychiatric condition. There are less and less resources available to serve the patient. There is a pressure to treat the ‘acute’ condition and then discharge”. Other comments referenced the stress arising from the unpredictability of the work, fluctuating demand as well as workplace changes including decreased access to administrative support as well as fluctuating bed pressure: • “The biggest stress is lack of beds to admit people when working on-call, as well as the general risks of caring for people who at times may be violent or suicidal as part of their illness”. • “Workload is increasing. Not enough staff and too little time to complete paperwork which the DHB keeps increasing”.

16 HEALTH DIALOGUE NOVEMBER 2021

“On-call shifts are busy and I struggle to work 24 hours on-call, be woken overnight and then go to work the next day. I now take annual leave to manage my on-call shifts”

• “High workload, minimal assistance from registrars as their scope of practice is limited especially regarding what they can do in terms of the Mental Health Act”. • “We are constantly expected to manage complex clients without the support of psychological therapies, due to a significant shortage of psychologists and an inability/reluctance to recruit more. We are also expected to manage client’s poor mental states that are unrelated to mental illness, but rather significant social stressors such as homelessness and poverty. I sometimes feel like I’m a social worker rather than a psychiatrist”.

Levels of workplace demands and peer and managerial support Table 3 outlines the mean scores for the three different variables used from the HSE management standards tool. An interpretation and comparison of the mean scores is presented later in Figure 6. In this presentation, low scores are negative for all three domains. The scores were highly correlated with each other, as outlined in Table 4, suggesting that high demands are associated with work environments with low peer and managerial support and vice versa.

TABLE 3: MEAN HSE QUESTIONNAIRE EXECUTIVE SCORES

VARIABLE

Mean

Perceived managerial support

365

13.57

4.31

Perceived peer support

366

14.05

2.87

Perceived workplace demands*

361

20.75

5.25

*The questions for this factor are negatively phrased but to aid comparison with other factors the scores have been reversed so that a higher figure represents lower demands at work.

TABLE 4: CORRELATIONS BETWEEN HSE QUESTIONNAIRE SCORES

WORKPLACE DEMANDS

LEVEL OF PEER SUPPORT

Level of peer support

0.257**

Level of managerial support

0.224**

0.560**

**p<0.001

“Workload is increasing. Not enough staff and too little time to complete paperwork which the DHB keeps increasing”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 17

Detailed trichotomized scores for individual questions comprising the three variables are presented in Figure 4 and Figure 5.

I have to work very intensively Different groups at work demand things from me that are hard to combine I have to work very fast I have to neglect some tasks because I have too much to do I am unable to take sufficient breaks I have unrealistic time pressures I have unachievable deadlines I am pressured to work long hours 0%

10%

20%

30% 40% Never/seldom

50% 60% 70% 80% Sometimes Often/always

90%

100%

FIGURE 4: RESULTS FROM THE HSE QUESTIONNAIRE ON WORKPLACE DEMANDS (Note bars coloured red represent a negative outcome)

The findings suggest that the main demand factor is the need to work very intensively. Over half of respondents found that combining demands from different groups and having to work very fast were significant pressure factors. In terms of peer and managerial support, the negative factors most experienced were not receiving enough supportive feedback on work and not feeling supported through emotionally demanding work. Fewer than half responded often/always to the other measures except for receiving the respect at work from colleagues.

“If I could I’d work in mental health but not in a clinical role because of work pressure and stress”

18 HEALTH DIALOGUE NOVEMBER 2021

My colleagues are willing to listen to my work-related problems I receive the respect at work I deserve from colleagues I get the help and support I need from colleagues If work gets difficult my colleagues will help me I can talk to my manager about something that has upset or annoyed me at work My manager encourages me at work I can rely on my manager to help me out with a work problem I am supported through emotionally demanding work I am given supportive feedback on the work I do 0%

10%

20%

30% 40% 50% Sometimes Never/seldom

60% 70% Often/always

80%

90% 100%

FIGURE 5: RESULTS FROM THE HSE QUESTIONNAIRE ON PERCEIVED LEVELS OF PEER AND MANAGERIAL SUPPORT

Comparisons between other studies using HSE data and the current HSE scores are outlined in Figure 6. The lower scores for 2021 show a decline in perceived working conditions as represented by increased demands and low peer and managerial support. Note that the demands score is reversed so that a lower score in the graph represents an increase in demands.

24.40

25.51 20.75

Mean HSE Score

20 18 16 14

17.35

15.55 13.57

15.20

14.90

14.05

12 10 8 6 4 2 0

Level of managerial support UK Study

Level of peer support Level of workplace demands* ASMS psychiatrists 2018 (n=177) ASMS psychiatrists 2021 (n=365)

FIGURE 6: COMPARISON 2021 HSE SCORES QUESTIONNAIRE WITH OTHER RELEVANT STUDIES *The questions in the questionnaire for this factor are negatively phrased, but to help comparison across the other factors in this figure the scores have been reversed so that a higher value in the figure indicates less demands at work, as with the support factors.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 19

Intentions to leave Questions pertaining to intentions to leave revealed only 37% of respondents disagreed or strongly disagreed with a desire to leave their job if they were able. Almost half (n=161) reported that they would leave their job if they were able. Conversely, however, for those who had a plan to leave their job within the next 6–12 months, only 23% (n=77) agreed or strongly agreed that they would leave. Over half disagreed or strongly disagreed that they had firm plans to leave their job (Figure 7). A supplementary question asked those who were planning to leave their job within the next 6–12 months what their intention was; there were no clear trends, but many signalled a desire to start or increase locum work (n=30) and/or start or increase private practice (n=27). Qualitative comments in this section revealed a good deal of indecision; some noted their desire to leave the more gruelling aspects of their current work, reduce hours, or find jobs with DHBs that have better staffing ratios: • “I am weighing up options from complete retirement to working overseas e.g., Australia • Continue mixed psychiatry and pain. I would not return to Community or Inpatient psychiatry”. • “If I could I’d work in mental health but not in a clinical role because of work pressure and stress”. • “Do the same job with a DHB that has better staffing e.g., greater number of SMO FTE per population”.

I plan to leave my job within the next 6-12 months

57%

If I was able, I would leave my current job

20%

37%

10%

20%

23%

18%

30%

Strongly disagree/disagree

40%

45%

50%

60%

Neither agree nor disagree

70%

80%

90%

100%

Strongly agree/agree

FIGURE 7: INTENTIONS TO LEAVE WORK

Comments in this section noted that there were limited opportunities for other work, and many noted that “their value system places [them] in the public sector”. Others noted the impact of the recent public sector remuneration freeze, noting that for some the pay freeze was the final straw: • “Hearing about the pay freeze and being expected to manage increasing workload and risk without adequate compensation is driving this decision”. • “Really like the work, but there are also lots of difficult things about the work which I don’t like at all – mostly poor systems, staff shortages, poor models of care, lack of time and resources”. • “I think the mental health system in New Zealand is so broken that no matter where you work, you experience the same levels of disillusion”.

20 HEALTH DIALOGUE NOVEMBER 2021

Job satisfaction Nearly half of all respondents reported being either satisfied or very satisfied with their job. Only 29% were either dissatisfied or very dissatisfied (Figure 8).

3.5% 4.6%

25.3%

40.6%

Very satisfied Satisfied Neither satisfied nor dissatisfied Dissatisfied Very dissatisfied

25.9%

FIGURE 8: DEGREE OF SATISFACTION WITH JOB

Comments in this section noted that many doctors’ feelings of satisfaction came from their passion for their work, but noted that increasingly they were feeling overwhelmed by the demands on their services and the time and dedication it required: • “Escalation of referrals over the past year with no resources to see them and unable to follow existing treatment pathways due to lack of qualified therapists”. • “I like seeing people and helping them. I do not like feeling that I could do better with both my patients and the service as a whole”. • “However, I am concerned about the chronic long-term under-resourcing of the sector, and the lack of any holistic, integrated approach to early intervention for better health outcomes”. • “Really enjoy the work, seeing patients and whānau. But sometimes the demands and time required for paperwork, completing forms, MHA requirements, reports, etc. reduces the satisfaction”.

“I think the mental health system in New Zealand is so broken that no matter where you work, you experience the same levels of disillusion”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 21

Time, resourcing, and workloads As detailed in Figure 9, while a third either agreed or strongly agreed that they had sufficient time to spend with patients to deliver patient-centred care, respondents overwhelmingly disagreed that they are provided with sufficient administrative support (62% disagree/strongly disagree) and resourcing (87% disagree/ strongly disagree).

In your view, do feel you have sufficient time with patients and, (where appropriate) their families, to provide your preferred level of patient-centred care?

29%

Do you feel you are provided with sufficient administrative support in your current role?

Do you feel you are working in a well-resourced mental health service?

23%

48%

15%

62%

10%

87%

20%

30%

Strongly agree/agree

40%

50%

60%

Neither agree nor disagree

70%

80%

90%

100%

Strongly disagree/disagree

FIGURE 9: VIEWS CONCERNING TIME WITH PATIENTS, ADMINISTRATIVE SUPPORT, AND RESOURCING IN MENTAL HEALTH

Comments in this section noted a raft of issues including poor or unreliable IT services, poor administrative support and general frustration with the growing demands and poor resourcing. Respondents emphasised their frustration with the reduction of admin time, and statements like the following were common: • “the team is down in capacity by 20% and about to lose the clinical co-ordinator but referrals and demands on service have increased. The team administrator is not able to accommodate changes … I have to use personal study time to finish admin/place clinics during peer reviews and shorten appointment times to fit more people in a week”. • “[there is] no admin support. As a registrar, I used to have admin staff who would type letters I dictated. Now, all this is done in one go: seeing patients and whānau, organising follow up and scripts and typing letters and sending out reports and in the same time frame. Luckily there is still the joy of working with the patients.” “I like seeing people • “We have one administrative person (reception, typing, phone calls ...) and no cover for her leave.” • “Stuff doesn’t work. Today there is one teaspoon on our floor for hot drinks, for 40 people, someone keeps removing them? My chair is broken, the light tube is rattling again.”

22 HEALTH DIALOGUE NOVEMBER 2021

and helping them. I do not like feeling that I could do better with both my patients and the service as a whole”

• “Admin capability has decreased. Typing/spelling is atrocious and makes for tedious editing. Management of the patient flow from waiting area to office is impersonal and indifferent. We still do not have adequate hardware and technology to be able to handle AVL [audio visual link] clinical work should we experience another lockdown.” • “Growing caseloads and resourcing which doesn’t match the demand.” • “My caseload is 150+ patients, well above average for other parts of the service. This is reflective of significant population growth. I highlight these concerns regularly, but additional resource is not forthcoming. I submitted data for service sizing over two years ago, without a response yet from management.” • “My service has had no FTE increase for over 13 years, but the rest of the hospital staffing has increased by 40%.”

As detailed in Figure 10, very few respondents felt able to access the recommended amount of non-clinical time (NCT) for their service. The majority agreed to some extent that they were able to see patients for their follow-up appointments within clinically appropriate time frames, while 41% of respondents felt that they always or usually covered their colleagues’ caseloads (Figure 10A).

How often are you able to see patients for follow-up appointments within clinically appropriate time-frames?

54%

How often are you able to access the recommended non-clinical time for your service?

16%

10%

34%

20%

30%

64%

40%

Always/Usually

How often do you end up covering other colleague’s case loads?

50% Sometimes

60%

10%

20%

70%

80%

48%

30%

40%

Always/Usually

50% Sometimes

90%

100%

Rarely/Never

41%

12%

60%

11%

70%

80%

90%

100%

Rarely/Never

FIGURE 10 AND 10A: VIEWS CONCERNING TIMELINESS OF FOLLOW-UP APPOINTMENTS, ACCESS TO CME AND COVER FOR COLLEAGUES’ CASELOADS. Note the reverse scoring for the second graph.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 23

Perceptions of workload change over time Respondents were also asked to consider how their current workload demands, size and complexity had changed with recall to their conditions of work in 2018. Over three quarters of respondents reported an increase or significant increase to their workloads; none reported that the complexity of their caseload had decreased during this time.

How have the demands of your on-call workload changed?

77%

How has the size of your caseload changed?

76%

22% 1% 22% 2%

How has the complexity of your caseload changed?

86%

How has the demand for specialist mental health services changed?

14%

95%

4% 1%

10%

20% 30% 40% Significantly increased/increased

50% No change

60% 70% 80% 90% Significantly decreased/decreased

100%

FIGURE 11: COMPARISON OF KEY INDICATORS WITH 2018 WORKLOADS AND DEMANDS

Comments left for this section included the following: • “I have worked for this DHB for twenty years this year. The amount of people we see and complexity inexorably rises, and it gets harder and harder. I still think psychiatry is a good job, but most of my senior registrars arrive at our team saying they don’t want to be psychiatrists as all the psychiatrist jobs look awful. It is very sad. When I look back on patient files, I am reminded how much care we could provide 5, 10 and 15 years ago to specific patients compared to now.” • “On-call shifts have become almost universally busy, with increased referrals, acuity, and complexity. In addition, inadequate resources can mean a disproportionate amount of time is spent in resource discussions, rather than clinical care.” • “Explosion of methamphetamine and drug related mental health problems, family violence and child protection problems.”

24 HEALTH DIALOGUE NOVEMBER 2021

“Stuff doesn’t work. Today there is one teaspoon on our floor for hot drinks, for 40 people, someone keeps removing them? My chair is broken, the light tube is rattling again”

• “We now see lots of patients who come straight from prison, as forensic services are overwhelmed and they have usually been significantly traumatised in prison also. I have discharged numerous patients as soon as they achieve some tenuous stability, but new ones keep pouring in. Most of my patients are poor also.” • “I am referred more complex patients, but with less resources and too little time.” • “Overall caseload of my community sector has increased by over 25% in just [the] last one year. While SMO resource has had no increase in years the workload has increased over 25% in just [the] last one year.” • “The demands have increased with the effect that patients are discharged faster, seen less frequently and there is more demand on GPs to manage patients.” While the survey did not attend to the issue of recruitment and interest in psychiatry as a medical specialty, several comments noted issues with consultants working without junior staff. For example, one respondent noted that: • “The acute services are entirely overrun with some consultants working without junior staff which would be expected in their area. This leaves consultants doing menial data-entry and phone calls, working well above 40 hours per week, patient care meetings encroaching on “non-clinical time” and units running above their supposed bed capacity. The needs of the service, … seem to constantly expand without any additional resources.” Related were comments that spoke to the impact of not having trainees or registered medical officers to assist in services: • “I am called upon to conduct tasks that can be done by junior doctors (registrars and house officers) in addition to what I am supposed to do as a consultant. I am also swamped with paperwork. The MHA is an extremely time-consuming process. Therefore, I am unable to use my time effectively to provide a specialist level of care for my patients and their whānau.” • “I love working with my clients/patients; however, the current system is unsustainable. We do not have enough staff or resources to retain staff, the staff around me are burnt out, unmotivated and it is painfully obvious.” The consequences of this growth in demand and concerns from clinicians for the impact on the quality of care for mental health patients were clearly articulated in the qualitative comments. Some described feeling they had little choice but to discharge patients before they were ready, resulting in greater workload pressures down the line: • “[I compensate] for lack of time by discharging outpatients who would benefit from remaining under MHS for longer.” • “There are less and less resources available to serve the patient. There is a pressure to treat the ‘acute’ condition and then discharge.” • “We often feel like patients are being discharged to the community to fail. This failure takes the form of suicide, homicide, estrangement and homelessness.”

“My service has had no FTE increase for over 13 years, but the rest of the hospital staffing has increased by 40%”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 25

• “Our population is increasing, alongside case complexity and public demand for best evidence medicine. Yet, our budget was cut this year. We are often asked to work against best practice guidelines because of a lack of resourcing. In this way, I do not enjoy my work and do not think I can continue to be part of a damaged system. It has been very difficult for me and my family making this decision.” • “If you are able to meet time frames for follow up that means the care is suboptimal.” • “Due to high caseloads, patients are not seen as often as required by best practice guidelines, often slowing their recovery. Management insists on no wait list so more and more patients are added to caseloads even when there are not enough resources, slowing everyone’s recovery. More and more complex patients require review of their medical records or proper diagnosis and treatment which often are in large volumes of files, which requires substantial time before and after seeing patients.”

Correlations between variables and main outcome measures Non-parametric Spearman’s correlations found significant associations between all variables and the degree of stress experienced in the role and whether individuals scored as burnt out. The exception to this was whether people felt that the demands of their on-call workload had changed; this was not positively associated with either finding the work stressful (p=0.358) or scoring as burnt out (p=0.111). Negative correlations were found between the positively worded questions and the main outcomes; these are highlighted in red in Table 5. These negative associations suggest that having good support, ability to access NCT and having sufficient time with patients, low demands, and high peer support, are all correlated with low burnout and stress scores.

“When I look back on patient files, I am reminded how much care we could provide 5, 10 and 15 years ago to specific patients compared to now”

26 HEALTH DIALOGUE NOVEMBER 2021

TABLE 5: CORRELATIONS BETWEEN ORGANISATIONAL VARIABLES AND THE TWO MAIN OUTCOME MEASURES

DEGREE OF STRESS ASSOCIATED WITH JOB

SCORING AS BURNT OUT FROM WORK

If I was able, I would leave my current job

0.484**

0.553**

I plan to leave my job within the next 6–12 months

0.267**

0.294**

Considering everything, how satisfied are you with your job?

0.533**

0.609**

Do you feel you are provided with sufficient administrative support

−0.223**

−0.208**

Do you feel you are working in a wellresourced mental health service?

−0.283**

−0.251**

In your view, do feel you have sufficient time with patients to provide your preferred level of patient−centred care?

−0.294**

−0.318**

How often do you end up covering other colleague's caseloads?

0.236**

0.318**

How often are you able to access the recommended non-clinical time for your service?

−0.294**

−0.224**

How often are you able to see patients for follow-up appointments within clinically appropriate timeframes?

−0.275**

−0.257**

How has the current demand for specialist mental health services changed?

0.176**

0.154**

How has the complexity of your caseload changed?

0.297**

0.232**

How has the size of your caseload changed?

0.205**

0.186**

How have the demands of your on-call workload changed?

0.060

0.103

Perceived level of workplace demands

−0.560**

−0.511**

Perceived level of peer support

−0.312**

−0.348**

Perceived level of managerial support

−0.251**

−0.314**

**p<0.001

“While SMO resource has had no increase in years the workload has increased over 25% in just [the] last one year”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 27

Demographic variables Analysis found no significant correlations with demographic variables and any of the two main outcome measures, with the exception that New Zealand trained graduates perceived their workplace demands to be higher than those who were IMGs. The Wilcoxon summed rank test did not reveal significant differences in either burnout or degree of stress related to the job by place of work. Nevertheless, as shown in Figure 12 and 13 the proportion of respondents with burnout and degree of stress varied from 8.3% to over 60%. Any places of work that had fewer than five respondents were excluded from this graph due to the risk of identification of individual respondents.

Survey average

35.3%

Waikato DHB

60.9%

Hawkes Bay DHB

58.3%

Nelson-Marlborough DHB

50%

Southern DHB

48.1%

Auckland DHB

43.8%

Hutt Valley DHB

40%

Waitematā DHB

32.2% 30%

Counties Manukau DHB

28.9%

Capital & Coast DHB 22.2%

Canterbury DHB 17.6%

Bay of Plenty DHB

16.7%

Taranaki DHB 0%

10%

20%

30% 40% 50% Proportion of psychiatrists with burnout

60%

70%

FIGURE 12: PROPORTION OF RESPONDENTS WITH BURNOUT BY DHB BY PLACE OF WORK

Waikato DHB Southern DHB Auckland DHB Counties Manukau DHB Taranaki DHB Hawkes Bay DHB Waitemata DHB Capital & Coast DHB Nelson-Marlborough DHB Bay of Plenty DHB Northland DHB Canterbury DHB Hutt Valley DHB 0%

10%

20%

30% 40% 50% 60% 70% 80% Proportion of psychiatrists by degree of stress related to work

Not at all stressful and mildly stressful

Moderately stressful

90%

100%

Very stressful and extremely stressful

FIGURE 13: PROPORTION OF RESPONDENTS BY DEGREE OF STRESS RELATED TO WORK BY PLACE OF WORK

28 HEALTH DIALOGUE NOVEMBER 2021

Regression analysis Analysis was undertaken to test the strength of association between the variables in the study and the two primary outcomes: experience of burnout and stress at work. The results found that all factors were significantly associated; Table 6 presents the variables and the odds ratio scores for the two outcomes. The results suggest that positive experiences with the workplace, for example, working in a well-resourced mental health service, having good levels of administrative support, and having regular access to recommended amounts of non-clinical time, are all likely to reduce the likelihood of experiencing burnout and finding work stressful. Having to cover caseloads of colleagues was found to increase likelihood of stress and burnout, as was having higher levels of dissatisfaction in the workplace (note this variable negatively worded so higher dissatisfaction coding linked to higher stress and burnout). Having time to provide desired levels of patient-centred care and being able to see patients within the clinically recommended time frame for follow-up appointments had the biggest impact on the likelihood of reducing burnout; being able to accomplish these two factors reduced the likelihood of experiencing burnout by nearly 46% and 48% respectively.

TABLE 6: ODDS RATIOS BY VARIABLES AND MAIN OUTCOMES

DEGREE OF STRESS ASSOCIATED WITH JOB p-VALUE

ODDS RATIO*

LOWER 95%

Desire to leave job

0.000

2.274

1.820

2.840

Plan to leave job

0.000

1.526

1.274

Work in wellresourced MHS

0.001

0.585

Degree of job satisfaction (Negatively worded variable)

0.000

Has admin support

FREQUENCY OF FEELING BURNT OUT

UPPER p-VALUE 95%

ODDS RATIO*

LOWER 95%

UPPER 95%

0.000

3.213

2.454

4.205

1.829

0.000

1.709

1.417

2.061

0.429

0.798

0.000

0.489

0.350

0.682

3.402

2.564

4.514

0.000

5.029

3.612

7.000

0.001

0.719

0.590

0.878

0.000

0.635

0.516

0.782

Has sufficient time with patients

0.000

0.652

0.525

0.810

0.000

0.543

0.432

0.682

Covering colleagues' caseloads

0.001

1.602

1.224

2.097

0.000

2.114

1.587

2.815

Ability to access NCT

0.000

0.626

0.495

0.790

0.000

0.637

0.505

0.803

See patients for follow-up within clinically appropriate timeframes

0.000

0.545

0.403

0.736

0.000

0.517

0.382

0.702

* This represents the relative risk of the outcome if the risk factor increases by one category on the Likert scale. So, for example, for the variable ‘desire to leave job’ an endorsement of strongly agree is associated with 2.27 times greater risk of stress compared to agree.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 29

Discussion This Health Dialogue represents an in-depth account of the psychosocial wellbeing of psychiatrists working at the frontline of New Zealand’s mental health service. It extends previous research into the quantitative indicators of mental health demand and the mismatch between demand and current resourcing and explores the impact of this demand for the wellbeing of mental health providers. It contributes to existing literature on the psychosocial wellbeing of this crucial sector of New Zealand’s mental health workforce, and the high 70% response rate ensures good representativeness and reliability of the findings. The application of existing, pre-validated measures such as the single item measure of burnout, degree of job stress and the use of the HSE indicators of workload, peer and managerial support ensures comparability of these results with other peerreviewed literature and an objective assessment of the pressures this psychiatry workforce is under. The inclusion of questions pertaining to intentions to leave, ability to access admin support and nonclinical time, as well as subjective perceptions of changes to workload over time add depth to the likely drivers of the psychosocial wellbeing of this workforce. Application of the odds ratio analysis enables insight as to what factors may play the greatest role in terms of instituting wider change. The combination of qualitative and quantitative data with analysis of the former, describing in real terms the personal and professional impacts of the mental health crisis, further adds to the strengths of this study and provides insight into the consequences of a stretched and strained system for clinicians and patients alike.

Indicators of wellbeing The primary objective of this research was to investigate the psychosocial wellbeing of psychiatrists. This was achieved by focusing on levels of burnout as represented by a single item measure and the overall measure of work-

30 HEALTH DIALOGUE NOVEMBER 2021

related stress. Secondary outcome measures were represented by degree of job satisfaction, and intentions to leave work as well as dimensions of the HSE management standards tool. The results from this study find over a third of respondents reporting high levels of burnout and over half reporting high levels of work-related stress. The burnout scores are slightly higher than those in other research on psychiatrists; a similar study into rates of burnout in Irish psychiatrists found 33% suffering from burnout (McLoughlin, Casey et al. 2021). Kumar’s study into New Zealand psychiatrists (Kumar, Fischer et al. 2007) also reported 33% as suffering from high emotional exhaustion and found 35% suffering similarly in their follow-up study published in 2012 (Kumar, Sinha et al. 2012). The single item measure of burnout used in this survey is not as fulsome a measure as the full MBI or Copenhagen Burnout Inventory, but alongside the degree of stress experienced at work it provides a good overall snapshot of psychiatrist wellbeing. This study adds to the literature demonstrating the pervasiveness of burnout as an entrenched feature of the senior medical workforce and a pressing issue of concern for those working in psychiatry. The consequences of burnout in terms of risks of medical error and possible consequences for the quality of patient care has been well substantiated in other studies (Shanafelt, Balch et al. 2010, Montgomery, van der Doef et al. 2020). In psychiatry specifically, the consequences of an exhausted, stressed workforce for patient care are not to be taken lightly; for example, research into New Zealand SMOs has postulated a negative relationship between degree of burnout and degree of patient-related empathy (Reynolds, McCombie et al. 2021). As noted in the introduction to this Health Dialogue, being able to hold empathy for patients in mental health has high therapeutic salience (McNicholas, Sharma et al. 2020). This issue was directly described by one

respondent: “...being over worked and stressed in our job as an SMO reduces our empathy towards our patients and their families. This is not helpful.” While the cross-sectional nature of this research means a causal relationship cannot be inferred, the possibility that burnout and stress may have an impact on the degree of empathy towards patients is concerning and must be taken seriously by those involved with staffing and resourcing of services.

Risk factors The second objective of this research was to identify the main risk factors for work-related stress. This was achieved by correlation analysis of the two primary outcome measures with demographic and non-demographic variables as well as the odds ratio analysis. These are discussed in turn. Correlation analysis suggests that experiences of burnout and work-related stress were widespread; there were no significant patterns of association in terms of demographic variables. This suggests that psychiatrists across the board are equally at risk of stress and burnout. Nevertheless, it is worth noting the variance in burnout scores; some DHBs such as Waikato and Hawke’s Bay had over half of their psychiatrists responding to the survey scoring with high rates of burnout (61% and 58% respectively) whereas Bay of Plenty, Taranaki and Northland all had fewer than 20% scoring with burnout. Similarly, no respondents at either Hutt or Taranaki reported that they found their work not at all stressful or mildly stressful. This suggests that there may be issues at certain DHBs that are more pronounced than others and are worthy of further investigation. Correlation analysis with non-demographic variables found that experiencing burnout and work-related stress was significantly correlated with all predictor variables except for having experienced a change to the demands of the on-call workload (p=0.358 and 0.111 for stress and burnout respectively). While the crosssectional nature of this research means that causality cannot be inferred, it does suggest that clinician’s wellbeing is significantly associated

with perceptions of demand, how well resourced their mental health service is felt to be, and whether they feel they are provided with support. All these factors are recognised to constitute significant sources of stress and are understood as contributing factors in burnout research more widely (Walsh, Hayes et al. 2019, Zhang, Mu et al. 2020, Chênevert, Kilroy et al. 2021). The findings of this research are consistent with those in other studies such as that conducted by McNicholas, Sharma et al. (2020) and an earlier study by Walsh, Hayes et al. (2019), which emphasised the contributory factors to burnout of staff shortages, inadequate resourcing and a disjunction between the views and priorities of managers. The odds ratio analysis enables a finer grained exploration as to key risk factors as well as potentially protective factors associated with workrelated stress and burnout. Feeling that clinicians had adequate time with patients and being able to see patients for follow up appointments within clinically recommended guidelines appear protective against stress and burnout. These factors tally with the qualitative comments left which emphasise the desire for clinicians to provide the best quality care for their patients while simultaneously finding this challenging to do because of heavy workloads and varying constraints upon their time. These connections between wellbeing and conditions of work are clearly substantiated in the wider literature; as defined by Maslach and Leitner (2016), risk factors include workload, degree of control, recognition or reward for work effort, and having a sense of community and support in the workplace. The strong associations between indicators of resourcing and demand and ability to provide

“Non-clinical time is a joke, it’s when I catch up with admin”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 31

quality patient care emphasises the significance of well-resourced mental health services for the wellbeing of those tasked with providing care. Other factors noteworthy in this research include the positive impacts on wellbeing of enabling clinicians to access their non-clinical time. At present, the ASMS MECA recommends 30% allocation of NCT, but many respondents suggested that access to NCT was extremely rare and is generally used to catch up with clinical administration tasks: “Non-clinical time is a joke, it’s when I catch up with admin”. Other studies have suggested that regular access to as little as 2 hours a week of NCT is likely to reduce the likelihood of doctors experiencing burnout (Stevens, Davey et al. 2020). Others noted little support for NCT: “Management look down on SMOs who even talk about NCT. It is considered taboo as if [the] MECA has not been read by the DHB management.” Ensuring access to genuine NCT is a relatively costneutral change that could be readily implemented if resourcing and staffing were better matched to demand for mental health services. Another aspect that the odds-ratio analysis highlighted was the importance of having adequate administrative support. Those respondents who reported good levels of administrative support were 28% less likely to find their job stressful and 37% less likely to experience burnout. This emphasis on administration is consistent with other studies in the New Zealand context such as that by Fischer and Kumar who note that lack of administrative support can also include the notion of “an aggressive administrative environment”, in which there is a feeling of vulnerability to persecution via complaints and inquiries” (Fischer, Kumar et al. 2007 p420). This was noted in some comments which referenced the burdensome nature of the mental health act but also the feeling that clinicians were ill supported if something was to go wrong:

“Administrative support is very limited. Admin staff are unhappy and often unhelpful. Time with patients and their families is sacrificed due to DHB demands of filling in useless

32 HEALTH DIALOGUE NOVEMBER 2021

forms. These forms are proxy measures of our performance and safe practice. We are thrown under the bus if there is an HDC [Health and Disability Commission] complaint, management tends to point out forms that were not filled in.” Again, while causality cannot be inferred, these findings suggest that readily achievable and relatively cost-neutral changes such as ensuring access to NCT and provision of more comprehensive administrative support would likely pay dividends in terms of improving wellbeing for this critical workforce. As one respondent summed up “If you could improve admin … ease all the bureaucracy of electronic notes etc ... if we could be listened to and valued ... before the system completely collapses …” The inclusion of the HSE management standards tool in this study provides further insight as to the connections between wellbeing and conditions of work. The correlation analysis suggests that emotional exhaustion and job stress are strongly associated with working situations where work demands are high, workloads are growing, and support levels are low. Comparison of mean HSE scores with a UK study and a previous ASMS survey finds that levels of workplace demands are higher in this current research while levels of support, both peer and managerial, are lower. The results further suggest that working intensively is the most significant aspect of work demands, while colleagues provide the main source of respect and willingness to listen to work-related problems. There was no significant variation in any of the three HSE scales by demographic variables suggesting again that the pressures are widespread

“Unrealistic expectations are placed on skilled/ competent clinicians who are working grossly in excess of their job size”

and not particular to any place of work or other demographic factor. The low scores for receiving supportive feedback and being supported through emotionally demanding work provide indicators as to what factors can be improved in workplaces. In line with other research, working in environments where clinicians feel supported by managers and have strong sense of collegiality are protective factors for wellbeing and have the capacity to ameliorate feelings of stress (Chambers, Frampton et al. 2018, Ike, Durand-Hill et al. 2021). In contrast, many respondents in this study expressed frustration with perceived disconnect with non-clinical managers regarding the pressures and stresses faced by clinicians at the coal face: • “There is a low level of trust with management... unrealistic expectations are placed on skilled/competent clinicians who are working grossly in excess of their job size.” • “Not feeling valued by management is my strongest reason for wanting to get out.” • “Mental health services are managed by nonpsychiatrists with poor knowledge of quality of care”. • “Very poor managerial support; practically none and their only concern is how to save money denying payment for additional duties and taking longer time to pay the money when claims are put forth”. As noted in previous research into psychiatrist burnout, efforts from non-clinical managers to build good quality relationships with their clinical staff can have a significant positive impact on morale (Kumar, Hatcher et al. 2010). Given the salience of morale for job satisfaction and in turn, likelihood of staff retention, management would do well to consider how to strengthen connections with clinical staff to improve perceptions of disconnect and lack of respect. The importance of this was also referenced in a comment from one participant who stated that “All I look for is treatment with respect and dignity from the management.” Implicit to much of the commentary was a latent sense of anxiety regarding the

burdensome workloads they were managing. The following section examines how respondents perceived their current workloads in comparison to 2018 when the mental health review was released.

Changes since 2018 The third objective of this research was to gauge perceptions of how workload (quantity and quality) had changed over time. This aspect of the research required those who had been working in the system at the time of the 2018 mental health inquiry to recall and compare their current work demands and case complexity with their workloads three years prior. Only those who had been working at the time of the 2018 inquiry were asked to complete this question. The long recall period and the subjective nature of this question is a limitation of this study. Nevertheless, the results corroborate the trends evidenced in the companion study released earlier by ASMS (2021), which suggests funding for mental health services has not kept pace with demand. The perception of the psychiatrists in this study was overwhelmingly negative, with nearly all reporting increases to the demands of their on-call workloads, caseloads, and the complexity of patients seen; 95% of respondents reported that demand for their services had increased or significantly increased. Many responded to the question with comments that expressed incredulity at the predictions in the 2018 mental health inquiry, which stated demand was expected to decrease because of promised funding increases. For example, one psychiatrist noted “I despair when I read that statement in the report - bears no resemblance AT ALL to reality” (emphasis in original). Other subjective indicators of demand included in this research included perceptions of adequacy of time with patients, whether participants felt their service was well resourced, and how frequently they covered the caseloads of their colleagues. Further longitudinal research would be helpful to track these indicators of demand and how they may change over time. The results from this study were closely matched by unpublished data from

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 33

a survey by the Royal Australasian New Zealand College of Psychiatrists (RANZCP). In their survey report, they noted 90% of college members felt that patient demand in their after-hours work had increased or increased significantly over the previous two years. The potential consequences of this increase in demand were foreshadowed by the high proportion of respondents who were indicating a desire to leave the profession. This issue and views regarding job satisfaction are discussed in the following section.

Job satisfaction and intentions to leave The final aim of this study was to assess the degree of job satisfaction and the possible consequences of all the factors assessed in this study for possible intentions to leave. Despite the moderate levels of burnout and work-related stress reported in this study, over 40% of participants stated they were satisfied with their job, with 4.6% very satisfied with their job. This finding was consistent with that reported by Kumar et al. in their 2012 survey, where job satisfaction appeared to persist at a rate higher than levels of burnout. Job satisfaction is known to be a protective factor against burnout, and the greater the levels of job satisfaction, the lower the levels of burnout and work-related stress. Nevertheless, as outlined in the logistic regression results, a decrease in job satisfaction was associated with a 3.4 times greater likelihood of finding the job stressful and 5.0 times greater likelihood of experiencing burnout. The combination of the quantitative and qualitative results suggests most psychiatrists surveyed gain pleasure and enjoyment from their interactions with patients. For example, one respondent emphasised the importance of their interactions with patients for their overall sense of satisfaction: • “My satisfaction is driven by my selfmotivation and interest in the work – my interactions with patients and unit staff keeps me focused and enjoying my work”.

34 HEALTH DIALOGUE NOVEMBER 2021

Despite this positivity, however, many spoke of the impact on their job satisfaction from what they described as a system set up to fail patients: • “Job satisfaction can be poor due to a lack of community rehabilitation and community placements, meaning we often feel like patients are being discharged to the community to fail. This failure takes the form of suicide, homicide, estrangement and homelessness.” • “Lack of housing, both state and supported and full tertiary service and under-resourced MHS, which have not kept up with increase in demand – are all contributing to a blocked inpatient unit, which in turn have further knock-on effects throughout the MHS – leading to increased risk of SMO burnout and job dissatisfaction.” One of the potential consequences of low job satisfaction was manifest in the respondents’ views concerning desire and plans to leave their work. There is an acknowledged relationship between low job satisfaction and intentions to leave medicine (Hann, Reeves et al. 2011, Nguyen and Tran 2021), as well as between intentions to leave medicine and rates of stress and burnout in doctors (Hämmig 2018, Chênevert, Kilroy et al. 2021). While understanding the relationship between burnout and work stress and intentions to leave psychiatry was not a specific focus of the research, the association found between low job satisfaction and thoughts of and intentions to leave work is noteworthy. This association is consistent with established literature and reinforces what is potentially at stake for a mental health workforce already experiencing staffing shortages and difficulties with growing demand for specialist interventions.

“Every week our small service hosts yet another farewell for a departing staff member. I have never experienced such disillusionment in a workplace”

The relatively high proportion of respondents indicating that they would leave their jobs if they were able should also be of concern. Up to 63% of the current specialist psychiatrist workforce represented in this study have some desire to leave their current job (combined strongly agree and agree and neither agree nor disagree). This amounts to 161 individuals who would like to leave and 66 who are on the fence. The comments indicated that for many changing jobs or leaving their work was not possible, despite what they might like to do, because of personal circumstances and financial pressures. While only 23% have firm plans to leave their job within the next 6–12 months, this percentage equates to 77 psychiatrists, a number that New Zealand can ill afford to lose. Comments in this section of the survey emphasised the impact of stress levels on decisions to leave work, as well as the disillusionment many participants felt: • “I am considering changing my work – e.g., going into private practice part time. If I could move entirely to another job that was equally well paid, and not as stressful then I would strongly consider it.” • “Every week our small service hosts yet another farewell for a departing staff member. I have never experienced such disillusionment in a workplace, across the board of colleagues in other disciplines. I enjoy the patient-contact work itself, but the lack of collegiality is a new and profoundly disturbing element in a mental health service.”

as the acknowledged issues with attracting trainees into psychiatry (Greig 2021). Research by Snyder and Kumar (2004) has already raised concerns with the maldistribution of psychiatrists in New Zealand, particularly with relation to the concentration of IMGs outside main centres. In their view, there needs to be more attention paid to factors that increase satisfaction with work and help to reduce stress and burnout so that potential churn and attrition is prevented. This possibility of losing doctors from an already stretched and strained service needs to be taken seriously by those with responsibility for planning and managing mental health services in New Zealand.

• “I think my health will deteriorate if I stay in my current job.” As noted in the introduction to this Health Dialogue, New Zealand’s health system has a very high reliance on international medical graduates, particularly in psychiatry which tops the tables for proportion of IMGs. This diversity and wealth of different cultural perspectives for our medical workforce is to be celebrated but it does mean New Zealand is vulnerable to international trends regarding shortages of medical specialists, as well

“I think my health will deteriorate if I stay in my current job”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 35

Conclusion This study provides new information on the relationship between key indicators of wellbeing, levels of job demand, perceptions of workload over time and job satisfaction in the cornerstone of our mental health workforce. This study is of key relevance for those tasked with workforce planning for psychiatry and emphasises potential consequences of the high demand for mental health services for the retention and recruitment of doctors in psychiatry. The high response rate to the survey and the use of pre-validated instruments allows for comparison with other studies as well as ensuring rigour and confidence in these findings. Nevertheless, the cross-sectional design of the survey and inclusion of subjective self-report measures with a relatively long recall period are limitations of this research. This also means that the associations noted between core variables cannot be causally determined. Since the completion of this study, COVID-19 continues to have a significant impact on the wellbeing of the health workforce in New Zealand. It is likely that demand for mental health services will continue to escalate as well as the pressures on health care workers to continue to provide care in these uncertain and stressful times. This means that some of the pressures outlined in this research are likely to worsen in the short term. Moreover, work to amend the Mental Health Act, may also institute changes to working requirements and the demands on psychiatrists. It will be important to monitor the impact of both these issues closely. The findings of this research – that over a third of psychiatrists in this study are experiencing high levels of burnout and work-related stress, and many are struggling with high work demands and moderate levels of peer and managerial support – are concerning. All of these factors are known to be associated with increased intentions to leave work and for those suffering from burnout, recovery is known to be a lengthy process. The New Zealand mental health system can ill afford to lose any more psychiatrists and departments will

36 HEALTH DIALOGUE NOVEMBER 2021

continue to struggle to cover leave of colleagues if burnout results in the need to take significant periods of time off work. Monitoring of levels of stress and burnout is recommended to ensure that these metrics do not worsen in the short term, particularly given the escalating impact of COVID-19. The near consensus of respondents regarding the growth in workloads, on-call demands, and patient complexity, provides a worrying complement to the statistics outlined in ASMS’ earlier Research Brief focussing on mental health. The findings from this Health Dialogue emphasise the human cost of the swelling growth in demand for mental health services and must serve as a clarion call to those with responsibility for workforce planning. This research backs the recommendations from the Research Brief, emphasising the need for an urgent workforce census, and the re-allocation of resources into mental health to enable more sustainable workloads in the short term. The findings from this research - that 40% of respondents are routinely covering their colleagues’ workloads - further suggests that better modelling of full-time equivalents (FTE) is required to ensure that leave cover is properly resourced by the service rather than covered at the expense of individual workloads. Vacancies too, put huge stress on a workforce already under extreme pressure; there must be better planning and strategies to cope with these. ASMS recommends that mental health services are urgently service sized to ensure that the demand is aligned with the number of psychiatrists able to provide care. Updated service sizes would account for growth in demand and ensure that existing staffing is able to deal with vacancies and cover in a sustainable way. The finding that 64% of respondents are never or rarely able to access their non-clinical time is further evidence of a mismatch between service planning and demand. Non-clinical time should not be used for routine clinical administration duties or to fill service shortfalls. The fact that many

respondents described non-clinical time as ‘a joke’ further indicates that demands on secondary mental health services are not being accurately translated into correct resourcing. ASMS has an existing safe staffing accord that recommends the use of job sizing to determine adequate staffing levels and address workforce shortfalls. ASMS recommends that all mental health providers are urgently job-sized to ensure that psychiatry workloads are safe and sustainable. Ensuring that each psychiatrist has a clearly defined workload which accounts for the growth in demand and patient acuity, is fundamental to achieving safe work, psychiatrist wellbeing and job satisfaction. This research also lays down a challenge for the new national health employer Health NZ. It must ensure staffing rates are adequate in mental health services across the country. This includes the staffing of other critical components of the mental health system such as nurses, key workers, psychologists, and counsellors. Health NZ must also ensure that buildings and infrastructure are fit for purpose. The emphasis in the comments provided in this report regarding the impact of poor physical work environments, absence of functional IT systems and logistical challenges to complete the simplest of tasks is not going to improve doctor wellbeing or health outcomes for mental health patients. This research provides pointers to Health NZ as to what factors could be readily adjusted to improve working conditions. The provision of better administrative support, functional and connected IT systems are all highlighted in this research and would bring significant benefits to improving wellbeing. The provision of fit for purpose workspaces in the form of individual offices would also be a sensible place to start given the sensitive nature of much mental health care provision as well as giving clinicians safe work environments to accomplish their core tasks. Given the demographic profile of psychiatrists in New Zealand, this research also emphasises the importance of succession planning. A number of comments referenced the dwindling supply of medical registrars to assist with workloads,

and ASMS has already noted the importance of attending to the medical pipeline in previous publications. Each department must have plans in place to ensure that as psychiatrists age, there is work in progress to find replacements and bring younger staff up through the specialist ranks. It is a cause for concern that a number of services employ few or no resident medical officers (RMOs). The high reliance on IMGs to staff psychiatry is further indication of the need to address medical pipeline planning to encourage medical students to consider psychiatry as a sound option for their specialist training. While the majority of IMGs surveyed in this research had been in New Zealand for 15 years or more, Medical Council data shows about a quarter of IMGs leave within three years of gaining vocational registration in this country. There must be national efforts to develop strategies to improve the retention of the IMG psychiatry workforce. Efforts to ensure that workloads are sustainable, manageable, and fulfilling for clinicians will assist in making psychiatry a more attractive medical specialty proposition for trainees. The persistence of relatively high job satisfaction for psychiatrists involved in this research is heartening. Psychiatrists continue to report significant satisfaction from their ability to provide high quality care for those in need. Nevertheless, the findings of this research that just over half of those surveyed are always or usually able to see patients for follow-ups within clinically appropriate timeframes and that many feel discharges are made too soon due to sheer volume of demand is of concern. Psychiatrists do not wish to discharge their patients to fail. This research makes clear the emotional toll for clinicians as well as the serious possible consequences for patients.

“Due to high caseloads, patients are not seen as often as required by best practice guidelines, often slowing their recovery”

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 37

ASMS recommends the establishment of clinically informed benchmarks for safe provision of inpatient and community-based services so that workloads are safe and sustainable for doctors and caseload limits enable best practice guidelines to be followed.

time with patients, adequate administrative support, and genuine time for non-clinical tasks were all associated with reductions in work-related stress, and the likelihood of experiencing burnout - all of which have the capacity to improve job satisfaction and reduce attrition.

Finally, those responsible for designing mental health models of care need to ensure that recovery time is in place in each service or department to allow for recovery from fatigue following afterhours call. Clinicians should not have to use annual leave provisions or donate their non-clinical time to ensure they have adequate recovery time following onerous on-call duties. Rosters require urgent updating to ensure they provide scheduled time to recover from increasingly onerous on-call work, as described by clinicians in this research.

It is not good enough to run our forensic, inpatient, and community-based mental health services on an assumption that only 3% of our population requires this level of care. Evidence already shows that figure is closer to 5% and it is imperative that funding is increased to better account for this demand. The health of both patients and providers depends on it.

Essentially, those designing and resourcing the work need to use established methodologies around job and service sizing to ensure they provide enough staff to meet current and future demand for mental health services. Factors such as ensuring adequate

38 HEALTH DIALOGUE NOVEMBER 2021

While we will continue to draw attention to the manifold funding and planning issues that undermine doctors’ ability to provide best care for the communities they serve, our preference is that those responsible for investing in our health system look to the evidence and address this urgent need. If we don’t care for those providing the care to the most vulnerable in our society, then who will?

Recommendations in summary • Measure and report on the impact of COVID-19 and the Mental Health Act on the workloads and wellbeing of psychiatrists. • Implement formal standardised monitoring of burnout and stress levels in psychiatrists. • Complete a national workforce census, along with annual reporting on staffing levels, training, and succession-planning. • Develop and adhere to agreed minimum safe staffing levels in mental health units. • Regular job and service sizing completed and adhered to as described in our MECA. • Mental health buildings and facilities are audited regularly and are fit for purpose, ensuring safe work for staff and safe care for patients. • Appropriate and sufficient administrative support is put in place to support clinical work in mental health settings. • Psychiatrists have private office spaces to work in. • Rosters are urgently updated to ensure scheduled recovery time is provided following on-call work. • Succession planning is implemented across all services. • Training and development is implemented with Health NZ to ensure we build and support a sustainable psychiatry workforce. • Specific supports are given to our IMG workforce, as part of retention planning and implementation.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 39

References ASMS (2017). International Medical Migration: How can New Zealand Compete as Specialist Shortages Intensify? Research Brief. Wellington, New Zealand, ASMS. Issue 6. ASMS (2020). Building the workforce pipeline, stopping the drain. Wellington, New Zealand, ASMS. ASMS (2021). What Price Mental Health? The Crisis and the Cure. Research Brief. Wellington, New Zealand, ASMS. Issue 29. Broderick, N., R. Vaughan, F. McNicholas and L. Feeney (2021). “Staff stress and burnout in a community adult mental health service.” Irish Journal of Psychological Medicine: 1-9. Chambers, C. (2021). “My Employer is Exhausting” Burnout in the Senior Medical Workforce Five Years On. Health Dialogue. Wellington, New Zealand, ASMS. 17. Chambers, C. N., C. M. Frampton, M. Barclay and M. McKee (2016). “Burnout prevalence in New Zealand’s public hospital senior medical workforce: a cross-sectional mixed methods study.” BMJ Open 6(11): e013947. Chambers, C. N. L., C. M. A. Frampton, M. McKee and M. Barclay (2018). “’It feels like being trapped in an abusive relationship’: bullying prevalence and consequences in the New Zealand senior medical workforce: a cross-sectional study.” BMJ Open 8(3): e020158. Chênevert, D., S. Kilroy, K. Johnson and P.-L. Fournier (2021). “The determinants of burnout and professional turnover intentions among Canadian physicians: application of the job demands-resources model.” BMC Health Services Research 21(1): 993. Choudry, A. and S. Farooq (2017). “Systematic review into factors associated with the recruitment crisis in psychiatry in the UK: students’, trainees’ and consultants’ views.” BJPsych bulletin 41(6): 345-352. Cousins, R., C. J. MacKay, S. D. Clarke, C. Kelly, P. J. Kelly and R. H. McCaig (2004). “‘Management Standards’ work-related stress in the UK: practical development.” Work & Stress 18(2): 113-136. Dolan, E. D., D. Mohr, M. Lempa, S. Joos, S. D. Fihn, K. M. Nelson and C. D. Helfrich (2015). “Using a single item to measure burnout in primary care staff: a psychometric evaluation.” J Gen Intern Med 30(5): 582-587. Dolbier, C. L., J. A. Webster, K. T. McCalister, M. W. Mallon and M. A. Steinhardt (2005). “Reliability and validity of a single-item measure of job satisfaction.” Am J Health Promot 19(3): 194-198. Edwards, J. A., S. Webster, D. Van Laar and S. Easton (2008). “Psychometric analysis of the UK Health and Safety Executive’s Management Standards work-related stress Indicator Tool.” Work & Stress 22(2): 96-107. Fischer, J., S. Kumar and S. Hatcher (2007). “What Makes Psychiatry such a Stressful Profession? A Qualitative Study.” Australasian Psychiatry 15(5): 417-421. Fukui, S., M. P. Salyers, G. Morse and A. L. Rollins (2021). “Factors that affect quality of care among mental health providers: Focusing on job stress and resources.” Psychiatr Rehabil J 44(3): 204-211. Goldacre, M. J., G. Turner, S. Fazel and T. Lambert (2018). “Career choices for psychiatry: national surveys of graduates of 1974–2000 from UK medical schools.” British Journal of Psychiatry 186(2): 158-164. Greig, F. (2020). “COVID-19, medical education and the impact on the future psychiatric workforce.” BJPsych Bulletin 45(3): 179-183.

40 HEALTH DIALOGUE NOVEMBER 2021

Greig, F. (2021). “COVID-19, medical education and the impact on the future psychiatric workforce.” BJPsych Bulletin 45(3): 179-183. Gunasekara, I., S. Patterson and J. G. Scott (2017). “‘What makes an excellent mental health doctor?’ A response integrating the experiences and views of service users with critical reflections of psychiatrists.” Health & Social Care in the Community 25(6): 1752-1762. Hämmig, O. (2018). “Explaining burnout and the intention to leave the profession among health professionals – a cross-sectional study in a hospital setting in Switzerland.” BMC Health Services Research 18(1): 785. Hann, M., D. Reeves and B. Sibbald (2011). “Relationships between job satisfaction, intentions to leave family practice and actually leaving among family physicians in England.” Eur J Public Health 21(4): 499-503. Holmqvist, R. and B. A. Armelius (2006). “Sources of psychiatric staff members’ feelings towards patients and treatment outcome.” Psychol Psychother 79(Pt 4): 571-584. Houdmont, J., L. Jachens, R. Randall, S. Hopson, S. Nuttall and S. Pamia (2019). “What Does a Single-Item Measure of Job Stressfulness Assess?” International journal of environmental research and public health 16(9): 1480. HSE (2012). Health and Safety Executive Psychosocial Working Conditions in Britain in 2010 Ike, I. D., M. Durand-Hill, E. Elmusharaf, N. Asemota, E. Silva, E. White and W. I. Awad (2021). “NHS staff mental health status in the active phase of the COVID-19 era: A staff survey in a large London hospital.” General Psychiatry 34(2). Johnson, J., L. H. Hall, K. Berzins, J. Baker, K. Melling and C. Thompson (2018). “Mental healthcare staff well-being and burnout: A narrative review of trends, causes, implications, and recommendations for future interventions.” International Journal of Mental Health Nursing 27(1): 20-32. Kavanagh, A. and L. Watters (2010). “Consultant psychiatrists’ experience of workplace violence–a national survey.” Irish journal of psychological medicine 27(2): 77-81. Kumar, S. (2007). “Burnout in psychiatrists.” World psychiatry : official journal of the World Psychiatric Association (WPA) 6(3): 186-189. Kumar, S., J. Fischer, E. Robinson, S. Hatcher and R. N. Bhagat (2007). “Burnout and Job Satisfaction in New Zealand Psychiatrists: a National Study.” International Journal of Social Psychiatry 53(4): 306-316. Kumar, S., S. Hatcher, G. Dutu, J. Fischer and E. Ma’u (2010). “Stresses Experienced By Psychiatrists and Their Role in Burnout: a National Follow-Up Study.” International Journal of Social Psychiatry 57(2): 166-179. Kumar, S., P. Sinha and G. Dutu (2012). “Being satisfied at work does affect burnout among psychiatrists: A national follow-up study from New Zealand.” International Journal of Social Psychiatry 59(5): 460-467. Looi, J. C. L. and P. Maguire (2019). “Surviving the Heart of Darkness: reflections on the experience of psychiatrists working in public mental health services in Australia.” Australasian Psychiatry 27(6): 634-636. Maslach, C. and M. P. Leiter (2016). “Understanding the burnout experience: recent research and its implications for psychiatry.” World psychiatry: official journal of the World Psychiatric Association (WPA) 15(2): 103-111. McLoughlin, C., S. Casey, A. Feeney, D. Weir, A. A. Abdalla and E. Barrett (2021). “Burnout, Work Satisfaction, and Well-being Among Non-consultant Psychiatrists in Ireland.” Academic Psychiatry 45(3): 322-328. McNicholas, F., S. Sharma, C. Oconnor and E. Barrett (2020). “Burnout in consultants in child and adolescent mental health services (CAMHS) in Ireland: a cross-sectional study.” BMJ Open 10(1): e030354.

WWW.ASMS.ORG.NZ HEALTH DIALOGUE 41

MoH (2021). https://www.health.govt.nz/publication/annual-update-key-results-2019-20-new-zealandhealth-survey. Wellington, Ministry of Health. Montgomery, A., M. van der Doef, E. Panagopoulou and M. P. Leiter (2020). Connecting Healthcare Worker Well-Being, Patient Safety and Organisational Change The Triple Challenge. Cham, Springer International Publishing. Morgan, L. J., G. M. Finn and P. A. Tiffin (2021). “Are efforts to recruit to psychiatry closing the stable door after the horse has bolted? Knowledge and attitudes towards a career in psychiatry amongst secondary (high) school students: a UK-based cross-sectional survey.” Journal of Mental Health: 1-8. MSOD(2021). National report on students graduating medical school in New Zealand in 2016-2020. Medical Schools Outcomes Database and Longitudinal Tracking (MSOD) project. Otago Medical School, Otago University. Nguyen, Q. A. and A. D. Tran (2021). “Job satisfaction and turnover intention of preventive medicine workers in northern Vietnam: Is there any relationship?” Health Services Insights 14: 1178632921995172. Priebe, S. and R. McCabe (2008). “Therapeutic relationships in psychiatry: The basis of therapy or therapy in itself?” International Review of Psychiatry 20(6): 521-526. Reynolds, M., A. McCombie, M. Jeffery, R. Mulder and F. Frizelle (2021). “Impact of burnout on empathy.” The New Zealand Medical Journal (Online) 134(1530): 12-20. Rotstein, S., A. R. Hudaib, A. Facey and J. Kulkarni (2019). “Psychiatrist burnout: a meta-analysis of Maslach Burnout Inventory means.” Australas Psychiatry 27(3): 249-254. Shanafelt, T. D., C. M. Balch, G. Bechamps, T. Russell, L. Dyrbye, D. Satele, P. Collicott, P. J. Novotny, J. Sloan and J. Freischlag (2010). “Burnout and medical errors among American surgeons.” Ann Surg 251(6): 995-1000. Shields, G., R. Ng, A. Ventriglio, J. Castaldelli-Maia, J. Torales and D. Bhugra (2017). “WPA Position Statement on Recruitment in Psychiatry.” World Psychiatry 16(1): 113-114. Singh, J., M. Karanika-Murray, T. Baguley and J. Hudson (2020). “A systematic review of job demands and resources associated with compassion fatigue in mental health professionals.” International Journal of Environmental Research and Public Health 17(19): 1-28. Snyder, T. G. and S. Kumar (2004). “A Perspective on the Problems in Retaining Psychiatrists in New Zealand.” Australasian Psychiatry 12(4): 401-403. Stevens, K., C. Davey and A. A. Lassig (2020). “Association of Weekly Protected Nonclinical Time With Resident Physician Burnout and Well-being.” JAMA Otolaryngology–Head & Neck Surgery 146(2): 168-175. Walsh, G., B. Hayes, Y. Freeney and S. McArdle (2019). “Doctor, how can we help you? Qualitative interview study to identify key interventions to target burnout in hospital doctors.” BMJ Open 9(9): e030209. West, C. P., L. N. Dyrbye, D. V. Satele, J. A. Sloan and T. D. Shanafelt (2012). “Concurrent Validity of SingleItem Measures of Emotional Exhaustion and Depersonalization in Burnout Assessment.” Journal of General Internal Medicine 27(11): 1445-1452. West, C. P., L. N. Dyrbye, J. A. Sloan and T. D. Shanafelt (2009). “Single Item Measures of Emotional Exhaustion and Depersonalization Are Useful for Assessing Burnout in Medical Professionals.” Journal of General Internal Medicine 24(12): 1318. Zhang, Q., M. C. Mu, Y. He, Z. L. Cai and Z. C. Li (2020). “Burnout in emergency medicine physicians: A metaanalysis and systematic review.” Medicine (Baltimore) 99(32): e21462.

42 HEALTH DIALOGUE NOVEMBER 2021

Appendix 1