Health Impact Assessment: Reducing Homicide using Drone Technology

CENTER FOR HEALTH EQUITY, LOUISVILLE METRO DEPARMENT OF PUBLIC HEALTH & WELLNESS

Health Impact Assessment: Reducing Homicide using Drone Technology Prepared by Center for Health Equity at the request of the Office of Civic Innovation & Technology

August 3, 2018

Table of Contents Introduction .................................................................................................................................................. 1 HIA Steps and Definitions ............................................................................................................................ 2 Screening ...................................................................................................................................................... 3 Scoping.......................................................................................................................................................... 4 Limitations ................................................................................................................................................ 7 Assessment ................................................................................................................................................... 8 Goal 1 ........................................................................................................................................................ 9 Goal 2 ...................................................................................................................................................... 11 Conclusion ............................................................................................................................................... 12 Recommendations ..................................................................................................................................... 13 Conclusion .................................................................................................................................................. 18 References .................................................................................................................................................. 19

Introduction In recent years, the homicide rate in Louisville, KY has spiked and is now double its 10-year average. According to the Louisville Metro Police Department (LMPD) End of the Year Crime Report, there were 117 homicides in 2016 and 107 in 2017 as opposed to 55 in 2014 and 82 in 2015 (January 4, 2018). While there are occurrences in every zip code, homicides are geographically concentrated in nine neighborhoods in the western area of the county, known as the West End. With some exception, residents in these neighborhoods are predominatelyBlack who are disproportionately navigating poverty, a pattern shaped by redlining, the historic practice of structural racism (Poe, 2015). According to the Louisville Metro Health Equity Report, Black men die by homicide at 5.5 times that of the Louisville Metro of 30 years old, making interpersonal violence an especially large contributor to premature death for Black men (Center for Health Equity, 2017).

Louisville Metro Government (LMG) has employed a range of initiatives to address the rising homicide rate, including the development of the Office for Safe and Healthy Neighborhoods in 2013 and directed funding to LMPD initiatives specific to reducing homicide and gun violence.

Most recently, LMG focused on homicide in their 2017 proposal to the Bloomberg Philanthropies’ 2018 Mayor’s Challenge which is designed to fund “bold inventive ideas that confront cities toughest problems” (2018 Mayors Challenge, 2018). LMG proposed videoequipped unmanned aerial vehicles, or drones, to complement existing LMPD gunshot detection technology. In doing so, LMG posited the drones would ultimately improve the health and safety in the most impacted neighborhoods, particularly through reducing homicide and decreasing gun violence. The proposal was one of 35 to advance to the Champion Cities round, which included pilot tests of the technology’s efficacy and perception among community.

In March 2018, Office of Civic Innovation and Technology (OCIT), the lead LMG agency for the proposal, and the Real Time Crime Data Center (RTTC) approached the Department of Public Health and Wellness’ Center for Health Equity to conduct a Health Impact Assessment (HIA) on the proposed project.

1

HIA Steps and Definitions According to Human Impact Partners, a health impact assessment is a tool which “aims to make the health impacts of public decisions explicit” and does so through the engagement of “health experts, decision-makers, and stakeholders including those with local knowledge” (2011).

Utilizing this tool, decision-makers can proactively assess the impact of a policy, program, or practice through a lens considering a holistic definition of health. HIAs offer a flexible, datadriven approach to identify potential health consequences through six steps; screening, scoping, assessment, recommendations, reporting, and monitoring/evaluation. See Figure 1 for a description of each step and its purpose.

This assessment was conducted through literature review, content expert consultation and engaging residents in the West End neighborhoods. Because there is no known literature identifying the health impacts for drone technology used in this way, the literature review focused on the goals of the project – increasing arrests and reducing rates of homicide and gun violence. This was complemented by diverse engagement activities, including community conversations, 1-on-1 interviews and focus groups. These efforts led to an analysis identifying opportunities to improve public health benefits for the proposal.

The first step, screening, occurred in March 2018 when OCIT and RTTC requested the completion of an HIA on the use of drone technology to reduce homicides (see the Screening section for further description). Because of the six-month timeline given by Bloomberg Philanthropies to report the results from prototype testing, the due date for the completed assessment was August 2018. In April 2018, the project scope was defined, engagement with stakeholders occurred June 1-July 16, and the assessment and recommendations were developed through August 1.

2

HIA Process Steps 1. Screening

Purpose Determine whether an HIA is warranted and would be valuable in the decision-making process.

2. Scoping Decide on the scope of the assessment, including data sources to use, groups/populations most likely to be impacted by the decision, possible areas of health consequences to investigate, and methodology to utilize.

3. Assessment Develop a health profile of the community, including baseline conditions for various health conditions, literature reviews, and quantitative methods to assess likely effects of the proposed project. Use community input to provide nuance, context, clarification, and prioritization 4. Recommendations Generate a series of recommendations on how to optimize the health benefits and reduce the health risks of the proposed plan. To maximize the chances that they will be adopted, recommendations should be practical, considering constraints on decision makers.

5. Reporting Produce a written report and disseminate what was learned to the community in question through varied methods, such as radio press releases, community meetings, or going door to door.

6. Monitoring and Monitor implemented recommendations to see if they are working as Evaluation expected. Evaluate the process itself, rate of implementation of recommendations, partnerships formed, and improvements to health over time.

Figure 1. Health Impact Assessment Steps and Purpose Screening Reducing homicide is critically important to improve public health, particularly considering the related intersecting health and wellness concerns. Of the greatest priority is the concentrated, abrupt, and violent premature death which inevitably leads to psychological trauma experienced by loved ones and community residents (Smith & Patton, 2016). As a measure of public health best practice, patterns in health outcomes are an embodiment of the patterned experiences with social determinants of health, or root causes (Krieger, 2005). These patterns are the result of deep, structural barriers shaped by systems of power, such as racism, which prevent residents from realities where institutional practices and policies encourage them to thrive. Geographic concentration signals the need for prioritized, innovative, and thoughtful solutions which center healing to create thriving communities for everyone.

3

While the goal to reduce homicide using drone technology is appropriate content for an HIA, it is particularly apt because there remains an opportunity to influence the proposal to improve public health. Recommendations resulting from the assessment will be used to shape the final proposal to Bloomberg Philanthropies. HIAs provide an opportunity to meaningfully improve the experiences residents have in their communities through an historically-aware, collaborative approach by decision-makers which is shaped by those who are most impacted by the decision and framed through a lens which seeks to improve public health for all.

Specifically, this HIA will provide an equitable, holistic assessment of the relationship between autonomous drone technology, homicide reduction, and public health. In doing so, the HIA will (1) assess the health impacts of deploying autonomous drone camera platforms in West End neighborhoods with existing gunshot detection technology, (2) assess this autonomous drone technology’s potential impact in reducing homicide and gun violence, (3) offer recommendations for maximizing positive health outcomes and minimizing negative health outcomes, and (4) identify challenges to implementing the recommendations.

Scoping The scoping phase of an HIA outlines “objectives…and an outline” for the research with a particular focus on identifying which “health effect” should be prioritized, which “concerns have stakeholders expressed,” and “who will be affected by the…project” (The Pew Charitable Trusts, August 26, 2014). OCIT proposed the following goals for drone technology; to improve the public health and safety of neighborhoods through 1) an increased quantity of solved homicide cases, 2) reduced homicide rates and deter gun violence (Simrall, 2017). In preliminary engagement conducted by OITC, residents were primarily concerned with privacy.

Considering these goals and the identified concern, this HIA prioritized identifying nuances to the resident feedback to assess potential unintended consequences and opportunities for improved health. Additionally, due to the innovative nature of this grant proposal, much of the existing literature describing similar projects are supportive for implementation but do not share evidence-based results and are not peer-reviewed. As a proxy, the conducted literature review was limited to identifying evidence relevant to the underlying assumptions of the two

4

stated goals, particularly increased arrests lead to reduced homicide rates and surveillance as a deterrent to gun violence.

The most valuable data was collected at the community level, through focus groups, interviews, and community conversations. This strategy was tailored to engage residents who would be most impacted. For example, opportunities to discuss the drone technology project were shared through word-of-mouth rather than social media channels. According to the Louisville Metro Government Digital Inclusion Plan (2017), the digital divide in Louisville means residents most impacted by neighborhoods with high rates of homicide have limited access to reliable internet to learn details about community conversation, if primarily shared electronically.

Engagement prioritizing social media would have been to the wrong audience and could have saturated the data with experiences from people who are not impacted by high rates of homicide in their neighborhood. In previous engagement outreach, residents have also shared in other spaces that personal connections are the preferred method of seeking community input. Additionally, content experts were engaged to better understand the nuances of policy and policing. See Figure 2 for identified priority areas of focus for community residents and Figure 3 for content experts.

5

Areas of Focus: Community

Reason

Residents

In Louisville, neighborhood associations often represent an older Neighborhood Associations within impacted neighborhoods

demographic who are engaged with decision-making in their neighborhoods. From initial conversations with OCIT and RTTC describing their community engagement, CHE was informed this demographic was supportive of the drone technology proposal.

Residents currently incarcerated

Residents currently incarcerated have valuable perspective on what is needed to reduce homicide Residents formerly incarcerated have valuable perspective on what is

Residents formerly incarcerated

needed to reduce homicide as well as the assumption being tested that incarceration reduces illegal activity, particularly homicide.

Participants in this Office of Safe and Healthy Neighborhood program build

Ambassador Institute Fellows

their capacity to engage in violence prevention work as community residents

Figure 2. Areas of Focus for Community Resident Engagement

6

Areas of Focus: Content

Reason

Experts Louisville Metro Police

CHE emphasized the importance of speaking with first-responding officers to

Department (LMPD) officers

the scene of gun violence or homicide, as they would be the most impacted

of various ranks

employees by the proposal. Additionally, investigative detectives, lieutenants, and sergeants were identified as important to engage.

No More Red Dots

This Louisville-based organization approach violence as a disease and seek to disrupt its transmission through the immediate interruption of violent escalation. No More Red Dots staff are well-respected in the community and their perspective, as individuals who directly intervene in violence, is valued.

American Civil Liberties

OCIT and RTTC communicated early on that they understood a significant

Union (ACLU) of Kentucky

concern for residents was privacy. ACLU was identified to discuss the nuances of these privacy concerns.

Figure 3. Areas of Focus for Content Expert Engagement Limitations Data limitations are constraints experienced by researchers with the potential to impact the assessment. The primary limitation to this project was the timeline. Because of the six-month due date for prototype testing and reporting results, this assessment was constrained to three months to engage stakeholders, analyze the data, and document the results. Additionally, the timing of the project required engagement to occur during the summer months, which is generally a busy time for residents. Next, when conducting qualitative data collection, it is preferential to record and transcribe interviews to reference specific language used. However, because identified audiences included young people currently incarcerated and people formerly incarcerated, the decision was made to refrain from recording to protect their privacy and create a safe space, so their full experiences could be shared. Rather than coding and theming transcribed data, researchers searched for common messages across engagement spaces.

7

Assessment This section describes the “baseline health status” of residents in the West End as well as the historical, structural causes for the health inequities they navigate (The Pew Charitable Trust, August 26, 2014). It is not comprehensive but a brief overview of facts critical to the analysis.

The consequences of homicide are vast and have rippling effects throughout the community.

While the research is limited, studies demonstrate residents living in neighborhoods with high rates of homicide and other violent crimes experience negative impacts to their physical and mental health, which can include psychological trauma such as post-traumatic stress disorder, grief, a sense of loss, and a reduced perception of safety (Smith, 2014; Smith & Patton, 2015; Graham et al., 2017). As homicide rates increase, survivors who witnessed the violence are also at an increased risk of gun carrying and dying by homicide (Tracy, Braga, & Papachristos, 2016; Reid et al., 2017). And, when homicide cases go unsolved, families lack feelings of closure, knowing the person responsible for the death of a loved one is likely a resident of their community (Kaste, 2015). In addition to the psychological consequences of concentrated homicide, residents in the West End inequitably experience other adverse health outcomes, including reports of “not good” mental health days, lead poisoning, infant mortality, and other chronic diseases (Center for Health Equity, 2017). The combination of these experiences has shaped a 12.6-year difference in life expectancy for residents living in the West End as opposed to other Louisville Metro neighborhoods. This is a significant public health crisis.

While there are many root causes to these health outcomes, neighborhood/built environment and the socioeconomic status of neighborhood residents are two which have a substantial influence on improving public health and reducing homicide (Center for Health Equity, 2017).

When residents have access to well-lit neighborhoods, safe sidewalks and greenspaces, and low rates of vacant property, the resulting experiences for residents is a lower occurrence of crime and violence (Culyba et al., 2016; Branas et al., 2016; Garvin et al., 2013; Wiebe et al., 2016).

Research also demonstrates investment in social protections, including policy and practice 8

decisions raising the standard of living, significantly reduce homicide rates (Rogers & Pridemore, 2013). There is a correlation to an increase in homicides when residents cannot meet their basic needs. And, because poverty socially and geographically isolates residents from the protections that come with thriving communities, individuals create their own mechanisms for navigating economics and safety outside of the law (Lee, 2011).

As stated previously, patterned experiences with root causes to health are shaped by structural systems of power (Krieger, 2005; Center for Health Equity, 2017). In particular, redlining and other structurally-racist methods of disinvesting in communities of color, specifically Louisville’s predominately-Black neighborhoods, have had long-term consequences of limited wealth building and neighborhood development which shape the current pattern of homicide (Poe, 2015; Herring & Henderson, 2016). An emerging body of research demonstrates the relationship between formerly-redlined neighborhoods and the current likelihood for there to be increased rates of violence (Jacoby et al., 2017). Because redlining resulted in the concentration of poverty coupled with neighborhoods with limited investment, residents navigate significant structural barriers to meeting basic needs for themselves and their families.

With this information guiding a baseline understanding of the health status for West End residents, the following describes each goal for the proposal, related literature, and key messages gathering during stakeholder engagement.

Goal 1: Drone surveillance will result in an increased quantity of solved homicide cases The proposed use for drone technology is to complement “existing gunshot detection software with autonomous aerial drones to enhance law enforcement response to firearm activity within high-crime areas of Louisville” (Simrall, 2017). Drones will deploy, only in safe weather conditions, to triangulated GPS coordinates of detected gunshot activity “within 90 seconds after shots are detected (based on a flight speed of 60 mph)” (2017). Drones will stream video of the neighborhood scene to the Real Time Crime Center, enabling staff to alert responders of “potential danger and collect information to help solve cases including license plates of vehicles 9

fleeing the scene and suspect hiding locations” (2017). Limited to a one-mile radius from their home locations, drones will be located on city-owned properties within prioritized hotspots, or areas with high rates of gun violence. Once LMPD arrives on the scene, drones will return to the city property. The primary goal of utilizing drone technology is to use their visual recording capability to increase the number of closed homicide cases, also known as clearance rate. The first assumption is that of the effectiveness of surveillance and its technology in aiding criminal investigations. While there is certainly some validity to these assumptions and applying them to homicide police work, they may not be ultimately sustainable, and there is some necessary nuance in approaching it from a public health perspective.

As surveillance has evolved with modern technology and become more ubiquitous in society, there is an increased body of research addressing the questions of its efficacy in reducing crime.

Research demonstrates surveillance cameras can be valuable to criminal investigations, particularly as a cost-effective method of documentation (La Vigne, 2013), and can produce “statistically significant, moderate crime reduction” (Braga, Weisburd, & Turchan, 2018).

However, there is a complicated relationship between police surveillance and communities of color, particularly predominately-Black neighborhoods, as these policies have been shown to disproportionately prioritize communities of color (Gellman & Adler-Bell, 2017). And, while new data-driven technology, such as predictive policing, are intended to increase accuracy and efficiency of investigations, surveillance monitoring is not exempt from the impact of implicit bias.

The prospective role of the drone initiative and its surveillance capabilities were met with various feedback when shared with community. The following describe the common messages shared by stakeholders: •

Residents expressed the need for cases to be solved to show that consequences existed for inflicting the harm of homicide on a community.

•

Several residents expressed drone technology would be valuable if it brought the families experiencing loss from homicide closure.

10

•

Residents were frustrated that the surveillance focused on predominately-Black, West End neighborhoods.

•

Many expressed privacy concerns, especially the prospect of the camera capturing other community members present in a shooting that may lead to false arrests or profiling o ACLU recommended policy language to protect privacy rights (see Appendix).

•

Some residents raised similar concerns about protection of witnesses or other people who may appear on video who may face retaliation from those involved

•

Many questioned the effectiveness of arrests in addressing homicides in comparison to other methods, including addressing the trauma navigated by individuals involved in gun violence, as well as built environment initiatives, employment, and recreation opportunities for youth

Goal 2: Drone surveillance will reduce homicide rates through an increased clearance rate as well as deter additional gun violence The second goal is developed in relationship with the first. OCIT posits the increased clearance rate for homicides will lead to increased incarceration for those who commit homicide, which will reduce future homicides. A literature review was conducted to assess the relationship between incarceration and reduced homicide rates.

Incarceration as a crime reduction tactic has been especially utilized in the United States for the past 40 years, with both incarceration rates and sentence length significantly increasing (Brennan Center for Justice [Brennan], 2015). This phenomenon, also known as mass incarceration, has its origins in responding to increasing crime rates during the 1970s and 1980s which led to more than 500% rise in incarceration while the general population rose by 40% (The Sentencing Project, 2005; Brennan, 2015). The relationship between incarceration and crime is complex and requires a nuanced approach for understanding. Research demonstrates that, while there is correlation between reduction in crime in the past several decades and an increase in incarceration, this generally only applies to property crime (Spelman, 2007). There is little association between an increase in incarceration and reduction in violent crime. For example, research demonstrates that, considering the reduction in crime during the 1990s, 11

incarceration was only responsible for about 25% with the remaining 75% are attributed to responses to factors outside of incarceration, including increased employment/wages (The Sentencing Project, 2005; Stemen, 2007; Brennan, 2015, Stemen, 2017). A fundamental reason for increased incarceration in response to crime “posits that incarceration not only incapacitates past offenders, but also deters future ones” (Brennan, 2015). However, longer prison sentences lead to increased structural barriers and limited social support upon release which can significantly impact recidivism (The Sentencing Project, 2005; Stemen, 2017).

The following describe the common messages shared from stakeholders: •

Residents agreed that increased surveillance would improve the likelihood of identifying the individual responsible for homicide

•

Residents do not believe that incarceration results in reduced crime, particularly due to the unaddressed trauma of individuals involved in homicides and structural barriers faced once the sentence is complete.

•

Residents disagreed that drone technology would act as a deterrent to gun violence because 1) they are concerned residents will attempt to shoot at drones and put people in danger of falling bullets and 2) they are concerned this will displace crime rather than address the reasons homicides are occurring.

•

Most importantly, residents recognized this proposal situates the drone technology as a response mechanism to homicides rather than used a method to prevent homicides.

Conclusion If anything can be immediately concluded upon, it is that this project and its relationship to community is complex. While the relationship between increased video surveillance leading to increased arrests resonates as possible and even desired by some, residents had concerns about their privacy and the police ownership over the technology. Additionally, residents were clear in their understanding this technology would not increase the safety in their communities, both through the targeted gunfire towards the drones and it being used as a reaction to homicide, rather than preventative method. A review of related literature identified benefits of camera surveillance for solving crime but did not support increased incarceration as a means of reducing 12

homicide. In conclusion, the results of the assessment offer opportunities to improve the proposal in ways that will significantly increase the likelihood of sustainably meeting the goals to reduce homicide, gun violence, and better neighborhood health and safety.

Recommendations Reshaping the experiences residents have in their communities through reducing the homicide rate is a significant public health priority. These recommendations are guided by the literature review and community engagement. They are intended to minimize adverse health impacts and increase the public health benefit of using drone technology to reduce homicide. There are five recommendations put forth to improve the public health benefits and reduce violence, three of which are specific to this proposal. The innovative nature of the proposed use of drone technology to complement existing gunshot detection technology is an opportunity to utilize a completely new approach in a way that fundamentally improves the experiences of residents in Louisville Metro. While Figure 4 describes the full series of recommendations and challenges, it is necessary to detail the recommendations specific to the proposal.

Research demonstrated that surveillance can add value to investigative tools and residents agreed the video capabilities could lead to the benefit of identifying those responsible for homicide. However, many also recognized the use of this technology placed an emphasis on responding to homicides rather than preventing its occurrence. This was an important, reoccurring message as people imagined a possibility where the trauma of homicide could be completely prevented. This community feedback guided the development of the first recommendation, to utilize the drone technology as a primary tool for emergency medical response to improve the transportation time to care for gunshot victims. The lifesaving benefits of improved medical response is well-documented (Pons et al., 2005; Band et al., 2011; Wandling et al., 2018). By prioritizing the use of drone technology for medical response time and situational awareness for first responders, there is an opportunity to prevent homicide – achieving the stated goal of homicide reduction.

13

In connection to the first recommendation, the second opportunity is to infuse the drone technology utilization into the existing ecosystem of evidence-based violence prevention initiatives, particularly hospital-based Pivot to Peace and community-based No More Red Dots.

Additionally, residents shared a desire to built infrastructure and neighborhood development to be prioritized. Considering the previously-named research drawing connections between the surrounding built environment and homicide rate, this proposal offers an opportunity to meaningfully prioritize Public Works projects in alignment with the drone technology activation.

In general, this creates collaboration between agencies to simultaneously, and sustainably, address homicide as a health outcome and a contributing root cause. Finally, while the engagement for this assessment demonstrated a shared vision to reduce homicide, residents also expressed the desire to be meaningfully included in ongoing dialogue of shared ownership over the transparency and accountability goals for the project. This final proposal-specific recommendation is designed to facilitate a collaborative, ongoing relationship between residents and LMG related to the use and fidelity of drone technology. It is critical to develop a process for accountability which involves residents, prioritizing those who engaged in the preliminary as well as the health impact assessment data collection.

Together, these recommendations have the potential to significantly improve the potential for achieving the stated goals through a sustainable, trauma-informed initiative.

14

Priority Audience Office of Civic

Recommendation 1) Reframe proposal from

Description

Challenges

Reconvene stakeholders with an emphasis on

The short timeline will require quick strategizing; It is

Innovation &

drone technology used as

Emergency Medical Services, Public Works, Office

important to consider the necessary relationship

Technology (OCIT)

police surveillance for

of Safe and Healthy Neighborhoods, No More Red

between first responders and this technology. While

increased arrests to

Dots, and Pivot to Peace to strategize what this

research shows incarceration is less effective at reducing

improve transportation

proposal and team would look like if the goal was

crime rates, community has also identified a hesitancy to

time for gunshot victims

to prevent interpersonal violence from becoming

increase police surveillance in this way but are eager for

to the Emergency Room

homicide.

innovative solutions to prevent homicide. Because police will need to be involved in the proposed project, as the gunshot detection technology alerts LMPD, embedded processes for accessible transparency and accountability will need to be identified prior to resubmitting the proposal.

OCIT

2) Prioritize the connection

Reconvene stakeholders with an emphasis on

It is possible there will be hesitation from Metro

between drone

Emergency Medical Services, Public Works, Office

employees to prioritize Public Works projects at locations

technology use and

of Safe and Healthy Neighborhoods, No More Red

where the drone technology was recently activated. This

existing resources

Dots, and Pivot to Peace to strategize what this

will require a dialogue to identify innovative solutions

proposal and team would look like if the goal was

taking into consideration the fear employees may have.

to prevent interpersonal violence from becoming homicide.

15

OCIT

3) Identify accountability

Explicitly define that activation of drone

Differing definitions for ‘transparency’ and

and transparency

technology will not be used for ongoing

‘accountability’ will impact the relationship between

measures

surveillance by LMPD; Develop Community

community residents and LMG. These will need to be

Advisory Board to work with LMG to develop

defined, agreed upon, and used to guide the roles of both

transparency and accountability goals (residents

the Advisory Board and LMG.

who engaged the HIA process should be prioritized for participation); utilize proposed policy language from ACLU to protect privacy rights OCIT, Office for Safe

4) Conduct an assessment

Residents expressed a desire for resources to

This is a time-intensive project which would require

and Healthy

on the six-point violence

prioritized for prevention interventions. There is

cross-functional collaboration and clear guidelines for

Neighborhoods

prevention plan

currently a six-point plan that allocates resources

improving the return on investment with resource

(OSHN), Public Health

(enforcement,

to each pillar, which includes prevention. It is

allocation recommendations.

and Wellness

intervention, prevention,

recommended to conduct an assessment on the

community mobilization,

pillars to assess the efficacy and return on

organizational change,

investment that each dollar spent has on violence

and reentry) including

prevention.

total money allocated as well as return on investment

16

OCIT, OSHN, Public Health and Wellness

5) Convene a cross-

Research demonstrates a relationship between

This team would require representation from a variety of

functional team to

structural racism and concentrated homicides. It is

equity and trauma content experts as well as residents.

identify structural

critical to recognize the role that policies have in

Most importantly, recommendations from this team

opportunities to address

the community-level trauma experienced by

would need to be actionable and resourced to have the

the community-level

residents in the West End. To interrupt the high

necessary impact.

trauma experienced in

rates of homicide, it is critical to address to cycle of

neighborhoods where

trauma experienced by residents.

homicides are concentrated

Figure 4. Recommendations for use of drone technology

17

Conclusion To address the increased rate of homicide, Louisville Metro Government’s Office of Civic Innovation and Technology submitted a proposal to Bloomberg Philanthropies 2018 Mayor’s Challenge proposing drone technology as a complement to existing gunshot detection technology to assist police investigations. Reducing homicide in Louisville Metro is a significant public health concern as communities navigating high rates of homicide are also more likely to navigate other health inequities, including limited access to mental health care, recreation spaces, and reduced feelings of safety and community cohesion. This leads to feelings of depression, anxiety, and increased risk of homicide.

Two goals were assessed through this HIA; to improve the public health and safety of neighborhoods through 1) an increased quantity of solved homicide cases, 2) reduced homicide rates and deter gun violence. A combination of literature review and stakeholder engagement identified opportunities to improve the public health benefit of this proposal. Police officers of various ranks, community residents, young people currently incarcerated, individuals formerly incarcerated, and related content experts shared similar desires to reduce the rate of homicide and address the immediacy of the trauma. However, the hesitancy from community residents about this technology’s emphasis as police surveillance created an opportunity to reframe the project for their benefit.

Utilizing drone technology as a complement to existing gunshot detection technology is innovative and, through the lens of public health, can prevent homicides through a focus on saving lives. With existing violence prevention programs and initiatives, this technology has the potential to fit into an ecosystem which connects emergency first responders, hospitals, violence interrupters, and wraparound services to better serve residents at risk of dying by homicide. Additionally, the relationship with Public Works has the potential to address built environment as a root cause to homicide. There are additional opportunities for assessment and action, including analyzing resource allocation and return on investment as well as identifying opportunities to structurally address community-level trauma.

18

Appendix The following is suggested language for policies related to drone technology, recommended by Kate Miller at the Louisville, Kentucky office of the American Civil Liberties Union.

An act to regulate the use of unmanned aerial vehicles.

Strategy Notes:

Section 1: If your state uses a particular term to describe law Section 1. Definitions. In this chapter: enforcement agents (for example, “peace officers”), you may want to substitute that term for “agent of the state or (a) Agent of the state or any political subdivision thereof has been given meaning in [Insert section any political subdivision of state code providing this definition]. thereof” both in the definitions and throughout.

(b) The term ‘‘unmanned aerial vehicle” means an aircraft that is operated without the possibility of direct human intervention from within or on the aircraft.

If your state does not have a

statutory definition of “Agent of the state or any political subdivision thereof” or an Section 2. Use of Unmanned Aerial Vehicles. analogous term, here is a definition you can use: “Agent of the state or any political subdivision thereof” means any stateAdministration or local agency, (a) Any use of unmanned aerial vehicles must fully comply with all Federal Aviation including, but not limited to, a requirements and guidelines, and acquisition of unmanned aerial vehicles must be approved by law enforcement entity or any the city council overseeing the agent of the state or any political subdivision other thereof seeking entity, investigative such acquisition. agency, department, division, bureau, board, or commission, anypolitical individual acting or (b) Except as provided in subsection (c), it is unlawful for an agent of the state ororany purporting to act for or on subdivision thereof to operate an unmanned aerial vehicle or to disclose or receive information behalf of a state or local acquired through the operation of an unmanned aerial vehicle. agency.

(c) Exceptions (1) Consent- It shall not be unlawful under this chapter to disclose or receive information about any person acquired through the operation of an unmanned aerial vehicle if such person has given written consent to such disclosure.

(2) Exception for Emergency Situations- It shall not be unlawful under this chapter for an agent of the state or any political subdivision thereof to operate an unmanned aerial vehicle, and for information from such operation to be disclosed or received, if the unmanned aerial vehicle is used in circumstances in which it is reasonable to believe that there is an imminent threat to the life or safety of a person, subject to the following limitations: 19

(A) the request shall document the factual basis for the emergency and (B) not later than 48 hours after the agent of the state or any political subdivision thereof begins operation of an unmanned aerial vehicle, a supervisory official shall file a sworn statement setting forth the grounds for the emergency access.

(3) Warrant or Order(A) Warrant- An unmanned aerial vehicle may be operated and information from such operation disclosed in order to collect information from a non-public area only pursuant to a warrant issued under [Insert reference to state rules of criminal procedure for issuing warrants].

(B) Order- An unmanned aerial vehicle may be operated and information from such Section 3(B): operation disclosed in order to collect information from a public area pursuant Obviously, if your state to a warrant authorized under subsection (3)(A) or pursuant to an order issued and prohibits surveillance dataof collection based by any court that is a court of competent jurisdiction if the agent the state or on reasonable suspicion, take any political subdivision thereof offers specific and articulable facts this part out and remove demonstrating reasonable suspicion of criminal activity, that “non-public� the operation of from 3(A). the public unmanned aircraft system will uncover such activity, and that alternative methods of data collection are either cost-prohibitive or present a significant risk to any person’s bodily safety. Such an order shall not be issued for a period greater than 48 hours. Extensions of an order may be granted but shall be no longer than the authorizing judge deems necessary to achieve the purposes for which it was granted and in no event for longer than thirty days.

(4) Exception for non-law enforcement operations - It shall not be unlawful under this chapter for an agent of the state or any political subdivision thereof to operate an unmanned aerial vehicle and for information from such operation to be disclosed if no part of any information and no evidence derived from such operation may be received in evidence in any trial, hearing, or other proceeding in or before any court, grand jury, department, officer, agency, regulatory body, legislative committee, or other authority of the [State] or a political subdivision thereof, or for any intelligence purpose. (d) When unmanned aerial vehicles are used pursuant to subsection (c), they shall be operated in a manner to collect data only on the target and to avoid data collection on individuals, homes, or areas other than the target. Neither facial recognition nor other biometric matching technology may be used on non-target data collected by an unmanned aerial vehicle. (e) Unmanned aerial vehicles may not be equipped with lethal or non-lethal weapons of any kind.

Section 3. Data Retention

20

(a) No data collected on an individual, home, or area other than the target that justified deployment may be used, copied, or disclosed for any purpose. Such data must be deleted as soon as possible, and in no event later than 24 hours after collection. (b) Whenever an agent of the state or any political subdivision thereof uses an unmanned aerial vehicle, no part of information acquired and no evidence derived therefrom may be received in evidence in any trial, hearing, or other proceeding in or before any court, grand jury, department, officer, agency, regulatory body, legislative committee, or other authority of the [State] or a political subdivision thereof if the disclosure of that information would be in violation of this chapter.

Section 4. Emergency Situations Exception

Section 4: This section is optional. If your state is unlikely to pass drones legislation without a national (a) Emergency Situation Exception- Notwithstanding any other provision of security this chapter, an you should exception, keep this language agent of the state or any political subdivision thereof may operate an unmanned aerial in; otherwise, you can remove it.

vehicle and disclose information from such operation if—

(1) such agent of the state or any political subdivision thereof reasonably determines that an emergency situation exists that-(A) involves-(i) immediate danger of death or serious physical injury to any person; (ii) conspiratorial activities threatening the national security interest; or (iii) conspiratorial activities characteristic of organized crime; and (B) requires operation of an unmanned aerial vehicle before a warrant or order authorizing such interception can, with due diligence, be obtained; (2) there are grounds upon which such a warrant or order could be entered to authorize such operation; and (3) an application for a warrant or order approving such operation is made within 48 hours after the operation has occurred or begins to occur. (b) Failure To Obtain a Warrant or Order(1) TERMINATION OF ACQUISITION- In the absence of a warrant or order, an operation of an unmanned aerial vehicle carried out under subsection (a) shall immediately terminate when the information sought is obtained or when the application for the warrant or order is denied, whichever is earlier.

21

(2) PROHIBITION ON USE AS EVIDENCE- In the event such application for approval is denied, the information obtained from the operation of an unmanned aerial vehicle shall be treated as having been obtained in violation of this chapter and an inventory shall be served on the person named in the application.

Section 5. Delayed Notice.

(a) An agent of the state or any political subdivision thereof under Section 2(c)(3) or Section 4 of this title may, where a court order or warrant is sought, include in the application a request, which the court shall grant, for an order delaying the notification required under [Insert reference to state rules of criminal procedure for issuing warrants] for a period not to exceed ninety days, if the court determines that there is reason to believe that notification of the existence of the court order may have an adverse result described in paragraph (b) of this subsection. (b) An adverse result for the purposes of paragraph (a) of this subsection is(1) endangering the life or physical safety of an individual; (2) flight from prosecution; (3) destruction of or tampering with evidence; (4) intimidation of potential witnesses; or (5) otherwise seriously jeopardizing an investigation or unduly delaying a trial. (c) Extensions of the delay of notification of up to ninety days each may be granted by the court upon application, or by certification by the agent of the state or any political subdivision thereof. (d) Upon expiration of the period of delay of notification, the agent of the state or any political subdivision thereof shall serve upon, or deliver by registered or first-class mail to, a copy of the warrant or order upon the person or persons upon whom information was collected together with notice that-(1) states with reasonable specificity the nature of the law enforcement inquiry; and (2) informs the person or persons upon whom information was collected -(A) that notification of such customer or subscriber was delayed; (B) what governmental entity or court made the certification or determination pursuant to which that delay was made; and 22

(C) which provision of this chapter allowed such delay.

Section 6. Administrative Discipline

(a)

Section 6.

You may want to consider having some sort of discipline section to give the bill teeth.

Administrative Discipline- If a court or appropriate department or agency determines that an

agent of the state or any political subdivision thereof has violated any provision of this chapter, and the court or appropriate department or agency finds that the circumstances surrounding the violation raise serious questions about whether or not the agent of the state or any political subdivision thereof acted willfully or intentionally with respect to the violation, the department or agency shall, upon receipt of a true and correct copy of the decision and findings of the court or appropriate department or agency promptly initiate a proceeding to determine whether disciplinary action against the agent of the state or any political subdivision thereof is warranted. If the head of the department or agency involved determines that disciplinary action is not warranted, such head shall notify the Inspector General with jurisdiction over the department or agency concerned and shall provide the Inspector General with the reasons for such determination. (b)

Improper Disclosure Is Violation- Any willful disclosure or use by an agent of the state or any political subdivision thereof of information beyond the extent permitted by this chapter is a violation of this chapter for purposes of this section.

Section 7. Reporting

Section 7.

This is a lot of required reporting. Subsection (a) is taken from Seattle’s draft (a) In June of each year, any agent of the state or political subdivision thereof that uses unmanned ordinance and Subsections (b), (c), and (d) are taken aerial vehicles shall report to the legislature and make public on its website: from the federal legislation.

It would be useful (1) The number of times an unmanned aerial vehicle was used, organized by the types of to include some sort of reporting incidents and the types of justification for deployment; requirements, but you can decide whether to include (2) The number of crime investigations aided by the use of unmanned aerialsome vehicles.

A or all of these and description of how the unmanned aerial vehicle was helpful to each investigation should which would be more useful – and more palatable – in be included; your state.

(3) The number of uses of unmanned aerial vehicles for reasons other than criminal investigations. A description of how the unmanned aerial vehicle was helpful in each instance should be included; (4) The frequency and type of data collected on individuals or areas other than targets; and 23

(5) The total cost of their unmanned aerial vehicle program. (b) In January of each year, any judge who has issued an order (or an extension thereof) under Section 2(c)(3) or Section 4 that expired during the preceding year, or who has denied approval during that year, shall report to the [Administrative Office of the Courts]— (1) the fact that an order or extension was applied for; (2) the kind of order or extension applied for; (3) the fact that the order or extension was granted as applied for, was modified, or was denied; (4) the period of unmanned aerial vehicle use authorized by the order, and the number and duration of any extensions of the order; (5) the offense specified in the order or application, or extension of an order; and (6) the identity of the applying agent of the state or political subdivision thereof making the application and the person authorizing the application. (c) In March of each year the Attorney General, an Assistant Attorney General specially designated by the Attorney General, or the principal prosecuting attorney of [State], or the principal prosecuting attorney for any political subdivision of the state, shall report to the [Administrative Office of the Courts]-(1) the information required by paragraphs (1) through (6) of subsection (b) of this section with respect to each application for an order or extension made during the preceding calendar year; (2) a general description of the information gathered under such order or extension, including (D) the approximate nature and frequency of incriminating conduct gathered, (E) the approximate number of persons upon whom information was gathered, and (F) the approximate nature, amount, and cost of the manpower and other resources used in the collection; (3) the number of arrests resulting from information gathered and the offenses for which arrests were made; (4) the number of trials resulting from such information; (5) the number of motions to suppress made with respect to such information, and the number granted or denied; 24

(6) the number of convictions resulting from such information and the offenses for which the convictions were obtained and a general assessment of the importance of the information; and (d) In June of each year the Director of the [Administrative Office of the Courts] shall transmit to the legislature and post on its website a full and complete report concerning the number of applications for orders authorizing or approving operation of unmanned aerial vehicles or disclosure of information from the operation of unmanned aerial vehicles pursuant to this chapter and the number of orders and extensions granted or denied pursuant to this chapter during the preceding calendar year. Such report shall include a summary and analysis of the data required to be filed with the [Administrative Office] by subsections (b) and (c) of this section.

25

References 2018 Mayors Challenge: The Time is Now. (2018). Retrieved from https://mayorschallenge.bloomberg.org/

Band, R., Pryor, J., Gaieski, D., Dickinson, E., Cummings, D., & Carr, B. (2011). Injury-adjusted mortality of patients transported by police following penetrating trauma. Academic Emergency Medicine : Official Journal of the Society for Academic Emergency Medicine,1 8(1), 32-7. doi:10.1111/j.1553-2712.2010.00948.x

Braga, A., Weisburd, D., & Turchan, B. (2018). Focused deterrence strategies and crime control: An updated systematic review and meta-Analysis of the empirical evidence. Criminology & Public Policy, 17(1), 205-250. doi:10.1111/1745-9133.12353

Center for Health Equity. (2017). 2017 Health Equity Report: Uncovering the Root Causes to Our Health. Retrieved from healthequityreport.com

Center for Popular Democracy, Law for Black Lives, & Black Youth Project 100. Freedom to Thrive: Reimagining Safety & Security in Our Communities. Dhondt, G. L. (2012). The relationship between mass incarceration and crime in the Neoliberal period in the United States(Doctoral dissertation). ScholarWorks@UMassAmherst.

Culyba, A., Jacoby, S., Richmond, T., Fein, J., Hohl, B., & Branas, C. (2016). Modifiable neighborhood features associated with adolescent homicide. Jama Pediatrics, 170(5), 473-80. doi:10.1001/jamapediatrics.2015.4697

Garvin, E., Branas, C., Keddem, S., Sellman, J., & Cannuscio, C. (2013). More than just an eyesore: Local insights and solutions on vacant land and urban health. Journal of Urban Health:Bulletin of the New York Academy of Medicine,90(3), 412-426. doi:10.1007/s1152 26

4-012-9782-7

Graham, P., Yaros, A., Lowe, A., & McDaniel, M. (2017). Nurturing environments for boys and men of color with trauma exposure. Clinical Child and Family Psychology Review, 20(2), 105-116. doi:10.1007/s10567-017-0241-6

Herring, C., & Henderson, L. (2016). Wealth inequality in black and white: Cultural and structural sources of the racial wealth gap. Race and Social Problems, 8(1), 4-17. doi:10.1007/s12552-016-9159-8

Human Impact Partners. (2011). A Health Impact Assessment Toolkit: A Handbook to Conducting HIA. 3rd Edition. Retrieved from https://humanimpact.org/hipprojects/a health-impact-assessment-toolkit-a-handbook-to-conducting-hia/

Jacoby, S., Dong, B., Beard, J., Wiebe, D., & Morrison, C. (2018). The enduring impact of historical and structural racism on urban violence in Philadelphia. Social Science & Medicine (1982), 199, 87-95. doi:10.1016/j.socscimed.2017.05.038

Kaste, M. (2015, March 30). Open Cases: Why One-Third Of Murders In America Go Unresolved.

Retrieved from https://www.npr.org/2015/03/30/395069137/open-cases-why-onethird-of-murders-in-america-go-unresolved

Krieger, N. (2005). Embodiment: A conceptual glossary for epidemiology. Journal of Epidemiology and Community Health, 59(5), 350-350. doi:10.1136/jech.2004.024562

Lee, M. (2011). Reconsidering culture and homicide. Homicide Studies, 15(4), 319-340.

Louisville Metro Government. (May 2017). Louisville’s Digital Inclusion Plan. Retrieved from http://digitalinclusion.louisvilleky.gov/ 27

Louisville Metro Police Department. (January 4, 2018). 2012-2017 End of the Year Crime Report.

Retrieved from https://www.louisville-police.org/ArchiveCenter/ViewFile/Item/97

Poe, J. (2015). Redlining Louisville: The History of Race, Class, and Real Estate. Retrieved from http://www.arcgis.com/apps/MapSeries/index.html?appid=a73ce5ba85ce4c3f80d365a b1ff89010

Pons, P., Haukoos, J., Bludworth, W., Cribley, T., Pons, K., & Markovchick, V. (2005). Paramedic response time: Does it affect patient survival? Academic Emergency Medicine, 12(7), 594-600. doi:10.1197/j.aem.2005.02.013

Reid, J., PhD, Richards, T., PhD, Loughran, T., PhD, & Mulvey, E., PhD. (2017). The relationships among exposure to violence, psychological distress, and gun carrying among male adolescents found guilty of serious legal offenses: A longitudinal cohort study. Annals of Internal Medicine,166(6), 412-418. doi:10.7326/M16-1648

Roeder, O., Eisen, L., & Bowling, J. (2015). What Caused The Crime Decline? Retrieved July 26, 2018, from Brennan Center for Justice website: https://www.brennancenter.org/sites/default/files/analysis/What_Caused_The_Crime_ Decline.pdf

Rogers, M., & Pridemore, W. (2013). The effect of poverty and social protection on national homicide rates: Direct and moderating effects. Social Science Research, 42(3), 584-595. doi:10.1016/j.ssresearch.2012.12.005

Roodman, D. (2017, September). The impacts of incarceration on crime. Retrieved from https://blog.givewell.org/wp-content/uploads/2017/09/The-impacts-of-incarcerationon-crime-10.pdf?_sm_au_=iVVF5WnJMS0sslP5 28

Smith, J., & Patton, D. (2016). Posttraumatic stress symptoms in context: Examining trauma responses to violent exposures and homicide death among black males in urban neighborhoods. The American Journal of Orthopsychiatry,86(2), 212-23. doi:10.1037/ort0000101

Smith, S. (2014). Traumatic loss in low-income communities of color. Focus, 31(1), 32-34.

Stemen, D. (2007, January). Reconsidering Incarceration: New Directions for Reducing Crime(Publication). Retrieved July 26, 2018, from Vera Institute for Justice website: https://www.prisonpolicy.org/scans/vera/379_727.pdf

Stemen, D. (2017, July). The Prison Paradox: More Incarceration Will Not Make Us Safer Retrieved July 26, 2018, from Vera Institute of Justice website: https://storage.googleapis.com/vera-web-assets/downloads/Publications/for-therecord-prison-paradox-incarceration-not-safer/legacy_downloads/for-the-recordprison-paradox_02.pdf

The Pew Charitable Trusts. (August 26, 2014). The HIA Process. Retrieved from http://www.pewtrusts.org/en/research-and-analysis/articles/2014/08/28/the-hiaprocess

Tracy, M., Braga, A. A., & Papachristos, A. V. (2016). The Transmission of Gun and Other Weapon-Involved Violence Within Social Networks. Epidemiologic Reviews, 38(1), 70– 86. http://doi.org/10.1093/epirev/mxv009

Wiebe, D., Richmond, T., Guo, W., Allison, P., Hollander, J., Nance, M., & Branas, C. (2016).

Mapping activity patterns to quantify risk of violent assault in urban environments. Epidemiology, 27(1), 32-41. doi:10.1097/EDE.0000000000000395 29

Wandling, M., Nathens, A., Shapiro, M., & Haut, E. (2018). Association of prehospital mode of transport with mortality in penetrating trauma: A trauma system-Level assessment of private vehicle transportation vs ground emergency medical services. Jama Surgery, 153(2), 107-113. doi:10.1001/jamasurg.2017.3601

30