An Evaluation of the City of Philadelphia's Kensington Encampment Resolution Pilot

University of Pennsylvania From the SelectedWorks of Dennis P. Culhane March 5, 2019 An Evaluation of the City of Philadelphia's Kensington Encampment Resolution Pilot Stephen Metraux, University of Delaware Meagan Cusack Fritz Graham, University of Pennsylvania David S Metzger, University of Pennsylvania Dennis P Culhane Available at: https://works.bepress.com/dennis_culhane/227/ :yb dednuF .D.hP ,enahluC sinneD .D.hP ,regzteM divaD .H.P.M ,maharG ztirF .S.M ,kcasuC nagaeM .D.hP ,xuarteM nehpetS toliP noituloseR tnempmacnE notgnisneK s’aihpledalihP fo ytiC eht fo noitaulavE nA 9102 ,5 HCRAM :TROPER LANIF :yb deraperP  Acknowledgements The authors are grateful for the support, assistance and contributions of the following departments and individuals throughout this evaluation: - numerous City of Philadelphia personnel who offered their time and energy to support our efforts and assist us in understanding and assessing numerous facets of this pilot, including councilpersons Maria Quinones Sanchez and Mark Squilla; the Managing Director’s Office and their divisions of Health & Human Services and Community Services; the Office of Homeless Services; the Department of Behavioral Health and Intellectual disAbility Services; Community Behavioral Health; the Philadelphia Police Department; and the Law Department; - individuals living in Kensington (both experiencing homelessness and housed) who participated in interviews and community meetings, including the civic engagement group, El Barrio es Nuestro, the Kensington and Allegheny Businesses Association, and Somerset Neighbors for Better Living; - service providers and advocates who participated in interviews and contributed to the important work of engaging individuals in need of services, including Megan Abrams, Caroline Bourscheid, Daryl Chapman, Lizanne Cooley, Chris Dean, Jeff Deeney, Brooke Feldman, Courtenay Harris-Bond, Sterling Johnson, Milan Marinkovic, Chris Moraff, Carol Rostucher, Katie Perch, Jennifer Powell Folks, April Steele, and Jeffrey Stockbridge; - the School of Social Policy & Practice of the University of Pennsylvania (especially TC Burnett) for administrative support; and - Beverly Woods, of the City of Philadelphia’s Managing Director’s Office, for serving as the liaison between the evaluators and the City, and who, in this role, tirelessly facilitated access for the evaluators to data, meetings, internal documents, and personnel related to the Encampment Resolution Pilot. All opinions, assessments and conclusions in this evaluation, unless noted otherwise, are those of the authors. Any errors and other limitations of this evaluation remain the responsibility of the authors. EXECUTIVE SUMMARY ......................................................................................................................................... 5 CHAPTER 1- INTRODUCTION ................................................................................................................................ 9 CHAPTER 2- BACKGROUND ................................................................................................................................ 10 2.1- GEOGRAPHY........................................................................................................................................................ 10 2.2- HISTORICAL CONTEXT AND RISE OF ENCAMPMENTS .................................................................................................... 11 CHAPTER 3- PHASE 1: PLANNING AND DEVELOPING THE ENCAMPMENT RESOLUTION PILOT............................. 14 3.1- USING SAN FRANCISCO AS A MODEL........................................................................................................................ 14 3.2- LEGAL CONSIDERATIONS ........................................................................................................................................ 15 3.3- FRAMEWORK FOR THE ERP .................................................................................................................................... 17 CHAPTER 4- PHASE 2: IMPLEMENTING THE ENCAMPMENT RESOLUTION PLAN .................................................. 20 4.1- TARGET POPULATION............................................................................................................................................ 20 4.2- SERVICES ............................................................................................................................................................ 23 4.2.1- Respite and Navigation Housing.............................................................................................................. 24 4.2.2- Substance Use Treatment........................................................................................................................ 25 4.2.3- Other Services .......................................................................................................................................... 26 4.2.4- Services Summary .................................................................................................................................... 26 4.3- SERVICE ENGAGEMENT.......................................................................................................................................... 27 4.3.1- Engagement Numbers ............................................................................................................................. 27 4.3.2- Police Counts............................................................................................................................................ 28 4.3.3- Service Engagement Summary ................................................................................................................ 29 4.4- PERSPECTIVES FROM PEOPLE STAYING IN ENCAMPMENTS ............................................................................................ 30 4.4.1- Living Situation ........................................................................................................................................ 30 4.4.2- Typical Day............................................................................................................................................... 31 4.4.3- Background and Service Use.................................................................................................................... 31 4.4.4- Perspective on the Encampment Closure................................................................................................. 32 4.4.5- Perspectives from People Staying in Encampments Summary ................................................................ 33 4.5- STAKEHOLDERS .................................................................................................................................................... 33 4.5.1- Community Members .............................................................................................................................. 34 4.5.2- Advocates ................................................................................................................................................ 36 4.5.3- Stakeholders Summary ............................................................................................................................ 38 4.6- ENCAMPMENT CLOSURES ...................................................................................................................................... 38 CHAPTER 5- PHASE 3: SUSTAINMENT THROUGH CONTINUED ENGAGEMENT AND MONITORING....................... 42 5.1- SERVICES USE AND OUTCOMES AMONG PEOPLE ON THE BY-NAME LIST ......................................................................... 42 5.1.1- The By-Name List ..................................................................................................................................... 43 5.1.2- Services Use Data from the CARES Integrated Data System.................................................................... 45 5.1.3- ERP Outcomes.......................................................................................................................................... 52 5.2- PERSPECTIVES ON SERVICES: PEOPLE SERVED AND ERP OFFICIALS AND PROVIDERS........................................................... 57 5.2.1- People Served Through ERP ..................................................................................................................... 57 5.2.2- ERP Officials and Providers ...................................................................................................................... 60 5.3- IMPACT OF THE ENCAMPMENT CLOSURES ON THE SURROUNDING AREA ......................................................................... 65 5.3.1- Perceived Impact ..................................................................................................................................... 66 5.3.2- Data-based Measures.............................................................................................................................. 70 5.3.3- Impact of the Encampment Closures on the Surrounding Area Summary............................................... 74 CHAPTER 6- CONCLUSION AND LESSONS LEARNED ............................................................................................ 76 6.1- LESSONS LEARNED IN THE PLANNING PHASE .............................................................................................................. 76 6.1.1- Don’t reinvent the wheel. ........................................................................................................................ 76 6.1.2- Effective coordination between participating entities is essential to overall success.............................. 76 3 6.1.3- Keep expectations in perspective............................................................................................................. 77 6.2- LESSONS LEARNED IN THE IMPLEMENTATION PHASE.................................................................................................... 77 6.2.1- Encampments are an opioid-related problem. ........................................................................................ 77 6.2.2- Encampments are a homelessness-related problem. .............................................................................. 77 6.2.3- Homelessness in Kensington is a Philadelphia problem........................................................................... 77 6.2.4- Temporary housing availability is limited by local resistance.................................................................. 78 6.2.5- Closing encampments means balancing competing interests. ................................................................ 78 6.2.6- Individual placements do little to relieve population pressures at the encampments............................. 78 6.2.7- Involving people experiencing homelessness in resolving encampments is difficult. .............................. 78 6.2.8- Effective services require removing access barriers................................................................................. 79 6.2.9- Closing the encampments is the easy part. ............................................................................................. 79 6.3- LESSONS LEARNED IN THE SUSTAINMENT PHASE......................................................................................................... 79 6.3.1- Most of the encampment population is involved with municipal services systems................................. 79 6.3.2- Use of a by-name list (BNL) is essential to coordinating individuals’ services. ........................................ 79 6.3.3- Having data makes a difference. ............................................................................................................. 79 6.3.4- Be deliberate in identifying outcomes and setting benchmarks.............................................................. 80 6.3.5- There is no model for policing an area overwhelmed by homelessness and opioid use. ......................... 80 6.3.6- Availability of short-term resources contrasts with scarcity of long-term resources............................... 80 6.3.7- Consolidating gains made by pilot program requires routinizing pilot services. ..................................... 80 6.3.8- Summer 2018 was long and difficult in Kensington................................................................................. 81 6.3.9- Crisis creates opportunity. ....................................................................................................................... 81 6.3.10- Pilots should lead to larger initiatives.................................................................................................... 81 APPENDIX A- METHODS ..................................................................................................................................... 82 APPENDIX B- MAP OF AREA CONTAINING THE TARGETED ENCAMPMENTS ........................................................ 83 APPENDIX C- OUTREACH ENCAMPMENT SURVEY............................................................................................... 84 APPENDIX D- OUTREACH ENCAMPMENT SURVEY RESULTS ................................................................................ 85 APPENDIX E- SAMPLE WEEKLY OUTREACH SCHEDULE ........................................................................................ 89 APPENDIX F- INTERVIEW GUIDE FOR PEOPLE STAYING IN ENCAMPMENTS & SERVED THROUGH THE ERP ......... 90 APPENDIX G- INTERVIEW GUIDE FOR COMMUNITY MEMBERS .......................................................................... 92 APPENDIX H- INTERVIEW GUIDE FOR CITY REPRESENTATIVES AND SERVICE PROVIDERS.................................... 94 APPENDIX I- GLOSSARY ...................................................................................................................................... 96 4 Executive Summary This evaluation is an independent examination of the City of Philadelphia’s Encampment Resolution Pilot (ERP), an initiative to shut down two homeless encampments located in the Kensington section of Philadelphia. The closure process involved an extended period of active outreach that facilitated access to housing, substance use treatment, and other services to people who stayed in the encampments and faced displacement. We examined two primary outcomes: first, whether the encampments were closed as scheduled and remained closed; and, second, the extent to which the people who were sleeping in the encampments (i.e., the target population) received needed services. The evaluation is based upon five primary data collection activities that, combined, provided a timely, multifaceted assessment of ERP, and particularly of the impacts ERP had on both those displaced by the initiative and the area of Kensington surrounding these encampments. These data components are: 1. An “Outreach Encampment and Survey” of 169 persons who frequented the encampments; 2. Data collected as part of ERP outreach services and linked to the City of Philadelphia’s integrated data system of services records known as CARES; 3. Direct observation of community meetings, ERP planning meetings, and encampment sites; 4. Interviews with people who stayed in encampments and community members as well as advocates for those staying in the encampments, and ERP officials and providers; and 5. Documents and records from City of Philadelphia and other sources, and media coverage related to ERP and Kensington. The ERP involved three phases: planning, implementation, and sustainment. The planning phase spanned early 2018, corresponding with the growth of the encampments and area concerns. During this initial phase, the encampment resolution group, consisting of leadership from multiple City departments, completed a plan for closing the Kensington Avenue and Tulip Street encampments. Two encampments were targeted so that sufficient housing would be available for the estimated 90 persons regularly sleeping in these locations. The plan highlighted the exceptional nature of the situation in Kensington, providing an opportunity to innovate and promote novel approaches to aiding those staying in encampments and improving the quality of life in the area around the encampments. It also underscored the heroic dimension to the initiative, in which modest resources were pitted against the daunting problems of homelessness and opioid abuse that were particularly intertwined and concentrated in the Kensington area. ERP started on April 30 with a 30-day implementation period where on-demand, low barrier housing and substance use treatment were offered to the target population in advance of the encampment closures. We assessed the implementation phase using a six-pronged approach: (1) findings from the Outreach Encampment Survey, which demonstrated the characteristics and needs of those staying in the encampments; (2) cataloging available services, and the extent to which they were appropriate, in sufficient supply, and accessible to the target population; (3) monitoring service engagement and its impact; (4) collecting perspectives from people staying in the encampments; (5) collecting perspectives from individuals advocating on behalf of those staying in the encampments and community members; and (6) documenting the encampment clearance process. When the implementation phase ended on May 30 with the closure of the two encampments, 83 individuals had engaged in temporary housing or substance use treatment services. After the encampments were cleared, the sustainment phase commenced and stretched well into Fall 2018. Our monitoring of this period focused on the longer-term outcomes of those who had engaged in services through ERP and the impact that the encampment clearances had on the surrounding area. 5 There were three facets to this monitoring. First, we took an in-depth look at the use of data, both ad hoc records used by outreach workers to structure sustained engagement, and administrative records covering a range of City-based services provided both before and after the ERP initiative. This allowed us to assess changes in services use as well as housing and treatment outcomes four and a half months after the encampments closed. In addition, we examined perspectives and reflections on the ERP from those who engaged with ERP-related services and from those involved in providing services, offering on- the-ground views of the sustainment process as well as early assessments of what worked and how future initiatives could be improved. Finally, we explored how the encampment closures impacted the surrounding community and, specifically, what changed in the wake of the closures, both near the former encampment locations and, more generally, in terms of impacts on homelessness and substance use in the area. This report offers the most comprehensive examination of an encampment clearance process, including the outcomes of those who were displaced by the clearances and the impact on the community, of which we are aware. Integrating an array of quantitative and qualitative data, we distill the key findings from this evaluation into 22 “lessons learned.” These are presented by the phase by which they are most readily associated: the planning phase (3 lessons), the implementation phase (9 lessons), and the sustainment phase (10 lessons). 1- Lessons Learned in the Planning Phase 1.1- Don’t reinvent the wheel. The City of Philadelphia’s approach adapted a model used in San Francisco that combined services with encampment closure, and now provides its own model that could potentially be used to guide similar efforts in other jurisdictions. 1.2- Effective coordination between participating entities is essential to overall success. Based upon our access to many planning meetings, officials and staff, and field settings related to implementing the ERP, we observed a high and, in our experience, unusual degree of coordination and cooperation among an array of services. 1.3- Keep expectations in perspective. ERP organizers emphasized the heroic nature of pitting the limited resources of a pilot project against the twin public health crises of homelessness and opioids. This provided a justification for setting aside bureaucratic caution and offered an opportunity to implement novel approaches. 2- Lessons Learned in the Implementation Phase 2.1- The encampments are an opioid-related problem. There was near ubiquitous substance use among those staying in the encampments: among all respondents on the Outreach Encampment Survey, 94 percent reported current substance use and 73 percent (79 percent of those reporting substance use) reported opioids as their drug of choice. As such, the ready availability of substance use treatment services was a critical component of the ERP. 2.2- The encampments are a homelessness-related problem. Over half (57 percent) of those staying in the encampments reported having spent time in a homeless shelter, and forty percent reported being homeless for over a year on the Outreach Encampment Survey. This was a population with substantial housing needs, and not simply a population who was homeless “by choice” in order to facilitate their substance use. 6 2.3- Homelessness in Kensington is a Philadelphia problem. On the Outreach Encampment Survey, 84 percent of survey respondents were Philadelphia residents, with 65 percent stating they were from Philadelphia and 19 percent coming from elsewhere but had lived in Philadelphia for over a year. The need for services to treat unsheltered homelessness in Kensington has Philadelphia origins and requires Philadelphia-based solutions. 2.4- Temporary housing availability is limited by local resistance. While low-demand local housing was desirable for the target population, community interests resisted local siting of temporary housing and other services. The resulting temporary housing supply appeared to have been an inadvertent compromise: more than many residents wanted, but less than what was needed. More proximal services would likely have led to higher levels of engagement by those staying in the encampments. 2.5- Closing encampments means balancing competing interests. The standoff over housing is one example of contrasting views that divide community members and advocates. Both stakeholder groups criticized aspects of the closure process, though from different perspectives; the City tried to maintain a middle position and be responsive to concerns from both sides. 2.6- Individual placements did little to relieve population pressures at the encampments. As substantial numbers of individual placements to housing and substance use treatment services occurred, new persons seemed to take their places as the encampments maintained a rough population equilibrium. 2.7- Involving people experiencing homelessness in resolving encampments is difficult. The target population was difficult to engage in a participatory process on resolving the encampments. Despite initial failure in this objective, the goal remains important. This challenges future closure efforts to develop innovative ways to involve the target population in the closure process. 2.8- Effective services require removing access barriers. Providing amenable, effective, and accessible housing and substance use treatment services under ERP led to more widely adopted best practices and was instrumental in providing temporary housing and/or treatment services to a total of 126 persons at some point during the implementation phase. 2.9- Closing the encampments is the easy part. Consolidating the closure into a reduced presence of unsheltered homelessness and providing housing and recovery options to those displaced by the closures showcased the formidable problems associated with addressing homelessness and substance use in the Kensington area. 3- Lessons Learned in the Sustainment Phase 3.1- Most of the encampment population is involved with municipal services systems. Among those most targeted for services through ERP, 55 percent were actively enrolled in Medicaid, and 90 percent were matched with some type of record in the City’s homeless, behavioral health, and/or prison systems. This could serve as a potential basis for creating more coordinated and effective services. 3.2- Use of a by-name list (BNL) is essential to coordinating individuals’ services. Creating and maintaining a BNL was the centerpiece of engaging with and managing services for 189 persons targeted for services. Six months after the encampments were cleared, the BNL was 7 instrumental in enabling outreach staff and caseworkers to maintain ongoing contact with 62 percent of the target population and engaging 41 percent with housing or substance use treatment services. 3.3- Having data makes a difference. Multiple data sources contributed to providing a rich report and evaluation of processes and outcomes related to the ERP. Future projects can learn from and build upon what was accomplished in this pilot. 3.4- Be deliberate in identifying outcomes and setting benchmarks. There was limited opportunity to design the evaluation beyond existing programmatic structures, and the evaluators largely adapted the evaluation to existing data. Benchmarks and successful program outcomes were at times unclear. 3.5- There is no model for policing an area overwhelmed by homelessness and opioid use. Many aspects of law enforcement’s role in the ERP were not within the bounds of traditional policing, and more support is needed to further formulate, implement, and communicate a clearer and more proactive role for the police. 3.6- Availability of short-term resources contrasts with scarcity of long-term resources. In October 2018, half of the 72 persons on the BNL who received either housing or substance use services were in long-term or permanent placements. This proportion would be higher, and thus the ERP outcomes would have improved, had more permanent housing and recovery housing resources been available citywide. 3.7- Consolidating gains made by a pilot program requires routinizing pilot services. Many ERP services required either diversions of existing resources or additional resource allocations. Converting these levels of services into ongoing services is necessary for continued access to housing and substance use treatment for the unsheltered homeless population targeted in this pilot. 3.8- Summer 2018 was long and difficult in Kensington. In the summer following the encampment closures, the number of persons counted as unsheltered and homeless increased to an unprecedented 700. This increase was unrelated to the encampment closures, but efforts to consolidate the gains from the closures were strained from this influx. This underscores the limited overall impact of a targeted initiative such as ERP. 3.9- Crisis creates opportunity. Among the lasting and most widely adopted innovations of the ERP was changing intake procedures to facilitate and expedite access to substance use treatment, and changing how temporary housing was provided to attract people who would otherwise have remained outdoors. Implementing such changes are more feasible in crisis conditions, and ERP was able to capitalize on local circumstances. 3.10- Pilots should lead to larger initiatives. We closed our examination of the ERP at the point at which the City implemented the Philadelphia Resilience Project, which would not have been possible without the groundwork in logistics, inter- departmental and agency coordination, and services provision that was developed through the ERP. 8 Chapter 1- Introduction This process evaluation is an independent examination of the City of Philadelphia’s Encampment Resolution Pilot (ERP), an initiative with the goal of shutting down two outdoor homeless encampments after actively reaching out to and providing assistance with housing, substance abuse and other services to people sleeping in the encampments. Process evaluations, in general, examine the functioning of an intervention and determine the extent to which the program’s implementation followed its design and led to attaining desired outcomes. In this process evaluation of ERP, there were two primary outcomes: first, whether the encampments were closed as scheduled and remained closed; and, second, the extent to which the people who were sleeping in the camps (i.e., the target population) received needed services. Additionally, an ethnographic component provided context regarding how the initiative was perceived by various stakeholders, including those staying in the encampments and community members and advocates. (See Appendix A for Methods.) The process of resolving encampments and relocating the target population was complex and multifaceted. Challenges included balancing the demands of the surrounding community with addressing the housing and substance use treatment needs of those in the target population. Conditions at the encampments raised substantial public health concerns, but closing them raised concerns related to due process, relocation, access to services and other health issues. Creating the atmosphere of trust necessary for an effective services-based and person-centered approach meant prioritizing a human services-based outreach over a more enforcement-based approach. All of this necessitated coordination between various entities, within and outside of City government, in the homeless services, behavioral health services, law enforcement, legal, sanitation, and communications domains. Finally, the ERP, as a pilot initiative, had limited resources with which to confront an extreme situation brought on by the larger crises of homelessness and opioid use that have been experienced nationwide. The ERP involved three phases: (1) planning, (2) implementation, and (3) sustainment. The planning phase spanned early 2018, followed by execution of the ERP. This started with a 30-day implementation phase that commenced on April 30 with posted announcements stating the City’s intention to close the two encampments. The objective of this phase was to make available on-demand, low barrier housing and substance use treatment services to the target population in advance of the closing of the encampments on May 30. The objectives of the sustainment phase were twofold: to assist those placed in temporary housing and substance use treatment as they continued to receive needed services; and to ensure against the formation of new encampment sites and the repopulation of those that were cleared. This report reviews each of the three phases of the ERP process, drawing on an array of data sources including: survey data from persons who frequented the encampments; integrated services use and outcomes data collected by the City; ethnographic observation; interviews with key stakeholders and persons directly involved with implementing the ERP; qualitative interviews at the encampments and in the surrounding community; and documents and records from City of Philadelphia and other sources; and media coverage related to ERP and Kensington. Taken together, this evaluation documents the ERP process; assesses planning, implementation, and outcomes; and reviews its strengths and limitations. Appendix I contains a glossary of terms and can be referenced for details on acronyms used throughout this report. 9 Chapter 2- Background 2.1- Geography The ERP targeted two of four encampments located in the Kensington section of Philadelphia. The encampments were on a series of streets that connected Somerset Street and Lehigh Avenue as they passed under a set of active Conrail railroad tracks. The camps were located in these underpasses on a corridor that was almost a half-mile long. The names of these encampments came from the names of these four streets: the Kensington Avenue and Tulip Street encampments were slated by the ERP for clearance; the other two camps, Frankford Avenue and Emerald Street (also called Emerald City) remained occupied. Figure 2a. Southern end of the Kensington Avenue underpass encampment, with Lehigh Avenue and Visitation Catholic School playground in the background. (source: Wall Street Journal)1 The encampments took advantage of the shelter provided by the underpass. Tents, mattresses and makeshift structures occupied the entire sidewalk on one side of each of the roadways. The sidewalk on the opposite side of each of these blocks, as well as the roadway itself, was kept clear of people camping and allowed pedestrians and vehicles to pass by unimpeded. Kensington Avenue, a major commercial thoroughfare, had steady streams of vehicular and pedestrian traffic going by the encampment. Down the block from the southern end of this underpass was the major intersection of Lehigh and Kensington Avenues, and on the other side of this intersection was a K-8 Catholic school. This epitomized the 1 Jon Kamp, “Wracked by Opioid Crisis, Philadelphia Braces for Tent-Camp Closures.” Wall Street Journal, May 26, 2018 (https://www.wsj.com/articles/wracked-by-opioid-crisis-philadelphia-braces-for-tent-camp-closures- 1527332400) 10 dissonant coexistence of the camps with the surrounding neighborhood. Figures 2a and 2b are photos of the Kensington encampment. In contrast, the Tulip Street camp was located on a less traveled street with Lehigh Avenue and Somerset Streets providing some buffer between the camp and adjoining residential blocks. Photos of this encampment are provided later in this report (Figures 4b through 4d), in conjunction with our account of the closure process. Police censuses indicated that, in April 2018, as many as 50 people per night slept in each camp (Kensington and Tulip), with a combined total for both bridges reaching as many as 90 people on a given night. The encampments were located in the Kensington section of North Philadelphia. Kensington encompasses several neighborhoods, and the railway under which the encampments lie forms a boundary that separates the Heart of Kensington, East Kensington, Somerset, Olde Richmond and Port Richmond neighborhoods. Because of this, a number of neighborhood-based organizations took an interest in the disposition of the encampments.2 In addition, the encampments were along the boundaries of the 1st and 7th City Council districts and of the 24th and 26th police districts. Appendix B contains a map with demarcated encampment locations. 2.2- Historical Context and Rise of Encampments The socioeconomic fortunes of Kensington have paralleled the rise and fall of industry and manufacturing in Philadelphia.3 This was an area where factories were located among residential blocks in a self-styled “workshop of the world.” As Philadelphia deindustrialized after World War II, its working class, predominantly Irish-American population declined and Hispanic and African American populations expanded. In contrast to other North Philadelphia neighborhoods, the Kensington area maintained a mix of races and ethnicities. As blue-collar jobs departed during the postindustrial era, unemployment, poverty and related socioeconomic woes increased and the industrial infrastructure that once provided this area with its identity and jobs fell into ruin. These conditions of postindustrial decay, particularly given a topography that includes railroad infrastructure, has provided the setting for Kensington’s rise as Philadelphia’s main hub for heroin and related activity. Abandoned railway infrastructure, vacant land and empty factories have created, in the words of one profile of the area, “a complete ideal place to be an open-air drug market.”4 Kensington Avenue has two intersections, at Allegheny Avenue and at Somerset Street, that rank among Philadelphia’s two most well-known drug corners and the street between these intersections is a known 2 Organizations that have participated in the ERP process include the East Kensington Neighborhood Association (EKNA), Harrowgate Civic Association (HCA), the Hispanic Association of Contractors and Enterprises (HACE) Community Development Corporation, the Kensington and Allegheny Business Association (KABA), New Kensington Community Development Corporation (NKCDC), Olde Richmond Civic Association (ORCA), Port Richmond Community Group (PRCG), Port Richmond on Patrol and Civic (PROPAC) Association, Somerset Neighbors for Better Living (SNBL), and West Kensington Neighbors (WKN). 3 For a more detailed account of economic, social and demographic indicators of the area around the encampments, see the report facilitated by Interface Studio LLC & V. Lamar Wilson Associates, Inc. (2016) Heart of Kensington: Collective Impact, 2022 (http://www.impactservices.org/neighborhood-plan/). 4 Alfred Lubrano (1/24/2018), “How Kensington got to be the center of Philly's opioid crisis.” Philly.com (www.philly.com/philly/news/kensington-opioid-crisis-history-philly-heroin-20180123.html). 11 prostitution strip.5 As opioid use has reached the level of public health crisis, Kensington has become the city’s hub of illicit opioid activity and a reputed source for particularly high-quality heroin. Heroin in Kensington received widespread attention in Summer 2017 with the clearance of a half-mile, semi-secluded stretch of railroad track that had functioned for several years as an open-air drug market and shooting gallery. Known alternately as El Campamento or Gurney Street (after a street that runs along that stretch of tracks), the area was located just west of Kensington Avenue (see Appendix B) and preceded three of the four underpass encampments (with the Emerald Street camp already established by Spring 2017). The scale and openness of the drug activity in this area, set against a large accumulation of needle debris and general squalor, received extensive media coverage.6 In July of that year, the City pressured Conrail, the owners of the railway property, to secure and clean up the area, and mobilized outreach workers to encourage people frequenting the Gurney Street area to accept temporary housing and substance use treatment. While consensus holds that the Gurney Street area was primarily used for drug transactions and use, it also contained a homeless encampment. A 2016 study, Kensington Counts, reported a “tent city” there of 60-75 individuals.7 The emergence of three of these encampments in Fall 2017, along with their close proximity to the Gurney Street area (see Appendix B), led to the widespread belief that the underpass encampments were a regrouping of displaced people and drug activity in a process akin to “a grim game of whack-a-mole.”8 City of Philadelphia officials cite surveys that showed a large majority of those staying in the underpass encampments to have reported not staying at the Gurney Street encampment.9 Regardless of perspective, the clearance of the Gurney Street area was frequently cited as a reference point in the context of the underpass encampments, and provided the City with a prototype from which to help plan their outreach and housing efforts. By the winter of 2017-18, police in the 24th District were regularly counting 200 unsheltered homeless in a series of nightly censuses, the majority of whom slept in the underpass encampments. This group was resistant to staying in shelters, as the demands of maintaining regular opioid use was difficult in a regulated shelter environment, and shelter locations were typically far from Kensington and a 5 Many media accounts highlight this notoriety of Kensington Avenue and these two intersections. A comprehensive account of the origins of the association between drugs, prostitution, and Kensington Avenue is in Jeff Deeney (8/13/2011), “Philadelphia's Kensington Avenue: Heroin, Prostitution, and No Police.” The Daily Beast (https://www.thedailybeast.com/philadelphias-kensington-avenue-heroin-prostitution-and-no-police?ref=author). 6 For examples of this, see Mehmet Oz (4/18/2017), “Inside Ground Zero of the Heroin Epidemic in Philadelphia” (https://www.doctoroz.com/episode/oz-inside-festering-epicenter-heroin-will-shock-you); Jeffrey Stockbridge & Courtenay Harris Bond (7/25/17), “Inside A Notorious Philadelphia Drug Market Before It Gets Shut Down” Time Magazine (http://time.com/4868823/philadelphia-campamento-cleanup/); and Alan Taylor (8/8/17) “Closing Down a Notorious Heroin Camp in Philadelphia” The Atlantic (https://www.theatlantic.com/photo/2017/08/closing-down-a-notorious-heroin-camp-in-philadelphia/536232/). 7 Kensington Counts Final Report (2016) (http://www.philadelphiaofficeofhomelessservices.org/news/reports/). 8 Edward Helmore (6/1/2018), “How Philadelphia closed homeless 'heroin camps' amid US opioid crisis.” The Guardian (https://www.theguardian.com/us-news/2018/jun/01/philadelphia-homeless-heroin-bridge-camps). 9 For a sample of opposing perspectives on the Gurney Street area clearance, see Thom Nickels (8/31/2017), “Neighborhoods Are Bearing the Cost of the Conrail Heroin Camp Cleanup.” Philadelphia Magazine (https://www.phillymag.com/news/2017/08/31/conrail-heroin-camp-port-richmond); and the response by Liz Hersh & David T. Jones (9/13/2017), “Response: The City was well-prepared for the Conrail Heroin Camp Cleanup.” Philadelphia Magazine (https://www.phillymag.com/news/2017/09/13/city-role-conrail-heroin-camp-cleanup/). 12 dependable heroin supply. This recalcitrance, despite the hazards of staying outdoors in December’s extreme weather conditions, prompted the City to open a “warming center” at the Cione Recreation Center, located near the encampments at 2600 Aramingo Avenue, that provided temporary overnight accommodations to roughly 75 people a night. Nonetheless, even on the coldest nights, roughly 100 people continued to sleep in the encampments.10 Figure 2b. Kensington Avenue underpass, with encampment on the eastern (northbound) sidewalk (source: photo posted on social media by Vanessa Baker11) As Spring approached and the encampments became increasingly entrenched, political and community pressure grew for the City of Philadelphia to address the issue. Two constituencies took a particular interest in the issue. On one hand, community members pointed to the encampments as the most visible example of how opioid-related activity had decimated the quality of community life in the neighborhoods abutting the encampments. On the other hand, advocates for a harm reduction approach to opioid use, along with groups that provided material assistance and support for persons in the encampments, opposed closing the encampments in the absence of access to sufficient housing, substance use treatment, and safe injection resources. They maintained that closing encampments without sufficient alternatives would increase the public health risk that led to 1,217 overdose deaths in Philadelphia in 2017. Amidst the high visibility of the encampments, and the competing concerns of these stakeholders, a City-led group of municipal entities and community non-profit organizations started to formulate a plan to close the underpass encampments and address the needs of those sleeping there. 10 Press release (1/2/18), “City Provides Extended Warming Center Hours and Other Supports for Homeless Living in Kensington-Fairhill.” City of Philadelphia (https://beta.phila.gov/2018-01-02-city-extends-warming-center-hours- for-homeless-in-kensington-fairhill/); Aubrey Whelan (1/4/18), “In Kensington, battling the cold means battling addiction.” Philly.com (http://www.philly.com/philly/health/addiction/in-kensington-hundreds-homeless- addiction-and-only-40-shelter-beds-20180302.html). 11 See https://www.facebook.com/events/959839337513714. 13  Chapter 3- Phase 1: Planning and Developing the Encampment Resolution Pilot In January 2018, planning began for what would become the Encampment Resolution Pilot (ERP). This chapter documents key elements of this planning process up to its culmination with the 30-day implementation period that is described in the next chapter. Leadership of this group came from: - the City of Philadelphia’s Managing Director’s Office (MDO)12 and their divisions of Health & Human Services (Eva Gladstein, Deputy Managing Director) and Community Services (Joanna Otero-Cruz, Deputy Managing Director); - the Office of Homeless Services (OHS, Liz Hersh, Director); - the Philadelphia Police Department (PPD, Ray Convery, Inspector, East Division); - the Department of Behavioral Health and Intellectual disAbility Services (DBHIDS, Jill Bowen); - Community Behavioral Health (CBH, Geoffrey Neimark); and - the City of Philadelphia Law Department (Kristin Bray). Many of the departments in this resolution group were involved with clearing the Gurney Street area in 2017. This clearance process was different in key respects to the clearance of Gurney Street in that it was on public property and insofar as the functions of the four underpass sites were primarily as encampments of people experiencing homelessness. This led the group to examine models for specifically addressing homeless encampments, and in developing an approach in which making housing and substance abuse services available became a key element of preparations for physically clearing the encampments. 3.1- Using San Francisco as a Model In developing a plan, the group looked to an approach developed and used by the City of San Francisco. This approach connects persons sleeping in camps to services and housing as part of physically removing the encampment. This adds a social services component to a process that has traditionally been primarily a law enforcement operation. The structure of this approach is also mindful of potential legal challenges to removing encampments on the basis that there are insufficient alternative accommodations available to those displaced by the closures. To avoid such legal challenges, one component of the closure process was for the City to have sufficient housing and services available to accommodate those in the targeted encampments. With this approach, known as encampment resolution, San Francisco closed 17 encampments in its first year of operation (August 2016 through July 2017). This approach starts by setting a deadline for removing the encampments, and subsequently focusing outreach efforts on people staying in the encampments. The imminent camp closure creates some urgency for accepting housing, treatment and other services as an alternative to displacement. In their plan, Philadelphia adapted many of the same measures and nomenclature from the San Francisco approach. Key features of San Francisco’s process include:13 12 See MDO’s website: http://www.phila.gov/mdo/pages/default.aspx 13 The information presented here about San Francisco’s approach uses an outline and descriptions that closely follow (and in some cases paraphrase) City of San Francisco webpages. See: http://hsh.sfgov.org/street- homelessness/encampment-resolution-team and https://www.sfpublicworks.org/navigationcenter. 14 1. Input from all stakeholders including people staying in encampments and community members impacted by encampments in planning and implementing an encampment resolution strategy. 2. Collaboration across a variety of City government agencies that have jurisdiction over various aspects of the camp resolution. 3. Intensive and persistent outreach and engagement by trained, experienced outreach workers. 4. Establishment of Navigation Centers, which offer housing and related services specifically for those staying in the encampments targeted for closure. Navigation Centers provide lodging and those served are encouraged to work with case managers who can connect them to income, public benefits, health services, shelter, and housing. Navigation Centers are different from traditional shelters in that they have few barriers to entry and ready access to intensive case management. 5. Evaluation and documentation of the encampment resolution process and outcomes in order to assess the success of the resolution effort and to apply findings to future resolution initiatives. 6. Prevention of encampments from re-forming in previously addressed areas through a variety of means, including police monitoring, outreach efforts, community involvement and physical changes to the site. As was the case with San Francisco, Philadelphia’s emerging resolution plan prioritized both the physical clearance of the encampments and making available housing and other services for those facing displacement from the encampments. With this design, there was a partnership between human services and law enforcement in which the lead role consisted of intensive efforts to connect persons with needed services. The specifics of these services will be described later in this report. A secondary function of this prominent service component was to reduce the risk of legal challenges to the resolution efforts. 3.2- Legal Considerations14 Philadelphia Code § 10-611 clearly precludes establishing encampments on public sidewalks insofar as it specifically prohibits sitting, standing, lying or otherwise using the public sidewalk, or placing one's belongings or other objects upon the public sidewalk, so as to impede, block, or obstruct the free passage of pedestrians. Furthermore, the underpass encampments, insofar as they generated considerable needle debris and had inadequate facilities to accommodate human waste, could also be considered a public nuisance. Either ordinances, restricting the uses of the sidewalk, or the City’s general regulatory authority to abate curbing public nuisances, would provide legal grounding for the City to clear the camps. The City also needed to consider potential Constitutional claims under the US and Pennsylvania Constitutions. While we are unaware of any litigation regarding removal of encampments in Pennsylvania, other municipalities had been sued when they engaged in encampment removal strategies; those legal challenges were primarily based upon infringements to the Constitutional guarantee of due process and the Eighth Amendment prohibition against cruel and unusual punishment. In general, other jurisdictions were found to have violated the Constitutional guarantee of due process when orders were given to abruptly vacate a camp area, without notice, and/or the municipality 14 Information on legal considerations is largely based upon an interview of and materials provided by Kristin Bray from the City of Philadelphia’s Law Department. Stated opinions those of the report authors. 15 disposed of personal possessions without notice. Challenges under the Eighth Amendment were brought when municipalities engaged in punitive measures such as arrests or criminal citations that were executed in conjunction with clearing encampments. Courts generally reasoned that such actions were cruel and unusual insofar as they punished people for their status as a homeless individual, who were involuntarily living on the street and who had no other viable housing options. Figure 3a. Encampment closure notice posted in Kensington Avenue underpass (source: evaluation team) The structure of ERP, insofar as it gave a 30-day notice (Figure 3a) in addition to other documents notifying encampment residents of the City’s plans and options available to them in both English and Spanish, provided alternative housing and services, and outlined clear procedures for safeguarding and storing possessions, appeared to exceed measures that were used in other jurisdictions, including San Francisco, to facilitate transitions for those persons who would be displaced due to encampment clearance. While plans for developing the ERP were clearly mindful of past legal challenges in other jurisdictions, there is nothing that we found to suggest that the ERP was developed for the primary purpose of precluding legal liability; rather the primary purpose was to provide support and services and encourage individuals to seek treatment. 16 3.3- Framework for the ERP In March 2018, Philadelphia’s encampment resolution group completed a plan for closing two of the four Kensington underpass encampments, at Kensington Avenue and Tulip Street.15 Only two encampments were targeted so that sufficient housing would be available for those sleeping in the encampments, which were estimated to contain up to 90 persons who regularly slept in these locations. This plan was considered a pilot project, and an assessment of this resolution pilot would guide subsequent actions targeting the remaining two underpass encampments at Frankford Avenue and Emerald Street. The plan for the ERP that emerged from the planning process outlined an organizational structure and activities for the remainder of the planning period, for a 30-day implementation period starting with posting notices announcing the City’s intention to close the encampments and ending with the removal of all materials under in the encampments. The most visible activities during this implementation period included intensive outreach efforts, increased police monitoring activity, and twice per week cleaning of the encampment areas by the Streets Department. The implementation period would be followed by a sustainment period during which those who accepted services would have access to more long-term housing and case management services. Police and outreach workers would monitor the closed-down underpass areas to ensure against re-encampment. In the plan, the resolution group was divided into three teams. An Encampment Oversight Team (EOT) would provide largely logistical support with primary duties that included creating the Encampment Engagement Team (EET), securing needed resources, and removing barriers to housing and substance use treatment services to the extent possible. Once the EET was established, it would handle matters related to engaging those identified on a by-name list (BNL) as sleeping in the encampments, with the aim of facilitating their exit from the camp and acceptance of services. The EET would also coordinate outreach staff with police patrols to provide alternatives to enforcement actions. These two teams (EOT and EET) would be coordinated by a third team, the three-person Encampment Coordinating Team (ECT) representing DBHIDS (Bridgette Tobler and Tim Sheahan) and OHS (David Holloman). ECT would provide operations leadership. Table 3a contains a comprehensive list of the entities that were part of the resolution group, and indicates which ones participated on each of the three teams. The three teams were presented with a substantial number of operations challenges to address before a functioning engagement process was feasible. Primary among these challenges included: - Locating and establishing a Navigation Center with approximately 40 beds to supplement the existing, low-barrier respite bed supply at PPP and ODAAT shelters that was available for short- term housing. - Ensuring the availability of sufficient assessment services, appropriate treatment services, and other services, and arranging to expedite intake and assessment so that, to the extent possible, these services were available on demand. This included providing assistance with (or waiving requirements for) identification, suspending preauthorization for services, and providing services while any issues with insurance coverage were worked out. 15 The primary sources for the City of Philadelphia’s encampment resolution plan come from internal planning documents and interviews with David Holloman (Director of External Affairs, OHS) and Beverly Woods (Assistant Deputy Managing Director, Health and Human Services Division of MDO). 17 - Appropriating and coordinating intensive outreach services with outreach workers from DBHIDS, ODAAT, Project HOME, and PPP. - Ongoing case management and services coordination for those accepting services during the implementation phase, which included having sufficient long-term housing. - Collecting survey data and creating a BNL that included the names of all persons who were sleeping at the Kensington and Tulip encampments and who got priority for housing. - Coordinating activities of the many and varied entities that were involved in the encampment resolution process. - Formulating the logistics of shutting down the two encampments. - Addressing unanticipated problems. Table 3a. Entities Participating in the ERP Entity Role in Encampment Resolution Process Team Participation Managing Director’s Office (MDO) Direct the resolution group, media relations and other communications functions Office of Homeless Services (OHS) Direct housing resources and provide operations leadership ECT, EOT Department of Behavioral Health and Intellectual Disability Services (DBHIDS) Direct substance use treatment services, provide operations leadership, and coordinates outreach services ECT, EOT, EET Philadelphia Police Department (PPD) Provide security, coordinate with outreach, and lead role in shutting down the encampments and preventing (with outreach services) resettlement EOT, EET Prevention Point Philadelphia (PPP) Provide respite and navigation housing and outreach services (non-City, non-profit agency) EOT, EET Project HOME Provide outreach services (non-City, non-profit agency) EOT, EET One Day at a Time (ODAAT) Provide respite housing and outreach services (non- City, non-profit agency) EET Community Behavioral Health (CBH) Provide substance use treatment and mental health services for Philadelphia residents with Medicaid eligibility EOT Law Department Provide guidance in legal matters Streets Department Carry out regular cleanups of encampment areas Licenses and Inspections Assist with code compliance and logistics of property management in conjunction with encampment closures Note. ECT=Encampment Coordinating Team; EET=Encampment Engagement Team; EOT=Encampment Oversight Team A draft plan that was put together on March 18, 2018, called this the Kensington Encampment Resolution Pilot Strategy, a name that would be shortened to the Encampment Resolution Pilot (ERP). Calling it a pilot highlighted several aspects of this initiative that were emphasized in presenting it to the public. The exceptional nature of the situation provided an opportunity to innovate, an opportunity to, in the words of DBHIDS’s Tim Sheahan “push the system farther” in being more responsive to the needs of persons who are both unsheltered and opioid dependent. In calling this a pilot, officials also underscored the novelty of the approach they were taking. OHS’s David Holloman observed: 18 Models are made to be shaken, broken, and put back together. This process is as much about finding out what works and doesn’t work as it is about providing assistance. Finally, framing this as a pilot underscored a heroic dimension to this initiative, in which modest resources were pitted against the daunting problems of homelessness and opioid abuse that were particularly intertwined and concentrated in the Kensington area. Managing Director Mike DiBerardinis, commenting in the wake of the encampment closures, underscored this theme: We have never seen a crisis like this before in Philadelphia and doing nothing is not an option.16 16 Sheahan’s and Holloman’s quotes were from evaluation team member notes of a “Town Hall” meeting at the Visitation Church Community Center in Kensington, May 4, 2018. DiBerardinis quote was from notes of a press conference at MacPherson Square in Kensington on May 30, 2018. 19 Chapter 4- Phase 2: Implementing the Encampment Resolution Plan On April 30 notices went up in the Kensington and Tulip underpasses that gave people thirty days, by May 30, to “leave this location and remove your personal property” (Figure 3a). This ushered in the implementation phase of the ERP, with the objective of getting as many people as possible out of the encampments and into housing or substance use treatment before the encampments were closed on May 30. The implementation phase was a complex undertaking and, in chronicling these thirty days, we seek in this chapter to give it a structure that captures this complexity and renders it amenable to critical assessment based upon the evaluation questions presented in the introductory chapter. In order to do this, we chronicle the implementation phase in six dimensions. Each of these dimensions overlaps with some of the other dimensions, but provides a singular perspective and draws upon different data sources. The first dimension focuses on the target population, and uses survey findings to report their characteristics and needs as a basis for matching to available services. The second dimension reviews the available services, and the extent to which they were appropriate, in sufficient supply, and accessible to the target population. Service engagement, the third dimension, examined the interface between the target population and the available services in terms of how many persons received services and how this impacted the encampment populations. The next two dimensions take qualitative approaches to examining key groups involved with the ERP. In the fourth dimension, findings based upon open-ended interviews of those staying in the encampments address questions related to how people come to stay in the encampments, the reasons why they accept or decline services, and other challenges to living at or leaving the encampments. The fifth dimension then broadens this focus of target population and available services to consider two key stakeholder groups: housed residents of the communities around the encampments and groups and individuals who advocate on the behalf of the target population. Both had specific interests that overlapped and at times conflicted. Finally, the entire implementation phase steadily built to the day of the encampment clearances, and we detail this process as the sixth dimension. The process of physically removing the encampments, and coordinating the relocation of those in the encampments, required coordination between multiple actors, and involved a considerable amount of uncertainty as to how things would go. The degree to which the encampments were cleared would, along with the number of people receiving services, be the key indicator of whether or not the implementation phase was successful. 4.1- Target Population In the week leading up to the closure announcement (April 23-27), outreach workers surveyed persons at the Kensington and Tulip Street encampments. The object of this survey, called the Outreach Encampment Survey, was both to identify a discrete target population, the basis of a BNL, and to collect information about this group. The BNL would include all those who were sleeping in the two encampments at the time of the closure notice posting. This survey, in providing a profile of the personal characteristics and service needs of unsheltered persons experiencing homelessness in Kensington, was only the second survey to date focusing on this population.17 The Outreach Encampment Survey was collected from a convenience sample in that no sampling methodology was used and many of those who responded were not sleeping in the two targeted encampments at the time of the survey (although nearly all respondents reported being homeless). 17 The 2016 Kensington Counts report contained the first such survey (see note #7). 20  While the results must be viewed with these limitations in mind, they provide a profile of this population that is useful for planning purposes. The survey document is included in this report as Appendix C. Respondents were provided with a $5 convenience store gift card for participating. Name and date of birth was collected for those willing to provide it. Key findings are presented here; detailed survey results are provided as Appendix D. Findings from the 169 respondents surveyed are summarized below by the following topics: demographics; current living conditions and place of origin; homeless service use and interest, employment history and service needs; and substance use, treatment and mental health. - Demographics. o Thesurveyedpopulationwasaboutthree-quarters(76percent)male. o Thepopulationwasofmixedracial/ethniccomposition,withtherespondentsbeing majority non-Hispanic White (57 percent), about one-third (31 percent) non-Hispanic Black, and 12 percent Hispanic. o Themedianageoftherespondentswasinthe35-44agerange,and88percentwere under age 55. o Thesedemographicswereverysimilartothosefoundinthe2016KensingtonCounts survey of unsheltered homeless in Kensington.7 o Comparedtothe2018city-widehomelessPoint-in-Time(PIT)countresultsfor unsheltered homeless, the proportion of male and Hispanic were similar. In contrast to this survey, where the majority of respondents was non-Hispanic White (57 percent), the majority of the unsheltered population in the PIT count was Black (54 percent).18 - Current Living Conditions and Place of Origin. o Almostallrespondents(94percent)reportedliving“onthestreet,”and92percent reported having spent the previous night “on the street” or in a bridge encampment. At the time the survey was administered, about 90 persons slept in the Kensington and Tulip encampments on a given night. o Themediantimespentlivingonthestreetwas6to9months.Twentypercentreported living on the streets for less than three months, while forty percent reported being homeless for over a year. Given that virtually all respondents likely had substance use issues (among other conditions) that may have been considered disabling, a large proportion of this 40 percent would likely have fit the federal definition of "chronically homeless."19 If so, such a proportion would have been consistent with the corresponding proportion of chronically homeless in Philadelphia’s unsheltered PIT count. o One-thirdofthosesurveyeddisclosedbeingwithapartner.Thishadimplicationsfor providing housing, as the respites and navigation centers often had to make special arrangements to accommodate couples. No data was collected to indicate families with children. o Almosttwo-thirdsoftherespondents(65percent)statedtheywerefromPhiladelphia, with a little over half of those from outside of Philadelphia reporting that they had lived 18 City of Philadelphia Office of Homeless Services, 2018 Philadelphia CoC Point in Time Count Summary with Subpopulations (http://www.philadelphiaofficeofhomelessservices.org/know-homelessness/point-in-time-count/). 19 For the federal definition of chronic homelessness, see https://www.hudexchange.info/resources/documents/DefiningChronicHomelessness.pdf 21 in Philadelphia for over a year. This is inconsistent with the belief that a large proportion of the encampment populations recently migrated to Kensington from other cities (and states) to engage in substance use. - Homeless Service Use and Interest, Employment History, and Service Needs. o Useandawarenessofshelterswasaboutevenlysplitamongrespondents.Alittleover half had gone to a shelter in Philadelphia (57 percent), and an equal number expressed willingness to go into shelter (56 percent). That number climbs to 91 percent when respondents were asked whether they would go to a shelter with minimal rules. o Mostindividualsreportedhavinganemploymenthistory(82percent),aboutthree- quarters expressed interest in employment assistance (76 percent), and over half reported looking for help with finding work (62 percent). o Whenaskediftheywerelookingforhelpinotherareasinadditiontohousingand employment, the most common responses were substance use treatment (56 percent), regaining identification documents (42 percent), applying for entitlements/benefits (33 percent), mental health treatment (30 percent), and medical treatment (21 percent). Among those who felt unable to access needed assistance, the most common barrier was lack of identification (65 percent), followed by substance use (33 percent), and trouble navigating the system independently (19 percent). - Substance Use, Treatment, Mental Health, and HIV. o Nearlyalloftherespondentsreportedcurrentdruguse(93percent)andone-quarter reported current alcohol use (25 percent). Opioids were by far the most popular drug of choice (79 percent), followed by cocaine (43 percent), and marijuana (18 percent). K2, benzodiazepines, and PCP were favored by a minority of respondents (9 percent, 8 percent, and 3 percent respectively). o Nearlythree-quartersofrespondentshadpreviouslybeentosubstanceusetreatment (73 percent), and an equal number were interested in getting treatment for their drug use (74 percent), with the majority indicating interest in medically assisted treatment (61 percent; MAT). Further, the vast majority endorsed interest in long-term treatment that could lead to permanent supportive housing (82 percent). However, the primary treatment barrier respondents affirmed was “not [being] interested or ready for treatment at this time” (52 percent), while the cost and lack of insurance was also noted as a barrier by about one-third of those surveyed (31 percent). o Overhalfofrespondentsreportedmentalhealthchallenges(65percent).Amongthose with such challenges, those reported most often were depression (68 percent), anxiety (50 percent), bipolar disorder (44 percent), post-traumatic stress disorder (19 percent), and schizophrenia (19 percent). If this self-report were accurate, conservatively upwards of one third of the overall surveyed population would be diagnosed with a major mental diagnosis (schizophrenia, bipolar disorder, and major depression). o Thosereceivinghelpformentalhealthchallenges(32percent)amountedtolessthan half the proportion who affirmed interest in such assistance (71 percent). o Whileonly3%reportedwantinghelpwiththeirHIVtreatment,thisrepresenteda potential opportunity for transmission and intervention. The low percentage may also reflect the fact that individuals diagnosed with HIV are often more easily able to receive case management, medical services and housing support. To the extent that the respondents in this survey representative of those sleeping in the underpass encampments, the portrait that emerges is one of a target population defined as much by its lack of 22 housing as by its substance use. The majority of survey respondents reported extended experiences with homelessness, including roughly 40 percent who likely could have been characterized as chronically homeless. A majority of the respondents had also received some sort of homeless service. In reviewing the survey results, we found patterns of homelessness that were roughly consistent with those of Philadelphia’s more general unsheltered homeless population. What is different among this encampment population, however, is the near ubiquity of substance use, and particularly opioid use. High levels of substance use are typically present among unsheltered homeless populations, but rarely at the levels found here. This supports assertions by OHS that the “homeless encampments are the result of the opioid epidemic.”20 In casting this population in terms of both homelessness and substance abuse, these survey results confirmed some basic assumptions about those sleeping in the underpass encampments. However, they also called into question other assumptions, such as the presence of large numbers of out-of-town migrants and more casual drug “tourists.” Caution is warranted here about the limits of drawing extensive conclusions based solely upon homelessness and substance use, as such a two-dimensional portrait will inevitably oversimplify and can readily lead to stigmatization of the target population. Also noted is the openness to housing and treatment services expressed by the survey respondents. The results support the central premise of the ERP: that the encampment closure process offered an opportunity to provide housing and treatment services to those who are facing displacement. 4.2- Services Coupling the availability of services with closing encampments is the defining feature of the approach taken by the ERP. Services provide a possible way out of homelessness, to the backdrop of the pressure of encampment clearance and subsequent displacement. The survey results support the importance of providing both housing and substance use treatment services.21 In an ideal scenario, all those staying in the encampments would receive housing and/or substance use treatment services, and there would be no need for further closure action. However, the goal of placing those in the encampments into services faced three primary sets of limitations: - Availability of the housing services was sufficient for targeting two of the four underpass encampments for closure, and then was prioritized to those persons on the BNL who were documented to be sleeping in the encampments at the time that the closure notification was posted (i.e., at the start of the 30-day implementation period). Given the “churn” among the encampment populations, this meant that as the closure date neared, those given priority for these services were a shrinking proportion of those at the encampments. - Access to the services was broadened by addressing logistical and bureaucratic obstacles. Efforts to reduce barriers to getting services increased the likelihood that targeted persons would follow through with accessing services. For housing, this meant implementing low-demand approaches that include better accommodating pets and possessions, not requiring identification, no sobriety requirements, and exit and entry at will. For treatment, providing 20 Internal City of Philadelphia Office of Homeless Services document. “Opioid Driven Homeless Camps: Update on Pilot, June 2018.” PowerPoint presentation. 21 The availability of both treatment and housing services intentionally leaves options for whether or not to engage in one of these services or both. Under the framework set up by ERP, treatment is not prerequisite to accessing housing services. 23 transportation, removing identification requirements, and reducing or removing pre- authorization requirements all facilitated a “treatment on demand” approach that ensured a smoother entry into treatment and increased the chances that the target population would seek services. - Suitability describes the perception that the services offered would facilitate desired outcomes. Services received criticism for being temporary, with no clear path to more long-term housing and recovery options. There was also resistance from those in the encampments who stated they were otherwise unwilling or not “ready” to accept services, even in the face of imminent camp closure. These limitations provide a context for a closer examination of the principal services offered in the ERP. This section provides further detail as to the nature of housing, substance use treatment and other services, with specific details pertaining to availability, access, and suitability of each. Better understanding of these services provides insights as to the challenges to implementing the ERP so that it could offer meaningful alternatives to unsheltered homelessness. 4.2.1- Respite and Navigation Housing The terms “respite” and “navigation” were both used to designate housing that was earmarked for those in the encampments. Respite housing comprised of the already existing shelter beds that were made available for the ERP, located at two shelters: One Day at a Time (ODAAT) at 2432 West Lehigh Avenue, and Prevention Point Philadelphia (PPP) at 2913 Kensington Avenue. Navigation housing contained beds in a facility that was set up specifically for the ERP, located at 3247 Kensington Avenue. Altogether, this housing provided capacity for approximately 90 persons from the Kensington and Tulip underpass encampments: 50 as respite beds and 40 in the navigation center. Persons who were homeless and actively using opioids typically eschew use of Philadelphia’s shelter system. Simply put, the need to obtain the opioids, keep the drugs and related works, and find suitable arrangements for injecting the drugs multiple times daily is incompatible with the structured environment typical in a homeless shelter.22 While the unstable and irregular housing of many who use opioids is consistent with definitions of homelessness, the drug-using population in the Kensington area represented a secondary nexus of homelessness, set apart from the primary nexus of people experiencing homelessness and homeless services in the Center City area and from the City’s shelter network. The respites and navigation center set up for ERP were designed to make housing more attractive to those in the Kensington and Tulip camps through a “low demand” structure and locations that were close to the bridge encampments. Low demand meant that there were both few requirements for admission and a reduced set of rules and expectations that governed stays. There were no requirements for sobriety, mental health treatment, or service participation, and there were no curfews or restrictions to leaving. Partners and possessions could accompany those in the respite and navigation centers, although other items such as pets and drugs were still prohibited. The PPP respite and navigation center were within walking distance of the underpass encampments, while the ODAAT beds were far enough 22 Perceptions of shelters as structured by such mechanisms as rules, curfews and limited privacy may conflict with efforts by OHS towards a more housing first approach that provides more low-barrier options for temporary housing. More detail on OHS shelter standards is available at http://philadelphiaofficeofhomelessservices.org/wp- content/uploads/2016/08/eh-standards-final.pdf. 24 away so that, despite transportation being made available to a drop-in program near the encampments, people in the encampments complained about the distance. The navigation center did not open until May 7, but its forty beds quickly filled and it remained at capacity. Similarly, the PPP respite beds also filled and remained filled. ODAAT respite beds were less desirable (primarily due to the greater distance from the underpass encampments) and never filled to capacity. This meant that outreach workers had housing available for people all through the implementation period, yet additional respite and navigation center beds in locations closer to the underpass encampments would likely have led to more persons accepting such housing accommodations. The respite and navigation center beds were referred to as emergency temporary housing. While no commitments were made by the City with respect to accessing more permanent housing accommodations, questions remained concerning paths to permanent housing for those in the temporary housing. City officials acknowledged that permanent housing of all sorts was scarce for persons throughout the shelter system, and that they did not have a sufficient source of permanent housing set aside for those in the respite and navigation centers. While one non-profit housing-first provider, Pathways to Housing, did have some permanent housing specifically for persons experiencing homelessness with substance use disorders, such a designation would fit many persons experiencing homelessness in the Kensington area and fell far short of the need for such housing. Beyond issues of availability, the prospect of housing earmarked specifically for people from the Kensington encampments would have raised policy dilemmas. For example, a dedicated housing resource would have conflicted with efforts by OHS to provide more equitable, city-wide access to housing resources through a centralized intake system. Targeting persons in the Kensington area with additional housing resources could have risked creating a “housing magnet” where people might migrate to Kensington in search of housing services. Those staying in the respite and navigation centers did have access to case management services that worked with these individuals and couples on locating housing over a time period that extended beyond the camp closures. This included accessing the scarce supply of low-income housing resources, accessing recovery-based housing (see next subsection), and leveraging individual and social resources, such as employment or family assistance, focused toward more permanent arrangements. 4.2.2- Substance Use Treatment Given the ubiquity of substance use among those staying in the underpass encampments, the availability of substance use treatment services was a critical component of Philadelphia’s ERP. Entering treatment was another means for someone to exit the encampments and have access to services that could address both recovery and short-term housing needs. Once someone was agreeable to receiving treatment, the first step was for outreach workers to get that person assessed through one of multiple NorthEast Treatment Centers (NET, netcenters.org), which opened a substance use assessment center in conjunction with the ERP. After an assessment, the person who sought treatment underwent detoxification or withdrawal management and was then matched with a treatment regimen. There were a range of available treatment regimens, provided both on inpatient and outpatient bases, and for varying lengths of time. For the ERP, treatment slots were coordinated by outreach workers and CBH in an effort to get as close to on demand access as possible through breaking down barriers, engaging on site at the encampments and providing transportation to assessment and treatment sites. The availability of treatment slots appeared sufficient to where 25 outreach workers extended such services for those in the encampments that were not among those on the BNL. Agreeing to treatment was voluntary and this underscored the importance of a speedy transition from requesting treatment to receiving services. There had been concern expressed about the length of the intake process, particularly for people with opioid use disorder for whom rapid access to treatment increases chances for successful engagement. Barriers that were addressed during this pilot included identification acquisition support and quick and reduced pre-authorization from third-party payors, as well as withdrawal management. Outreach workers recounted how the desire for entering treatment was often impulsive, catalyzed by particular circumstances and state of mind. Delays or barriers to accessing treatment services often meant that the requesting individual would have a change of mind or will not follow through. For this pilot, CBH, the primary payor of treatment services for Medicaid-eligible individuals in Philadelphia, made substantial changes in the intake process to facilitate smooth, expedient access to treatment, many of which were lessons learned that informed system improvements going forward. This included providing assessment and authorization on demand, with requirements for identification and preauthorization waived and, more generally, minimizing the time required for intake and authorization after admission. Outreach workers were positive about these changes, and these policy and programmatic changes have now been adapted more broadly throughout the substance use treatment system. Similar to housing services, there was a question of transitioning persons who entered treatment through the ERP to longer-term recovery services that address both recovery and housing needs. This includes placement in more long-term recovery housing, medication assisted treatment, and long-term residential services such as DBHIDS’s Journey of Hope program. Altogether, such services are in short supply. As with housing, City officials acknowledge the need for more long-term options and, at the time of the encampment closures, were looking to create sufficient long-term services for engaging anyone entering treatment through ERP and looking to enter longer term recovery or housing services. 4.2.3- Other Services Outreach workers were also able to offer a number of other services to individuals living in the encampments. One popular service included transportation to and assistance with getting identification, which was offered two days per week through OHS. Daily transportation for identification and other destinations was provided. Mobile clinics visited the encampments twice weekly and an outreach nurse accompanied outreach teams once a week. Included among these medical services on a limited trial basis was induction of buprenorphine at the encampments. Buprenorphine is a partial opioid agonist that can be dispensed as an alternative to illicitly acquired opioids. Appendix E contains a sample weekly outreach schedule of services made available at the encampments. 4.2.4- Services Summary As part of the ERP, impressive efforts were made to establish sufficient emergency housing and treatment resources to accommodate the target population, as identified on the BNL. This created a situation where a limited number of those in the encampments had housing and treatment options available to them. Respite and navigation center beds would have been accessed more widely if more of this housing were available in the more immediate neighborhood. Similarly, with treatment services, while questions were frequently raised about the available treatment modalities, there were services 26 available upon demand. For both housing and treatment services, the ERP underscored the need for removing barriers which discouraged unsheltered, substance-using persons from accessing services, and promoted the use and proliferation of more on-demand and low-demand services. These are noteworthy accomplishments. While there have been substantial achievements in the availability and accessibility of services under ERP, questions about the availability of follow-up services that provide long-term housing stability and recovery remained. Information in this area, particularly with regards to specific services received and outcomes of persons who were provided housing and treatment services through ERP, continued to be a major focus in the period following the encampment closures. Provision of suitable long-term services required a continued resourcefulness along the lines of what had already been achieved with the more short-term services described here. 4.3- Service Engagement Having profiled both the population sleeping in the underpass encampments and the services available to them, we now examine how many persons received services during the implementation phase. Specifically, we provide results on the initial placements made and the impact that they had on the encampment populations. Facilitating placements in the housing and treatment services reviewed in the previous section was a primary duty of outreach workers. Their first task was to forge relationships so that they were able to assess service needs and gain the trust of those in the encampments. When people did agree to leave the encampments, the outreach worker facilitated their service entry. In this process, the suitability, availability and accessibility of the services offered, described in the previous section, were the key to maximizing both the effectiveness of this outreach work and the potential for using the closure process to facilitate a lasting exit from homelessness. 4.3.1- Engagement Numbers The numbers of persons placed in services is based upon reports provided by City representatives and presented at ERP meetings. As acknowledged in an ERP planning meeting, there was a need to produce more exact accounts of placements and services received.23 Still, the following numbers provide an impression of how many people were moved from the underpass encampments to either housing and/or substance use treatment, or received other services during the implementation phase of the ERP. The day after the Kensington and Tulip encampments were closed, an internal DBHIDS memo24 reported: - Outreach workers engaged with 256 individuals at the two camps. This meant that they had at least one meaningful contact with each of those persons where they got a chance to assess services needs and discuss services options with the engaged individuals. Of the 256 who were engaged, 126 persons (49 percent) received at least one placement to housing or treatment services. On June 1, 90 people (35 percent) remained in that placement. - Of the 110 individuals who were on the BNL (a more targeted subset of the 256 people who were engaged by outreach workers), 83 (75 percent) had at least one placement. On June 1, 61 people (55 percent) had retained a placement. 23 Based on notes taken by evaluation team at the June 15, 2018 ERP planning meeting. 24 Benjamin Lambertsen, DBHIDS internal report, 6/5/2018. 27 - Placements in the navigation and respite centers totaled 83 (18 at PPP respite; 11 at the ODAAT respite; and 54 at the navigation center). As per June 1, 60 remained in housing placements (12 at PPP; 9 at ODAAT; and 39 at the navigation center). - An additional 20 persons received housing through other services (10 through Journey of Hope; 8 through a Safe Haven facility; and 1 each through veteran-specific housing and a recovery housing program). Additionally, one individual returned home to an out-of-town location. - Fifty persons accepted placements to substance use treatment and, on June 1, 15 remained in treatment. - Among other services, 51 individuals were linked to identification services. In addition, 31 referrals were made to intensive case management through the Behavioral Health Special Initiative (BHSI). All in all, the ultimate number of persons on the BNL, 110, was somewhat higher than anticipated, but the 83 persons placed into respite or navigation housing was very close to the 90 beds established to accommodate the projected demand. While a target did not seem to be set for treatment placements, adding the 50 individuals connected to treatment (with some overlap with housing) and other miscellaneous housing options resulted in three-quarters of those on the BNL and just under half of all engaged persons accepting housing or treatment services. 4.3.2- Police Counts The PPD maintained a presence in the encampment area with the objective of ensuring safety and enforcing the 30-day evacuation mandate. It was widely understood that they would not arrest or cite persons in the encampments for other reasons. The PPD also conducted weekly overnight counts of the number of individuals sleeping at the four underpass encampments. While the number of people in the encampments increased during the day with people who were not staying at the encampments, these late-night counts offered snapshots of the actual size of the unsheltered homeless population sleeping at the encampments. As such, this series of counts offers a blunt measure of the impact of the ERP activities on the number of people sleeping at the camps. 28 250 200 Emerald  Frankford Kensington  Tulip 30 35 49 40 150 22 100 50 0 48 47 46 38 40 38 38 43 41 40 34 40 23 163327 24 42 21 37 31 33 32 51 62 35 52 53 51 52 46 67 23 28 78 102 91 48 48 48 48 45 73 87 54 51 49 47 62 36 51 Figure 4a- Overnight census counts for underpass encampments25 4.3.3- Service Engagement Summary Based upon these counts, there was no measurable impact of the service placements on the size of the remaining encampment populations. The most apparent explanation for this was that the spaces in the encampments vacated by those entering housing or treatment were promptly taken up by others moving into the encampments as part of the population turnover. Outreach workers pointed out, with some degree of frustration, this paradox of rising placement numbers set against steady encampment population counts. On the other hand, the post-closure reduction was likely a function of space limitations in the remaining encampments, as police observed considerably more crowding at the Frankford and Emerald encampments in the wake of the closures. After tallying the placements and moves to the other encampments, police estimated that approximately 40 people staying at the Kensington and Tulip encampments at the time of the clearance were subsequently unaccounted for.26 Where these people went is unclear. Police observed an increase in people sleeping in visible outdoor locations along Kensington Avenue and a few other areas. However, in the immediate aftermath of the closures, no new encampments arose and the Kensington and Tulip underpasses remained empty. Outreach services also continued to engage persons in the area. In the month following the encampment closures, however, it was premature to draw any conclusions 25 Data source is Philadelphia Police Department, East Division. Data points for Kensington and Tulip encampments for May 3 were missing and were extrapolated from respective time series. 26 Based on notes taken by evaluation team at the June 8, 2018 ERP planning meeting. This population of those sleeping at encampments in late May 2018 is different from those tracked by outreach services based upon the BNL and subsequent contacts. 29 March 1 March 7 March 15 March 21 March 29 April 5 April 12 April 18 April 26 May 3 May 10 May 16 May 24 May 31 June 7 about how the camp displacements would affect the long-term geography of unsheltered homelessness in the Kensington area. 4.4- Perspectives from People Staying in Encampments After the closure announcement on April 30, the evaluation team conducted interviews with a sample of individuals staying in the Kensington and Tulip Street encampments. A semi-structured interview guide elicited open-ended responses along four domains informed by the Outreach Encampment Survey and previous RARE assessments27: living situation, typical day, background and service use, and perspectives on the encampment closure. (See Appendix F Interview Guide.) Qualitative data included in this report was collected during interviews with 25 individuals and 2 couples between May 16-29 in the Kensington (n=11) and Tulip Street (n=16) encampments. The demographic make-up of individuals interviewed was similar to respondents who completed the Outreach Encampment Survey: the majority of those interviewed were white, male, and under age 45. 4.4.1- Living Situation Two couples were interviewed, and others reported living with a partner at the camp or staying in the camp while waiting for their partner to return from incarceration. Most reported being homeless, though five individuals did not; two because they had accessed respite shelter through the ERP and had only returned to the bridges to use drugs. Long-term experiences of homelessness (i.e., a year or more) were uncommon among those interviewed. Typically, those interviewed had heard of the camps from someone they knew who lived there or had noticed the camps while in Kensington to buy or use drugs, and then a series of precipitating events (e.g., job loss, eviction), often related to their substance use, led them to stay there. Most were from Philadelphia or the surrounding areas, including New Jersey, Delaware, and other parts of Pennsylvania. Interviewees identified a number of positive aspects of staying in the camps. Generally, people staying there looked out for one another, and there was a feeling of “safety in numbers,” especially as it related to substance use. Other people under the bridge could—and often did—respond if someone experienced an overdose to ensure they received medical attention. The relationship with law enforcement was also generally seen as positive as PPD did not bother those living or using under the bridges and even brought trash bags and helped organize clean-ups to keep the spaces habitable. Shelter from the elements, including rain, cold, and heat, were also described as benefits of staying in the underpass encampments. In addition, community groups and residents who shared food, clothing, and other items, were seen as helpful. Finally, the proximity to drug dealers as well as markets and other places that sold items people needed, was an important benefit of staying in these locations. One interviewee described the benefits and disadvantages of living in the camp in the same way: the bridges are like “a drug addict's vacation.” While staying under the bridges conferred the advantages described above, several interviewees felt that the negatives outweighed the positives. While the underpasses offered some shelter from the elements, those living in the camps would still get wet in the rain and cold in the winter, and despite regularly scheduled clean-ups, the areas were still described as smelling bad, in no small part because residents had nowhere else to urinate and defecate and nowhere to bathe. One interviewee described her camp as an eyesore on a major thoroughfare and expressed exasperation and frustration that other addicts couldn’t be bothered to put needles into sharps collection containers. Indeed, another bemoaned that “kids can't even make snowballs in the neighborhood because of all of the needles everywhere.” Individuals under the bridges faced 27 Rapid Assessment interview guides provided by evaluation team member David Metzger. 30  harassment from local youths and other neighbors who would drive through shouting and sometimes throwing things at individuals in the camps, prompting one interviewee to ask, “why beat a dog while it's down?” In addition, while interviewees expressed that they generally looked out for one another, several also reported assaults and robberies. One woman described being physically and sexually assaulted and faced ongoing retribution from others staying in the camps when she attempted to refuse their sexual advances. 4.4.2- Typical Day All but one of the individuals who participated in interviews were heroin users, though cocaine use was also quite common; their daily schedules revolved around getting money to purchase drugs and doing drugs. The most common ways of financing their habits were panhandling and scrapping (which involves searching for metal-based products on the street and exchanging them for money). Others made money by participating in the drug trade or sex work. One reported that he would do anything to make money for drugs “except go against my morals” and reported having at least 30 different “hustles” to make cash. At the end of the day, as another described, if he had enough money left over, he would get something to eat, but otherwise, life revolved around drugs. “It’s like groundhog’s day out here,” reported one about this ongoing cycle, referencing the film where the protagonist wakes up each morning to find himself repeating the same day, including every individual encounter, utterance, and dialogue, over and over again. 4.4.3- Background and Service Use Some of those living in the encampments were interviewed and the following are their stories about how they got to the encampments. The individuals interviewed described the trajectory of their substance use that brought them to the encampments. Some described using substances when they were as young as 11 years old out of curiosity, boredom, or simply the accessibility of substances and then eventually moving on to heroin. Others reported transitioning to heroin following injuries for which a physician prescribed legal opioid medication. For example, one interviewee described getting injured on the job and receiving a prescription for pain medication. He subsequently became addicted to them, and when his physician “cut him off” from the medication, he began to purchase pills illicitly. In response to the high price of illicit pain pills, he turned to using heroin, which was much less expensive. Most, if not all, of the individuals interviewed had participated in substance use treatment, usually more than once, at some point in their lives, and many had experienced extended periods of sobriety. However, all had relapsed and come to the Kensington area bridges. Their stories were often similar: a common theme to several responses was that it wasn't necessarily that treatment didn't work; instead they referred to their own desire to go back to using drugs that contributed to relapse. One interviewee went into detail about treatment challenges—including overcrowding, understaffing, and lack of helpful resources and information—but still described the biggest barrier to recovery as herself. Another acknowledged that the “aftercare part” was what didn't work for him, stating that he felt like he wasn't being pushed or really engaged by treatment staff and that he felt like he was just “there to be there.” For those interested in getting back into treatment, not having the necessary identification was a major barrier, and for one interviewee from outside of Philadelphia, accessing treatment proved to be particularly challenging due to residency issues. Despite prior and current setbacks, interviewees often had plans to access treatment again, and several had concrete plans to do so within the coming days or weeks, in part because they were being encouraged to seek treatment through the ERP. Individuals staying under the bridges also described other resources that they had accessed in the past, were actively engaged with, or planned to use, often as a result of the ERP. Shelter use was described by a couple of interviewees and several were already staying at respite centers. The respite center in which 31 one was staying was described as “nice; there's no curfew, you can clean up, do laundry, and they offer breakfast and dinner.” Having options for couples to stay together was important to those who didn't want to be separated from a partner. In contrast, others considered available options to be restrictive in regards to the hours you could stay there, with rules and regulations that were seen as onerous. Distinctions between shelters and respite centers were not always made, and perceived negative conditions of temporary housing were often emphasized.28 Several also pointed out that the number of individuals in need of services exceeded the amount of resources available. 4.4.4- Perspective on the Encampment Closure While a handful of interviewees didn't respond positively to outreach efforts—describing outreach workers as “pushy” and “in your face”—most who had interacted with them reported that they were helpful— “beautiful people” who were “awesome” and “should be put on a pedestal.” However, overall, feelings were mixed about the ERP. One described it as “putting a band aid on a bullet wound." Many felt that the initiative would be futile as there weren't enough resources to serve everyone. One interviewee suggested that without having a long-term housing strategy in place for those the City is seeking to serve, the ERP ends up “hurting the hurting.” Another felt that the initiative wasn't meant to help people get into treatment, but rather get them out of highly visible areas, particularly since drugs, and heroin in particular, are such an ingrained issue in Kensington and the City is only now beginning to address it. This interviewee had reflected on the complexity of the issues and concluded that until the root of the issue of drug abuse was addressed, other solutions would not be sustainable in the long term. In general, those who participated in interviews didn't expect the ERP to be successful—even those who were accessing shelter and treatment through these efforts. They felt that homelessness would continue to exist in Kensington, but it would move to other places in the neighborhood, including the other camps, abandoned properties, and the steps of the transit station, before likely returning to the Kensington and Tulip Street bridges when things settled down. One predicted that following the closure, people experiencing homelessness would “be in the neighbors' faces,” using drugs and doing other things related to drug behavior in an even less conspicuous manner, and one likened the City’s efforts to “an animal being backed into a corner,” portending that it would lead the people staying in the encampments to fight back out of fear. Many of the individuals interviewed did not have any specific plans for where they would go following the encampment closures. While several were planning to enter respites or other temporary housing when the encampments were shut down, others were darkly dismissive when asked what they would do: “I don’t know—commit suicide.” Some reported that they would delay even considering what they would do until they were forced with that reality; as one interviewee put it, he hadn’t taken the time to figure out what he would do because he’s too busy “figuring out ways to get enough money to survive” (i.e., earn money for food and drugs). For some, this perception extended to long-term goals as well. Like one interviewee who lamented that she had formerly “had every goal” but that heroin had taken their place, several reported that getting high was their only goal. Most, however, wanted to get sober and “live a normal life.” These individuals had goals that included finding housing, accessing treatment and services, returning to school, and working. They expressed a desire to “put this all behind” them and, particularly for those with children, reconnect with family. Some hoped that their experiences on the street could be used to help others dealing with addiction through art, advocacy, or business ventures that could provide meaningful activities for those caught up in addiction. 28 Examples of such negative perceptions included one assertion that respites were “madhouses” where theft was rampant due to the large number of people confined to a small space, and another with concerns about being forced to move to temporary housing, where he stated “that’s not what America’s about.” 32 4.4.5- Perspectives from People Staying in Encampments Summary Individuals staying in the Kensington and Tulip Street encampments did not describe their living conditions much differently than those connected with the ERP, community members, or advocacy groups. They acknowledged the benefits of staying in encampments that offered some refuge from the elements; regular street cleanings provided by the City; a relatively safe place to use drugs in a somewhat supportive milieu; and little fear of legal retribution for openly using illegal substances. Additionally, with advocacy groups and others bringing food and supplies to the encampments, those staying there were able to procure basic necessities as they focused on securing and using drugs. The interviewees also echoed the same concerns about needed housing and treatment options. However, they also expressed in more explicit terms what it meant to "not be ready" for treatment, especially since most had been in some sort of treatment in the past. This begs the question of the types of treatment these individuals accessed and the ways in which it could provide a better fit to engage individuals. Finally, those staying in the targeted encampments were almost completely absent from the ERP planning and implementation efforts. Two “town hall” meetings were held in locations designed to maximize participation from those in the encampments, but just one from this constituency showed up and plans for two additional meetings were scrapped. While there were rumors of camp leaders organizing from within the Tulip Street encampment, no one ever emerged as a spokesperson representing either encampment. During interviews, those staying in the camps reported that they did not expect the CIty's efforts to resolve the housing and opioid crises in Kensington. In some ways, they thought closing the camps would make the issue worse, forcing people in the throes of their addiction away from cover of tents and the safety of their communities into the neighborhood and even closer to the watchful eyes of community members. 4.5- Stakeholders Beyond the unsheltered people experiencing homelessness in the Kensington area, there were two other larger constituencies who were involved in the encampment closure process. One group, referred to here as community members, were residents and business owners in the area directly impacted by the presence of the underpass encampments and the widespread drug-related activity in the area. The second group, referred to here as advocates, were people who saw themselves as taking up the concerns of those staying in the encampments. The advocates addressed immediate subsistence needs, such as food and health care, as well as broader issues related to the closure process, such as greater availability of affordable housing and substance use treatment, and for various harm reduction approaches towards substance use such as the establishment of safe injection facilities. The community members and the advocates typically expressed different sets of interests. These interests, while not monolithic, were often at odds with each other, as community members generally prioritized eliminating the encampments and improving the quality of life and reputation of the area while advocates prioritized the well-being of those staying in the encampments, whose lives they saw as potentially endangered by the encampment closures. This section provides a more detailed overview of these positions, and are based upon open-ended interviews conducted with individuals representing each of these two stakeholder groups; from observations gained from attending various public meetings; and from comments left on social media sites. Interviews were based upon an interview guide that is in this report as Appendix G. 33 4.5.1- Community Members The encampments of homeless opioid users are emblematic of the national opioid crisis, and they are also the most immediate concern [bold in original] of neighbors. With the encampments come real and perceived threats to child safety, very real health hazards such as the spread of Hepatitis C (as is the case in San Diego), open sex trade which is viewed as the victimization of men and women who are supporting their addiction, open drug use, trash, defecation and urination in the spaces that we live and play, as well as exposure to used syringes.29 ****** [P]lease take a moment to put yourself in my shoes. Right now, my child is in his bed sleeping less than a hundred feet away from this encampment that is full of people with addictions and some possible mental issues. He also has to walk by them to go to school. ... I see treatment vans pull up here trying to get them to go to rehab facilities and the vans leave empty. Something has to give, this city put a plan in place and I don't think it is fair for it to be protested against without it even starting.30 These passages represent the concerns of community members. The first, in a measured yet urgent tone, was from a coalition of neighborhood organizations, and the second, more unvarnished message came from a neighborhood resident. Central to both messages was the untenable nature of the underpass encampments coexisting amidst the commercial and residential areas of Kensington. While the voices from the Kensington community expressed varying degrees of empathy for those staying in the encampments, the central theme of this position was that the encampments needed to go. The politically untenable scenes created by these encampments contributed to the formation of what would become the ERP planning group. In an interview, a Prevention Point staff member recalled knowing the clearances were “going to happen” given the community push-back and “how fed-up the neighborhood was getting” with this situation. In response to this, the ERP planning group was careful to solicit input from the community throughout the planning process. In return, the community response, as represented by civic organizations and the majority of residents and business owners that participated in community forums, was guardedly positive. During the implementation phase, part of the ERP was to have four “town hall” meetings held in the immediate encampment area to provide an opportunity for those staying in the encampments as well as any other interested persons to participate. Additionally, City officials presented on the ERP in various community settings. In one meeting of “El Barrio es Nuestro,” a series of meetings organized by the Community Services Cabinet of the City’s MDO to address more general community issues in the Kensington/Fairhill area, Deputy Managing Director Joanna Otero-Cruz presented the ERP in terms that highlighted: 29 From “Partnering to improve quality of life in Kensington, Harrowgate, and Port Richmond,” a joint letter from the leadership of the Harrowgate Civic Association, the Somerset Neighbors for Better Living, and West Kensington Neighbors to the ERP planning group, February 19, 2018. 30 Comment from a Kensington resident on a publicly accessible social media site that announced a protest against the encampment closures. May 21, 2018 (https://www.facebook.com/events/959839337513714). 34 - an opportunity to do something about the encampment problem and learn from it. The Kensington and Tulip camps would be closed initially, and this approach would then be adjusted and applied to the other two bridges; - how the pilot was limited by available resources, particularly shelter beds and, to a lesser extent, treatment beds; - how the ERP was modeled on a successful approach developed in San Francisco; - the BNL, and how this led to placements in housing and substance use treatment services in which the more people were connected to services, the better; - the process where the people would be cleared out of the two encampments on May 30, including storage procedures and retrieval opportunities, the need to address people with respect and dignity, and the awareness of how these camps have impacted neighbors’ lives; and - the assertion that the City couldn’t continue to do nothing, reiterating how they were using this pilot to find out what was working and what was not. Otero-Cruz’s presentation drew questions from meeting attendees that were representative of concerns that were more widely expressed by local residents and civic organizations. These questions focused upon: - Long-term housing resources for those staying in the encampments. Otero-Cruz acknowledged that such housing was scarce, and that funding for permanent housing arrangements were part of upcoming budget requests. This mirrors the concerns, expressed elsewhere by civic associations, that the navigation and respite housing resources established in conjunction with the ERP would extend beyond the temporary arrangements under which they were established.31 - Impact of displacing those staying in the encampments. There were concerns that those displaced by the encampment closures would regroup and create new encampments. Otero- Cruz responded frankly that the ERP planning group was working with input from “the civics” [organizations] on this and that “we have the same fears.” - Focusing on the encampments without focusing on the larger issue of drugs. In response, Otero- Cruz stated that “we can’t police our way out of this” and that the pilot “can’t be solely an enforcement strategy.” Along with deemphasizing an enforcement approach in closing the encampments, she cast this in a context of continuing police efforts that targeted drug trafficking and a “block by block strategy” where residents partner with the City to identify crime and blight.32 Many of the concerns expressed by community residents extend beyond the immediate objectives of the ERP. Establishing new beds at respites and navigation center feed into perceptions that the Kensington area has been a dumping ground for undesirable social services facilities such as shelters, 31 This concern is expressed in the aforementioned joint letter from three neighborhood associations (see note #20) as well as in a March 19, 2018 letter to the ERP from the North Kensington Community Development Corporation. 32 The extended account of the El Barrio es Nuestro meeting were based upon an evaluation team member notes. The meeting was held on May 22, 2018 at the McPherson Square branch of the Free Library of Philadelphia. 35 recovery houses, and, most recently, for a proposed overdose prevention site (OPS). The encampments contribution to petty crime, drug dealing, needle debris, and public health concerns were parts of larger frustrations expressed by community members. Finally, there was an awareness of the limited resources at the disposal of the ERP, and how cleaning up two encampments represented a very specific response against a much larger problem. 4.5.2- Advocates We [protest organizers] want the city to create enough treatment and shelter beds for them [people sleeping in encampments] as well as a safe injection site where those that aren't ready for recovery can at least do their drugs safely indoors and away from the eyes of children. We are very upset that 1200 people OD'd last year and we're on track for an even deadlier 2018.33 Where community members placed an emphasis on the neighborhood, advocates centered their position on the population that would be displaced by the ERP. The advocates were generally less able to mobilize political support than community members and had less representation and participation in the ERP planning process. Some held positions contrary to the encampment closures, based upon concerns over the availability of permanent housing, treatment and safe injection opportunities for those in the encampments, and their vulnerabilities for public health risks given their injection drug use. Three organizations had some degree of visibility in representing the advocacy position during the implementation phase of the ERP. Mental Health Partnership (MHP), the only prominent advocacy organization to take an active role in the encampment closure process, has been involved in Kensington on the opioid issue since the clearing of the Gurney Street area and has subcommittee representation on the Mayor’s Task Force to Combat the Opioid Epidemic.34 In May, MHP performed outreach (independent of ERP-affiliated outreach) in the encampments and, as part of this, observed cleanups, provided material support, and disseminated information about rights and available services related to housing, mental health and substance use. An MHP program manager characterized MHP as having a monitoring function in this process. While acknowledging substantial agreement with the housing- oriented approach toward the encampment closures, he was critical of the lack of resources the City provided for housing for those staying in the encampments. The MHP sent an official team of observers to monitor the encampment closures on May 30 but did not take a stance as an organization as supporting or opposing the ERP.35 A more grassroots advocacy group with a visible presence at the encampments was Angels in Motion (AiM). AiM’s activities centered on going to the encampments and providing food, toiletries and other material assistance to those staying in camps while helping to “spread awareness about the disease of addiction.”36 The founding director, well-known and respected by many in the encampment milieu, by all accounts had good relationships with ERP-affiliated outreach workers. When asked, she questioned 33 Comment from an organizer clarifying the aims of a protest against the encampment closures. Comment was made on a publicly accessible social media site that announced the protest. May 21, 2018 (https://www.facebook.com/events/959839337513714). 34 See https://dbhids.org/opioid. 35 Accounts were taken from an evaluation team member notes of a phone interview held on May 16, 2018, and on-site at the Kensington encampment on May 30, 2018. 36 Quote is from AiM website (https://aimangelsinmotion.org/about-aim/). 36  the wisdom of shutting down encampments without providing accessible alternatives, stating that many of those staying in the encampments were not ready to “come inside.” She also spoke of advantages, including a sense of community, that emerged in encampments, and described efforts to open an alternative, sanctioned encampment more removed from commercial and residential areas where people could access portable toilets, medical care and substance use treatment coordination. She mentioned, as a possible site, property under an interstate highway overpass located east of the encampments, which is the property of the Commonwealth of Pennsylvania.37 This position was contrary to the City's policy of discouraging encampments from being established, and members of the ERP planning group pointed to the self-defeating prospect of facilitating a sanctioned camp that they might one day have to dismantle.38 The third visible advocacy proponent was an individual with no known group affiliations who organized protests against the encampment closures on the evening before and the morning of the encampment closures (May 29 and 30). This advocate, who described herself as a neighborhood resident who was in recovery, emphasized, in her rationale for opposing the closure, the harm that people displaced from the encampments would face in the absence of sufficient housing, substance use treatment and safe injection facilities. The protests functioned to pull together smaller, like-minded and less visible advocacy groups.39 While public protest was not unexpected, preparing for the possibility of an adversarial demonstration required additional planning to clear the encampments. Ultimately, the protests each drew approximately 40 people and did not interfere with the closure process, but did draw press coverage. Finally, there were various other “good Samaritan” groups, many affiliated with faith organizations or consisting of people in recovery, that similarly aided persons in the underpass encampments. An evaluation team member encountered a group of three women from the surrounding community who visited the Tulip Street encampment at least once a week to provide water, sandwiches and toiletries to residents. When asked for their thoughts on the encampment closure, they stated that it was a good idea, but were concerned about where people would go and the availability of resources for persons staying in the encampments. This ambivalent perspective, with most advocates ultimately expressing opposition to clearing the encampments, was relatively consistent among those providing aid in the encampments. Some advocates felt that they, along with those facing displacement, were shut out of the ERP planning and implementation processes. One advocate expressed in an email to an evaluation team member that: This decision was made without the leadership and involvement of the people most impacted -- those living in the encampments. While there has been talk from City officials of some survey having been done, I have not seen it nor has anybody I've spoken to in the encampments had any knowledge of it. For the City to decide on a timeline, approach, array of services, etc. without the involvement of those most impacted makes 37 Accounts were taken from an evaluation team member notes of a discussion with AiM, held at the Kensington encampment on May 22, 2018. 38 Views of ERP planning committee on encampments were taken from an evaluation team member notes of an ERP planning meeting held at the Municipal Services Building on May 18, 2018. 39 Examples of such groups, many with social media sites, include SOL Collective, The Philadelphia Chapter of the Young Patriots Association, In My Backyard, The Sunday Love Project. 37 this whole initiative problematic from the start. Also, have people most impacted informed the evaluation process? 4.5.3- Stakeholders Summary Antoinette Kraus, Director of the Pennsylvania Health Access Network, wrote that with ERP the City: is stuck in a predicament between responding to the safety concerns of area residents and addressing the needs of those who are homeless and have behavioral health issues.40 Here the City, in trying to implement the ERP, found themselves in the middle of two stakeholder groups that were both concerned and critical, in different ways, with the adequacy of the resources available to implement this approach. Both community residents and advocates pressed the City in different ways to be responsive to their views. One success of the ERP process has been some capacity to accommodate both, though each side sees the ERP as falling far short of what they would consider an ideal outcome. Community residents had a network of community organizations whose views were solicited and considered during the ERP planning and implementation processes. Nonetheless, one of their primary concerns, that the respite and navigation center arrangements in the Kensington area remain temporary, remained unresolved during the implementation phase, as there were no stated plans about the future of these beds. Community members also were left with more general questions about the extent to which the two camp closures would ultimately impact the quality of their neighborhoods. Advocates, and by extension the those in the camps, did not feel that their voices were adequately considered in the closure process. The ways in which they mobilized were outside of the local political structure of community groups and municipal political representation, and consisted primarily of external, more confrontational means such as demonstrations and media coverage. The success of this tack was limited due to the small number of active advocates and the limited involvement of those staying in the encampments. While City officials were resistant to their demands of keeping the encampments operating, there was substantial overlap with advocates’ more general positions around such issues as housing and harm reduction. One member of the City’s ERP leadership team, OHS Director Liz Hersh acknowledged, in an ERP planning meeting after the closures, that the advocate perspective had not been sufficiently represented in the ERP and resolved to reach out to this constituency to find ways to more effectively solicit their input.41 4.6- Encampment Closures The implementation phase of the ERP culminated with the closing of the Kensington and Tulip underpass encampments on May 30. Planning for this event had been going on for several weeks, coordinated out of the City’s MDO. The key components of the closure process were: - Outreach services continued to be the primary point of contact for anyone sleeping in the encampments and were prepared to provide housing and treatment services during the course of the closure process. 40 Antoinette Krauss, “Philly's heroin camps illustrate how housing instability hurts health.” Philly.com, May 29, 2018 (http://www.philly.com/philly/health/addiction/phillys-heroin-camps-illustrate-how-housing-instability- hurts-health-20180529.html). 41 Based upon an evaluation team member notes of an ERP planning meeting held at the Municipal Services Building on June 1, 2018. 38 - PPD provided security for the closure operations and were prepared to issue citations to anyone not complying with the closure notice. After camp clearance and closure, they were responsible for preventing the repopulation of the camps. They also monitored protests against the encampment closure. - Licenses and Inspection and PPP collaborated in managing and storing property left at the encampments. This included a process whereby receipts were issued with which persons leaving the encampments could reclaim their property within six weeks.42 - Streets Department took the lead in disposing of the trash and debris left behind in the encampments, with the Community Life Improvement Program (CLIP) then power-washing the underpasses. - PPD and the MDO provided communications personnel to coordinate public and media relations. This included organizing a press conference immediately following the camp closures. Figures 4b and 4c. Tulip Street underpass before and after closure and cleanup (sources: photos posted on social media by Brooke Feldman (4b) and Joe DeFelice (4c)). Two evaluation team members attended the closure operations, which commenced at 8:00 a.m. at the Tulip Street encampment. At around 9:00 a.m., approximately 30 protesters gathered at the corner of Kensington and Lehigh Avenues to condemn the City’s action and march through a pre-approved route through Kensington. Press coverage was highly visible at the Kensington underpass and at the protest, as were observers from the American Civil Liberties Union and the MHP. At 10:00 a.m. the closure 42 Per policy, the customary time for retaining property turned in at encampments is 30 days. The City extended this time frame to accommodate the July 4th holiday and concerns this might have impeded accessibility. 39 process at Kensington Avenue formally started. At both encampments, many who had stayed there had already left with their belongings prior to the start of the closures. Few people submitted any belongings to be held in storage. Once the encampments were cleared of people, police inspected the area and trash trucks then picked up and hauled away large amounts of remaining trash and debris. Following this, both underpass areas were power-washed and closure operations were finished by noon. Both underpasses were essentially cleared of any indication that encampments had existed there (Figures 4b and 4c). The encampment closure operation was considered successful in that the encampments were cleared on schedule, with no citations or arrests, major disruptions, or controversies. A news conference at noon provided closure and official commentary to the closure process.43 The media coverage of the closure, which also featured accounts of the protest and quotes from displaced persons who stayed, was uncritical of the closure process.44 The Philadelphia Inquirer opined that “the City deserves credit for creating what is essentially a thoughtful response: one that puts humanity first and counters the usual government response of denial, incarceration, demonization, and despair.”45 In the ERP planning meeting following the closure, Deputy Managing Director Eva Gladstein congratulated the planning group, stating the closures “were accomplished as well as could be ... what had to be done had to be done.”46 43 A City of Philadelphia press release provided an official summary of the closure: “City Completes Encampment Pilot Project in Kensington Area,” May 30, 2018 (https://beta.phila.gov/2018-05-30-city-completes-encampment- pilot-project-in-kensington-area/). 44 A sample of major media coverage of the closure includes: Aubrey Whelan, “Philadelphia sweeps aside 2 heroin camps in Kensington. What happens now?” Philly.com, May 30, 2018 (www.philly.com/philly/health/addiction/ philadelphia-sweeps-aside-2-heroin-camps-in-kensington-amid-protests-20180530.html); Joel Wolfram, “Philadelphia clears encampments of homeless drug users in Kensington.” WHYY, May 30, 2018 (https://whyy.org/segments/philadelphia-clears-encampments-of-homeless-drug-users-in-kensington/); Edward Helmore, The Guardian (see note #43). 45 Editorial, “Philly deserves credit for clearing out heroin encampments, offering help,” June 1, 2018 (http://www.philly.com/philly/opinion/editorials/opioid-philly-kensington-camps-20180601.html). 46 From an evaluation team member notes of an ERP planning meeting held at the Municipal Services Building on June 1, 2018. 40 Figure 4d. Scene from the Tulip Street underpass during the closure process (source: The Guardian47) 47 Edward Helmore, “How Philadelphia closed homeless 'heroin camps' amid US opioid crisis.” The Guardian, June 2, 2018 (https://www.theguardian.com/us-news/2018/jun/01/philadelphia-homeless-heroin-bridge-camps). 41 Chapter 5- Phase 3: Sustainment through Continued Engagement and Monitoring At the close of the implementation phase, the two targeted encampments had been cleared and, by the best available counts, 83 individuals who had stayed in the encampments had engaged in housing or treatment services. While there was universal acknowledgment that this pilot had only a limited impact on the more general concentrations of homelessness and opioid use in the Kensington area, by most accounts, they had met their immediate objectives of closing the two encampments and using the closure process as a portal into housing and recovery services. Still, a key part of the ERP remained. This post-closure period, called the sustainment phase, had two basic objectives. The first objective was to provide continued monitoring and engagement of those who were targeted for services during the implementation phase with the goals of maintaining outreach engagement, continued availability of short-term housing and substance use services, and assistance with transitioning those already in these short-term services to more long-term housing and recovery services. The second objective was to ensure that the encampments remained clear and that new encampments did not supplant the gains made in the closure process. This formal framework for sustained engagement with those displaced by the encampment closure for an extended period beyond the closure is, to our knowledge, unprecedented among any prior US efforts to close encampments. This report also represents a unique means of monitoring this process and reporting on the outcomes of those targeted for assistance in conjunction with the closures. The first section of this chapter tracks this process and the related outcomes. Here, we take an in-depth look at the use of a BNL to structure this sustained engagement, including the use of a range of City-based services both before and after the ERP initiative and where the target population was four and a half months after the encampments closed. Extending the closure process beyond the physical clearing of the encampments also provides an opportunity of reflection on the process. The second section of this chapter examines perspectives and reflections on the ERP from those who remain engaged with ERP-related services and from those involved in providing services. This offers on-the-ground views of the sustainment process as well as early assessments of what worked and how future initiatives of this nature could be improved. The third and final section of this chapter is a more conventional examination of how the encampment closures impacted the surrounding community. The presence of large encampments of persons experiencing homelessness, many of whom were actively and openly using illicit drugs, had become logistically and politically unsustainable. This section covers what changed in the wake of the closures, both directly at the camp locations and more generally in terms of impacts on homelessness and substance use in the area. Taken together, this chapter provides a more comprehensive assessment of the human and community impacts of closing homeless camps than has heretofore been presented. 5.1- Services Use and Outcomes among People on the By-Name List A key feature of this study was the ability to follow those targeted for services through their use of services, both before and after the ERP. This generated information on the extent to which those in the group had engaged in various City-funded and administered services, and what types of services they had used. In the case of behavioral health services, it also provided an overview of diagnosed mental health and substance use disorders that supplemented findings reported in the Outreach Encampment Survey reported in Chapter 4 and Appendices C and D. This section first reviews the construction and 42 use of a by-name list (BNL) to identify the target group and monitor their engagement, then presents aggregate results from matching names from the BNL with the City of Philadelphia’s integrated database of services records known as CARES; and follows housing and treatment placements over the course of the sustainment phase to provide outcomes related to the ERP services provision. 5.1.1- The By-Name List The BNL was a comprehensive list of all persons who were eligible to receive housing and substance use services under the auspices of ERP. BNLs have emerged fairly recently as a key tool for identifying, tracking and placing people who are chronically homeless and otherwise challenging to serve. Outreach workers and other social services staff used the BNL to coordinate services for this group, and DBHIDS staff tracked housing and substance treatment outcomes for this group through the implementation phase and into the sustainment phase. Not only was the BNL the backbone of ERP services provision and coordination, it also provided the basis for the analyses of services use and services outcomes that are presented in the next section. The BNL started with 110 names from the Outreach Encampment Survey and others that outreach workers thought to have stayed overnight in the Kensington Avenue or Tulip Street encampments on April 30, 2018.48 DBHIDS staff described the BNL as “where the magic happens”—that is, where providers were able to make a difference in individuals’ lives, trouble shoot, remove barriers, and coordinate follow-up to make sure no one and nothing falls through the cracks. In late June, well after the encampment closures, the BNL expanded to 189 names by including people who were engaging with outreach workers for services but who could not be confirmed as having stayed in the Kensington or Tulip encampments prior to their closure, which would have qualified them for the original list. The addition reflected a desire by social services staff to assist additional people experiencing homelessness in the Kensington area who were actively seeking assistance, as well as to fill the empty housing beds that emerged as attrition progressively shrank the number of those on the original BNL who continued to engage in services. Those on the BNL became the study group for our tracking of services use and related outcomes for this evaluation. Identifiers were available for those on the BNL, and enabled matching those on the BNL to services records in the City of Philadelphia’s CARES Integrated Data System. Persons on the BNL were also tracked for services receipt and outcomes by DBHIDS staff, who kept records of services engagement based upon their ongoing coordination of outreach, and case workers who were in direct contact with those on the BNL throughout the ERP sustainment phase. The primary drawback to this approach was that the aforementioned 79 additional people whose names were added to the BNL in June were likely very different in their motivation to engage in services than the 110 persons included on the original BNL.49 There was also no control group available for comparison with the study group. Logistical Challenges. Though the Outreach Encampment Survey was useful for identifying necessary and desired services and barriers to access for those staying in the encampments, staff reported that, in hindsight, it was not the best basis for establishing who stayed in the bridge encampments on April 30. While outreach and casework staff reported uncertainty as to the encampment affiliations of persons on the BNL, we were not aware of anyone claiming eligibility on this basis whose name was omitted from 48 April 30 was the day when the City of Philadelphia issued 30 days notice to vacate these two encampments. 49 While expanding the BNL in June likely increased the number of persons with successful services outcomes, all indications, based upon evaluator observation and interviews with outreach workers and city officials, were that this was with the intent to increase access to services and not with any direct intent to improve outcome metrics. 43 this list. Revised procedures for compiling BNLs for subsequent encampment closures use improved outreach data collection on where engaged individuals were staying. This allows subsequent BNLs to be created through ongoing outreach, rather than at a single point in time. Table 5a. Demographic Characteristics of People on the BNL (n=189) Gender Male 128 (67.7%) Females 46 (24.3%) Transgender, non-binary 1 (0.5%) Unknown 14 (7.4%) Race/Ethnicity Black (non-Hispanic) 50 (26.5%) White (non-Hispanic) 93 (49.2%) Hispanic (any race) 20 (10.6%) Other 3 (1.6%) Unknown 23 (12.2%) Age 18-24 6 (3.2%) 25-34 62 (32.8%) 35-44 67 (35.4%) 45-54 34 (18.0%) 55-64 9 (4.8%) 65+ 2 (1.1%) Unknown 9 (4.8%) As part of creating the BNL, the outreach team created a specific code alert within the CBH system to make workers aware of individuals from the BNL who presented for treatment. This facilitated coordination between ERP efforts and regular DBHIDS and CBH-funded services. However, management of the BNL proved labor intensive, as information was collected from other databases (e.g., treatment data came from the CBH system; outreach data through DBHIDS; and shelter use data through the City’s Homeless Management Information System [HMIS]) and from casework and outreach worker reports, and was then manually entered onto an Excel spreadsheet. Administrative records were often not updated in real-time, and keeping records up to date sometimes necessitated weekly emails or confirmation from assigned points of contact. For those on the BNL who were lost to follow-up, tracking was often even more challenging and included even more data systems (e.g., criminal justice). Data sharing was further complicated by HIPAA compliance procedures. Demographic Characteristics and Medicaid Eligibility. Data available from the BNL and record matches provide a basic demographic portrait of those on the BNL. This group was about two-thirds (67.7 percent) male, half (49.2 percent) White (non-Hispanic) and one quarter (26.5 percent) Black (non- Hispanic), with the median age occurring somewhere in the largest age group: 35 to 44. Over 85 percent of the people on the BNL were between ages 25 and 54. Table 5a contains further details on these demographic characteristics. 44 Demographically, the BNL group was similar to the respondents of the Outreach Encampment Survey50 (Appendix D). One reason for this was that the BNL group overlapped substantially with the Outreach Encampment Survey group, as 77 of the 189 people (40.7 percent) on the BNL were linked to this survey. Despite this, survey results do not necessarily reflect those included in BNL group. Given this lack of comparability, survey results, when referenced, are only used here for general context. Table 5b shows that 89.6 percent (all but 20 of the 189 people on the BNL) matched to some form of CARES record. More than half (55.5 percent) matched with CBH records that indicated that they were Medicaid eligible. Most of those who were Medicaid eligible (45.5 percent out of the total 55.5 percent) qualified through the expanded Medicaid eligibility criteria that was part of the Affordable Care Act. An additional third of the people on the BNL (33.9 percent) were not on Medicaid but had some other record of services use that was collected in the CARES system. Table 5b. Medicaid Eligibility and CARES Match for People on the BNL (n=189) Note. See Appendix I for more information on CARES and TANF/HB/MAGI Medicaid eligibility category. 5.1.2- Services Use Data from the CARES Integrated Data System In this section we examine the services use histories of the 189 people on the BNL based upon a match with records in the CARES integrated data system, which contains data from a range of municipal services providers. Analysts from the City of Philadelphia’s Health and Human Services Data Management Office (DMO) and CBH matched the BNL with CARES data51 and provided aggregated results to the evaluators in tables that the evaluators formulated.52 The six sources of records contained in CARES that were matched to the BNL are listed in Table 5c. 50 When appropriate, we compare findings based upon CARES records with those reported from the Outreach Encampment Survey that was administered just before the launch of the ERP. These results are reported in detail in Chapter 4 and Appendix C. 51 Records were matched based on first and last names, and date of birth. Because of the limited number of available identifiers, these elements had to match exactly (i.e., a deterministic match) in order for a name on the BNL to correspond with a CARES record. 52 Evaluators did not have access to identifiable BNL data or to any individual CARES records. This maintained confidentiality and anonymity of the data and, more practically, bypassed the need for data use agreements and other time-consuming data-sharing logistics between the City and the evaluators. City officials also declined to match BNL data with arrest data, citing concerns that such a match, even if done only for aggregate and evaluative purposes, could be used to identify people on the BNL with open warrants and other pending legal issues. Medicaid Eligibility (most recent) HC Expansion - Newly Eligible 86 (45.5%) SSI w/Medicare & Healthy Horizons 4 (2.1%) SSI without Medicare Adult 10 (5.3%) TANF/HB/MAGI Adult 5 (2.6%) No Eligibility but had CARES Records 64 (33.9%) No Eligibility nor CARES Records Found 20 (10.6%) 45 Table 5c. Services Use Data The records include those available prior to the ERP period (that is, prior to closure announcement), and during the ERP (May 1, 2018 through September 30, 2018). The time prior to ERP serves as a baseline for the extent of services received, and the services received during the ERP provides insight into both the number of services provided in conjunction with ERP, as well as a rough impression of changes in the degree of services provided (i.e., whether a particular type of services use increased or decreased during the ERP). The two time periods covered here – that prior to ERP and that during ERP—are not directly comparable as the prior period is much longer than the ERP period. A limitation of this analysis is that the records of services use during the ERP period were incomplete due to a time lag between when services were provided and when they were added to CARES. In other words, not all services provided during the sustainment phase, and particularly those provided later in the sustainment phase, will be included in the data used here. The degree to which this time lag impacts the records varies for different data sources. This lag was most pronounced with the substance abuse and mental health services data. Thus, the reports on services use during the ERP period should be considered conservative, and we assess the results with this in mind. The primary questions that we address with the results of the CARES data match are, first, the extent to which people targeted for services under the ERP pilot (i.e., those on the BNL) received various types of services, and, second, whether these services patterns changed after implementation of the BNL. This is a difficult analysis to undertake based upon the aforementioned data lag, the differences in time periods that are compared, and the lack of any type of control group. 53 OHS administers or funds approximately 85 percent of all Philadelphia shelter beds, and provides or supports a variety of other homeless services. OHS also coordinates and maintains Philadelphia’s Homeless Management Information System (HMIS), which keeps records of homeless services use, including shelter use, for all services providers that receive any funding through OHS. OHS has collected HMIS data at least since 1990, although there were gaps in their HMIS data collection between 2014 and 2016 due to issues related to switching data collection platforms. 54 The prison records do not include information on incarcerations in the state prison systems, nor do they have data related to the arrests, charges and convictions that precipitated the jail incarceration. These data are not included in CARES and were unavailable for this evaluation. 55 Specific data made available to the evaluation team included regular aggregated updates on services outcomes that DBHIDS provided throughout the sustainment phase, based on weekly updates provided by service providers and outreach workers; a deidentified copy of the BNL with service utilization and outcomes for each individual; results from specific data queries requested by the evaluation team; and qualitative information that provided context for these data. Service Type Source Description Homeless Services OHS Shelter use, outreach contacts, and related services53 Substance Use and Mental Health DBHIDS, CBH Diagnoses, behavioral health services, and services provided to low-income uninsured persons Philadelphia Jail Incarceration PDP Jail stays54 Child Welfare Services DHS Service data for adults with children who are receiving services ERP Outcomes DBHIDS Engagement and housing placements55 46 Homeless Services. In the Outreach Encampment Survey, 57 percent of the respondents indicated having spent time in a Philadelphia homeless shelter. In contrast, Table 5d shows that 38 percent of the people on the BNL had a record of a shelter stay prior to the ERP implementation.56 That proportion increased substantially after the ERP started, as over half of those on the BNL (98 people or 51.9 percent) used some shelter or temporary housing during the ERP implementation and sustainment periods. Much of this latter proportion reflects use of the navigation and respite centers, although the finding reflects use of any shelter in the City.57 Table 5d. Prevalence of Involvement in Services Provided OHS by People on the BNL (n=189) Table 5d also shows that increases in placements to Safe Haven housing and to Journey of Hope residential treatment occurred after the ERP commenced. While nine people (4.8 percent) had pre-ERP histories of Safe Haven use, 16 people (8.5 percent) had documented Safe Haven use during the ERP period. And while 3 people (1.6 percent) had a history of Journey of Hope participation prior to ERP, 13 (6.9 percent) participated in Journey of Hope during the ERP period. Finally, homeless outreach services were provided by an array of City and non-profit agencies. Results on Table 5d indicate that over half of the people on the BNL (57.1 percent) had encountered outreach services prior to the implementation of ERP, and that roughly the same proportion (54.5%) had at least one outreach contact during the ERP period. All in all, 69.3 percent of the people on the BNL had some documented contact with outreach over their life course, with most of these having contacts both in the five-month ERP period and at some point prior to ERP. Results from the Outreach Encampment Survey indicated that many people living in and around the encampments had experienced substantial episodes of homelessness and had already engaged various homeless services. The CARE data reflected reasonably high prevalence levels of shelter use (though substantially lower than the levels reported on the survey) and outreach services even before the ERP intervention, and high levels of all four services tracked here. This suggests that many people responded to efforts made under the auspices of the ERP to engage people in the encampments in homeless services. Increases in the numbers of placements to Safe Havens and Journey of Hope, while relatively small, are important as they both provide longer term residential arrangements for persons displaced from the encampments than the temporary treatment and shelter accommodations that were more widely available under the ERP. Substance Use Services. Results reported in the Outreach Encampment Survey indicated high levels of opioid use and near ubiquitous substance use among those surveyed who stayed in the encampments. Here we examine an alternative measure of substance use: the prevalence of substance dependency diagnoses based upon substance use services received. This measure is more indicative of the extent to 56 Part of the disparity between self-reported shelter use and shelter use as indicated from the OHS records may be due to the data gap in OHS records mentioned in a footnote to Table 5c. 57 The CARES shelter data used here covers approximately 85% of shelter beds citywide, consisting primarily of those shelters that receive support from the City of Philadelphia’s Office of Homeless Services. Time Period Shelter or Temporary Housing Safe Haven Journey of Hope Outreach Services Pre-ERP (before May 2018) 72 (38.1%) 9 (4.8%) 3 (1.6%) 108 (57.1%) ERP period (after May 2018) 98 (51.9%) 16 (8.5%) 13 (6.9%) 103 (54.5%) Lifetime 119 (63.0%) 22 (11.6%) 15 (7.9%) 131 (69.3%) 47 which substance use has led to receipt of publicly funded treatment services. We also report summary findings on the extent to which people with names on the BNL used publicly funded substance use services. Table 5e. Prevalence of Substance Use Diagnoses Received in Conjunction with Services Received by People on the BNL (n=189) Substance Abuse Diagnoses in Conjunction with Services Heroin/Opiates Dependency 115 (60.8%) Cocaine Dependency 28 (14.8%) Alcohol Dependency 35 (18.5%) Other SA Dependency 19 (10.1%) Number of Dependency Diagnoses per Person None 66 (34.9%) 1 74 (39.2%) 2 32 (16.9%) 3 9 (4.8%) 4 8 (4.2%) Any Abuse Diagnoses (excluding tobacco) 123 (65.1%) Table 5e shows that almost two-thirds of those on the BNL (65.1 percent) received at least one substance dependency diagnosis in conjunction with their use of CBH or DBHIDS substance use services.58 Almost all of those diagnosed with a substance dependency (i.e., the aforementioned 65.1 percent) had opioid dependency diagnoses (60.8 percent of total people on BNL). Just over one quarter of people on the BNL (25.9 percent) were diagnosed with multiple dependencies, and most of these involved one or more additional dependencies co-occurring with an opioid dependency. Almost two- thirds of this group (65.1 percent) had a substance abuse diagnosis, which indicates substance use that did not meet the criteria of substance dependency.59 These diagnosis prevalence rates, while high, are conservative indicators of problematic substance use, as only persons who received health care services covered by public reimbursement mechanisms in Philadelphia would be in a position to be diagnosed, and diagnoses were made only if patients met specific sets of rigorous criteria. These findings provide a more conservative set of parameters to complement results previously reported in the Outreach Encampment Survey that found 79 percent of respondents to be using opioids and 93 percent engaged in current substance use. Those who reported substance use on the survey do not necessarily meet the criteria of dependency, and thus almost certainly overstate the levels of problematic substance use among the BNL population. In addition to assessing the prevalence of substance dependency diagnoses, we also examined use of substance use services. Table 5f shows summary findings of four different types of such services that are reimbursed through DBHIDS and CBH. While data on overall use of substance services was unavailable, there is presumably a high degree of overlap between these four categories of substance abuse services. Noteworthy among these findings were: 58 To be considered to have a substance dependency diagnosis required a diagnosis to be present on either one inpatient or rehab/habilitation record, or at least two other records. 59 Abuse diagnoses are often made as secondary diagnoses to a dependency diagnosis for a different drug, but may also be given without a diagnosis when the criteria for dependency was not met. The equivalence in numbers of total people with dependency diagnoses and abuse diagnoses is coincidental and reflects overlapping subgroups. 48 - A majority of people (110, or 58.2 percent) received Rehab/Habilitation services60 at some point in their lifetime, with all but eight of these people first receiving a service in this category prior to the ERP implementation. Looking just at receipt of these services during the ERP period, at least 49 people (25.9 percent) spent, on average, 42.45 days in Rehab/Habilitation services. - A roughly similar number (114, or 60.3 percent) had a lifetime history of receiving some form of outpatient service.61 In a manner similar to the Rehab/Habilitation services, the vast majority (all but ten people) had first engaged outpatient services prior to ERP. Forty-eight people (25.4 percent) used, on average, 21.85 units of outpatient service during the ERP period. - Smaller proportions of people on the BNL had any lifetime record of substance abuse-related case management services (43.9 percent) and inpatient (i.e., hospital-based) treatment services (7.4 percent). Table 5f. Prevalence of Persons Receiving Substance Use Services and Frequency of Services Received Under Auspices of City of Philadelphia Addiction Services Agencies by People on the BNL (n=189) Mental Health Services. Prevalence of mental health diagnoses and the extent of services use is presented in a similar fashion as the substance use findings. The diagnoses and services provided are related to care for mental health disorders that do not include services whose primary focus is substance use treatment and recovery. Table 5g shows mental health diagnoses, by category, based upon specific mental health ICD9 and ICD10 diagnoses that appear on the mental health services records of people on the BNL. As with substance abuse diagnoses, a mental health diagnosis is counted when it is present on at least one inpatient or 2 outpatient claims. The rates diagnosed with severe mental illness,62 at roughly forty percent, are comparable to their prevalence in other single adult homeless populations. As best as can be determined, the prevalence rates for the BNL group on Table 5g are consistent with levels of mental health issues reported in the Outreach Encampment Survey. In the former, 44.4 percent had records with a mental health diagnosis; in comparison, 40 percent of the survey respondents reported that they “struggled with mental health challenges.” Similarly, while 37.6 percent had services records indicating a major depressive disorder, 43.5 percent of survey respondents cited depression as a “mental health challenge” they experienced. 60 Rehab/Habilitation services included detoxification and rehabilitation services that are provided in non-hospital settings. 61 Outpatient services included medication management (including methadone), ambulatory, counseling and therapeutic services related to substance use 62 Severe mental illness is reflected by diagnoses of major depressive disorder, bipolar disorder and manic depression, and schizophrenia. As an individual could have more than one of these diagnoses, the exact prevalence of major mental illness in this group could not be determined based upon the results received. Rehab/Habilitation Outpatient Case Management Inpatient Persons (% of total BNL) Pre-ERP (before May 2018) 102 (54.0%) 104 (55.0%) 60 (31.7%) 10 (5.3%) ERP period (after May 2018) 49 (25.9%) 48 (25.4%) 42 (22.2%) 5 (2.6%) Lifetime 110 (58.2%) 114 (60.3%) 83 (43.9%) 14 (7.4%) Contacts-Days – Total (contacts-days/people) ERP period (after May 2018) 2,080 (42.45) 1,049 (21.85) 262 (6.24) 38 (7.60) 49 Table 5g. Mental Health Diagnosis Groups in Conjunction with Services Received by People on the BNL (n=189) In a more exact comparison, 77 people on the BNL could be matched directly with their Outreach Encampment Survey responses. Here the levels of self-disclosed mental disorders were, in most cases, substantially higher than the levels of diagnosed disorders. Thus, while the numbers of schizophrenia diagnoses among these 77 are roughly similar (8 diagnosed and 11 reported), the numbers for major depressive disorder diagnosis compared to affirmative responses to experiencing “depression” is 20 and 33, respectively, and those diagnosed with anxiety compared to those reporting anxiety is 5 and 20, respectively. While self-report was based upon uncertain criteria and self-diagnosis, the administrative records typically give an undercount of the prevalence of disorders that also likely accounts for part of the disparities in these findings. Table 5h provides findings on the most frequently used categories of mental health services received by people on the BNL. While we could not unduplicate proportions of people on the BNL across categories, taken together, the majority of people on the BNL received at least one of these mental health services.63 Looking at acute services alone, just over a majority (52.4 percent) received inpatient services, and just under a majority (47.1 percent) received crisis or emergency services. Neither of these categories saw drastic changes in rates for receiving these services over the course of the ERP. In contrast, receipt of residential services, which include both mental health treatment and housing, increased substantially during the ERP period. Specifically, 29 of the 37 people who ever received residential services did so during the ERP. Table 5h. Prevalence of Persons Receiving Mental Health Services and Frequency of Services Received Under Auspices of City of Philadelphia Mental Health Services Agencies by People on the BNL (n=189) Philadelphia Jail Incarceration. Results in Table 5i show that almost two-thirds of those on the BNL (63.5 percent) had been incarcerated in a Philadelphia Department of Prisons facility (i.e., jail). Comparing incarceration rates for the year (27.5 percent) and the month (5.8 percent) immediately prior to the ERP 63 Other categories were examined and omitted due to marginal numbers of people on the BNL using these services. They included case management (20 people lifetime use); community integrated recovery centers (3 people lifetime use); employment (1 person lifetime use); rehabilitation and habilitation (5 persons lifetime use). Engagement services records were also used for the outreach records reported in homeless services, and are not included here to avoid confusion over duplicated services. Major Depressive Disorder 71 (37.6%) Bipolar Disorder/Manic Depression 24 (12.7%) Schizophrenia 24 (12.7%) PTSD 5 (2.6%) Adjustment Disorder (other than PTSD) 29 (15.3%) Anxiety 17 (9.0%) Other MH Disorder 11 (5.8%) Any MH Diagnosis 84 (44.4%) Emergency & Crisis Inpatient Outpatient Residential Persons (% of total BNL) Pre-ERP (before May 2018) 86 (45.5%) 93 (49.2%) 64 (33.9%) 17 (9.0%) ERP period (after May 2018) 17 (9.0%) 34 (18.0%) 14 (7.4%) 29 (15.3%) Lifetime 89 (47.1%) 99 (52.4%) 67 (35.4%) 37 (19.6%) Services Use – contacts or days (contacts/days per person using services) ERP period (after May 2018) 26 (1.53) 135 (3.97) 129 (9.21) 1651 (56.93) 50 launch in May (not on table) with the 20.1 percent rate during the ERP period gives no indication that incarceration rates dropped for this targeted group in conjunction with the ERP. At least 38 people had incarceration records during the ERP period, with only one of these 38 experiencing a Philadelphia jail incarceration for the first time during the ERP. Table 5i. Prevalence of Incarceration in the Philadelphia Department of Prisons (PDP) for People on the BNL (n=189) Beyond this, there is little information on the nature of these incarcerations, and incarceration history was not among the items on the Outreach Encampment Survey. Generally, the majority of jail incarcerations last a day or two, so many of these incarcerations were presumably short enough that the release dates also fell within the ERP period. Conversely, some of the incarcerations were likely long enough so that some people on the BNL were jailed for the remainder of the ERP period. There were no data on charges, warrants, or locations where people were taken into custody or other circumstances that were involved with these incarcerations. Child Welfare Services. The final CARES record match assessed the extent to which people on the BNL also were involved with the Philadelphia Department of Human Services (DHS) - either in a child welfare case (usually involving allegations of abuse or neglect) or in a situation where a child was involved with the juvenile justice system. Record matches were reported only in circumstances involving DHS involvement as adults, meaning that we do not report data on how many of the people on the BNL were included in DHS cases (e.g., in foster care, in neglect or abuse situations, as delinquents, etc.) as children. Furthermore, all persons on the BNL were adults and there were no noted instances of children or minors staying in any of the encampments. Table 5j. Prevalence of Involvement, as an Adult, in DHS by People on the BNL (n=189) Note. Records for the ERP period are not complete due to a delay in records transfer of as long as 90 days. As shown on Table 5j, over one-third of people on the BNL (37.0%) had a record, as an adult, of being in a household where some type of DHS service was received. The nature of such services was unclear and some proportion of these cases were presumably closed. However, as the remainder of the table indicates, as many as one quarter (the extent of duplication across the categories on the table is unclear) of the overall BNL group was involved with a DHS case that was substantial and ongoing. Most notably, 10 percent had ongoing cases in which they were the parent of a child in an out-of-home placement. A slightly higher proportion were parents of a child who was receiving in-home services in response to reported situations involving abuse or neglect. There was no further information available on the circumstances or the personal characteristics of those matched with records of DHS involvement; we cannot ascertain the gender of these parents, the ages of the children, the extent to which the parents are involved in the lives of the impacted children, or whether homelessness or opioid use preceded or followed DHS involvement. The proportions of DHS Pre-ERP (any time before May 2018) 119 (63.0%) One year prior to ERP (May 2017 through April 2018) 52 (27.5%) One month prior to ERP (April 2018) 11 (5.8%) ERP period (after May 2018) 38 (20.1%) Lifetime 120 (63.5%) Ever (as adult) 70 (37.0%) As Parent of a Child in Delinquency 8 (4.2%) As Parent of a Child in Dependency – In-home Services 21 (11.1%) As Parent of a Child in Dependency – Out-of-home Placement 19 (10.1%) As Parent of a Child Receiving Other DHS Services 3 (1.6%) 51 involvement would be higher if the denominator only included people on the BNL who were parents of minor children. Child welfare involvement was not among the items on the Outreach Encampment Survey. 5.1.3- ERP Outcomes In mid-August, a local radio station KYW, in a follow up story on the Kensington encampment closures, described how: Three months ago, SC was hopelessly addicted to heroin, living on the street, taking refuge in the tunnel encampments in Kensington. This week, sober and in shelter, he started a new job at a roofing company.64 SC’s experience represents a desirable trajectory for persons who accepted placement either into housing or treatment services as part of the ERP, insofar as his accepting services offered a means by which to initiate recovery, stable housing, and/or economic self-sufficiency. However, from the little information available in this story, SC was in temporary housing and it was unclear whether his recently obtained sobriety was supported by recovery services. SC’s situation speaks to broader questions on how people fared after they left the Kensington encampments and accepted housing and substance use services. Here we address this question of how the people who were sleeping in the encampments and targeted for services fared after the encampments were cleared on May 30. Ideally, the ERP provided a framework for persons on the BNL to access temporary housing and treatment services as an alternative to their encampment accommodations, and then provided them the opportunity to transition to more long-term housing and recovery arrangements. This analysis provides some basic data to gauge the extent to which that happened.65 In Chapter 4, we reported how 83 individuals on the BNL (75.5 percent) were placed in either housing or treatment services at some point during the 30-day implementation phase. Table 5k shows housing and treatment options at two points in time during the first month of the sustainment phase. On June 1, the first day of the sustainment phase, the number of people who remained in housing or treatment services settled to 61 (55.5 percent). This contraction of persons remaining in housing and treatment services continued until late June, when the BNL was expanded to 192 names (later unduplicated to 189) with the addition of people who had engaged with outreach workers and were receptive to receiving housing and treatment services. After adding these services-engaged people to the BNL, 64 Pat Loeb (8/19/18), “Kensington encampment eviction, 10 weeks later: Success or failure?” KYW newsradio (https://kywnewsradio.radio.com/articles/news/kensington-encampment-eviction-10-weeks-later-success-or- failure). 65 The findings are based upon data collected and provided by DBHIDS, including regular aggregated updates on services outcomes that DBHIDS provided; a deidentified copy of the BNL with individual services outcomes; results from specific data queries requested by the evaluation team; and qualitative information that provided context for these data. We use these data to follow progress from the day after the encampment closures on June 1 through October 15. We acknowledge Ben Lambertsen, Bridgette Tobler, Roberta Cancellier and Tim Sheahan for their assistance with compiling these data and assisting with their interpretation. 52 almost the same proportion of people were receiving housing and treatment services at the end of June (55.2 percent) as at the start of the month (55.5 percent).66 Table 5k. Engagement in Services for People on the BNL: June 1 and June 26, 2018 (n=192) June 1 June 26 Total on BNL 110 (100%) 192 (100%) Placements: PPP Navigation Center 35 (31.8%) 39 (20.3%) PPP Respite Center 8 (7.3%) 26 (13.5%) ODAAT Respite Center 2 (1.8%) 7 (3.6%) Drug Detox or Treatment Center 7 (6.4%) 19 (9.9%) Safe Haven Facility 6 (5.5%) 9 (4.7%) Recovery Housing 1 (0.9%) 1 (0.5%) Return Home 1 (0.9%) 1 (0.5%) VA Housing 1 (0.9%) 1 (0.5%) Salvation Army 0 1 (0.5%) No Placements 49 (44.5%) 86 (44.8%) Note. The 192 names listed in the June 26 report was subsequently unduplicated to 189 names. Table 5l. Summary of Placements for People on the BNL: October 15, 2018 (n=189) Subsequently, DBHIDS tracked more specific outcomes for persons on the BNL. These outcomes included the following categories, the results of which are summarized in Table 5l: - Long-term Placement or Permanent Housing. 36 people (19.0 percent) received placements that served as transitions from the immediate placements temporary housing and treatment services that were the centerpiece of the ERP intervention. Sixteen of these placements were to permanent housing—15 through Pathways to Housing and one placement through housing provided by the VA. Another nine persons were placed in Safe Haven facilities.67 On the recovery side, nine people were placed in Journey of Hope,68 and two others were placed in long-term recovery housing funded through the City of Philadelphia’s Office of Addiction Services. Although these accommodations are long-term, its residents are still considered homeless. 66 The proportional drop in Table 5k in navigation center placement is an artifact of the center’s capacity of 40 beds, and was offset by corresponding increases of those in respite centers. 67 In the CARES match, OHS records showed 16 people to have been placed in safe haven housing during the ERP time period, indicating that 7 left this housing before October. 68 In the CARES match, OHS records showed that 13 people were placed in Journey of Hope during the ERP time period, indicating that 4 left this service before October. October 15 Long-term placement or permanent housing 36 (19.0%) In time-limited placements 19 (10.1%) Unable to access placement 12 (6.3%) Contact but no ongoing placement 77 (40.7%) Unengaged 45 (23.8%) 53 - Time-limited Placements. On October 15, 19 people (10.0 percent) were in some sort of placement that did not resolve their homelessness. Of these, four were listed as remaining in either respites or the navigation center, despite the goal of transitioning everyone out. Some of these individuals may have been in the process of getting a limited-term rapid rehousing assistance69 that would permit them to move to other housing arrangements. The other fifteen people were receiving short-term substance use treatment services and presumably would receive assistance with living arrangements upon completion of the program. - Unable to Access Placement. On October 15, 12 people (6.3 percent) were either deceased (n=2) or known to be incarcerated (n=10). Further details on people in this category were unavailable beyond that one was staying in a respite center at the time of death. The CARES match identified 38 persons as being incarcerated in PDP facilities following the implementation of ERP (table 5i). Thus, the number who remained incarcerated as of October 15 may have been higher and might account for a few of those listed subsequently as “unengaged.” - Contact but No Ongoing Placement. On October 15, 77 people (40.7 percent) of those on the BNL engaged with some services but were not in housing or residential treatment arranged through the ERP. Of these, 37 people kept in contact with outreach workers, meaning that they continued to frequent the area near the encampments and formed some degree of relationship with outreach workers, but did not accept offers of temporary housing or treatment. Another 17 did accept offers of temporary housing or treatment and subsequently left these services without housing arrangements. The remaining 23 people in this group did not have contact with ERP-affiliated outreach workers, but did receive intensive case management services through either CBH or Behavioral Health Special initiative (BHSI), two City of Philadelphia entities that provide substance use services. To what extent these 23 people used these services was undetermined.70 As with those in the “unengaged” category, a substantial proportion likely remained homeless and substance using. Alternately, this group also maintained contact with services providers, and outreach coordinators pointed out that service engagement and relationships with outreach workers can provide the foundation for more meaningful engagement with services in the future. - Unengaged. 45 persons were listed as “whereabouts unknown since camp closing.” This means that 40.9 percent of the original 110 people (or 23.8 percent of the complete BNL) on the BNL did not engage in services that were provided by or linked to the ERP. Some, as mentioned earlier, may have experienced incarcerations that were unknown to ERP staff. Others in this category may have moved or made informal housing arrangements on their own, as substantial proportions of homeless populations experience homelessness relatively briefly before regaining some type of housing arrangement. Whether this holds true with people in this category is unknown, as Outreach Encampment Survey results (see previous chapter) showed 69 Rapid rehousing assistance most often came in the form of shallow rent subsidies, where people received assistance with a partial amount of monthly housing costs for a limited time duration, along with case management support services. Often such assistance facilitates households to either afford housing payments or negotiate shared living arrangements. 70 CBH services are described in greater detail elsewhere in this section; more information on BHSI and their case management services is available at: https://dbhids.org/about/organization/office-of-addiction-services/bhsi- intensive-case-management-services/ 54 that those staying in the encampments reported a median time spent living on the street of 6 to 9 months, with forty percent experiencing homelessness for over a year. Determining successful outcomes within each of these categories presents a challenge, as the terms for successful engagement were never formally laid out. Removing the 12 people from the BNL who were unable to access placements due to death or incarceration (leaving a total of 177), outcomes point to “success” for a number of individuals: - 36 (20.3 percent) received long-term housing or recovery services through the ERP; - 72 (40.7 percent) received at least temporary housing or treatment services after ERP engagement;71 and - 109 (61.6 percent) interacted with outreach workers to some degree and thereby had a direct opportunity to engage in services.72 This tracked very closely with the match between the BNL and OHS records, which indicated that 103 people had a record of an outreach contact during the time of the ERP sustainment phase. Outcomes differed between the 110 who were originally on the BNL and the 79 who were subsequently added. Those without any outreach worker contact (n=68) comprised 61.8 percent of the original subgroup and none of those added. All ten who were incarcerated at the end of the sustainment period came from the original BNL subgroup. Conversely, 11 of the 15 Pathways to Housing placements were from those added to the BNL, as were 7 of the 9 who were in Journey of Hope placements at the end of the ERP. Compared to those originally on the BNL, more among the 79 whose names were added to the BNL in June had desirable outcomes. OHS Director Liz Hersh73 confirms the lack of benchmarks for these outcomes, stating that, as a pilot, the “goal and design were to push forward offering the services we believed people wanted, be as person- centered as possible and learn from the experience.” Additionally, she emphasized how, in the context of the ERP, what determines a successful outcome is not necessarily the linear process that is implied by first engaging in services, then placing in substance use or temporary housing services, and then placing in long-term housing or recovery arrangements. Rather: People with opioid use disorders who are homeless rarely have a simple “Point A to Point B” linear path. They often come in and out of services, respite, treatment, and the street. This seems to be a pattern of this population around which service design must be tailored. Hersh characterized this as the “ebb and flow” nature of services engagement, and pointed to the number of people who maintained informal outreach contacts and relationships as indicative of those who may not produce short-term results but may nonetheless show housing and treatment outcomes over an extended time period. This assertion about such gains from extended engagement is supported by findings Hersh presented in early 2019, where 83 out of 189 (44 percent) people on the BNL had 71 This includes the 55 people in the top two categories of Table 5b, and the 17 from the fourth category who accepted offers of temporary housing or treatment but left before October 15. 72 This includes the aforementioned 72 people who received temporary housing or treatment services and an additional 37 people from the fourth category who interacted with outreach workers through the ERP program. 73 Based upon an interview with Liz Hersh, November 30, 2018, and subsequent correspondence. 55  received housing or treatment services;74 a number that grew from the 72 reported here as of mid- October 2018. Each of the three outcome categories considered here: long-term housing and/or recovery placements; temporary housing or treatment services; and ongoing outreach or case management engagements describe benefits none fully capture “success.” These findings form a potential basis for determining benchmarks for subsequent similar efforts as, despite their limitations, they reflect the most detailed follow-up of any initiative seeking to engage people with services as part of an encampment closure effort. Another problem with the numbers of persons in housing and/or treatment reported here is that some report a number on a particular day (i.e., a snapshot or point-in-time count) and others reflect the number over a more extended period of time, such as over the 30-day implementation period. We try to be clear what the timeframe is with the numbers we refer to, but this can lead to confusion. Finally, clear benchmarks for these outcomes are also not readily apparent. For example, how does placing 36 people, comprising 19.0 percent of the targeted population, into permanent or long-term housing arrangements square with outcomes that could realistically be expected from the ERP? On one hand, the population on the BNL, marked by homelessness and opioid use, is notoriously resistant to engaging and following through with available services. On the other hand, advocates assert that greater access to such resources as harm reduction-based permanent housing and recovery services using medication assisted therapies (resources that are beyond the direct scope of the ERP) would provide more amenable alternatives to continued homelessness. The outcomes reported in this section also strain the abilities of the current BNL efforts to track and monitor individual outcomes. These data limitations were mainly due to challenges involved with coordinating multiple sources of data and multiple services systems, and using an ad hoc data collection setup. Hersh acknowledges that issues with managing data from multiple systems is “a work in progress” and that data quality represents an ongoing concern. Services providers report that, based upon initial experiences reported here, procedures for compiling a BNL have already been improved and that they expected subsequent efforts to lead to more systematic data collection and more definitive reports of client outcomes. In the meantime, these limitations related to collecting, reporting and coordinating data led to inconsistencies between findings reported here and statements to the media concerning client outcomes made in conjunction with subsequent encampment closures.75 Taken together, the data structure and procedures informing the ERP have provided a framework for collecting outcomes, tracking collateral services use, and coordinating services that is more extensive than any other similar sized encampment clearance effort that we are aware of. This notwithstanding, it was still difficult under this data structure to consistently report outcomes and to put these outcomes into a readily understandable context. 74 These 2019 findings are from personal communication with Liz Hersh on February 8, 2019 and are stated in a City of Philadelphia press release (1/31/19) “City completes encampment pilot project in Kensington area” (https://www.phila.gov/2019-01-31-city-completes-encampment-pilot-project-in-kensington-area-2/). 75 See Joel Wolfram (11/15/18), “As Philly clears another encampment, what happened to homeless in previous sweep?” WHYY (https://whyy.org/articles/as-philly-clears-another-encampment-what-happened-to-homeless-in- previous-sweep/) and Aubrey Whelan (11/15/18), “As snow falls, Philly clears another Kensington heroin encampment.” Philadelphia Inquirer (http://www2.philly.com/philly/health/addiction/philly-clears-another- kensington-heroin-encampment-20181115.html). 56 5.2- Perspectives on Services: People Served and ERP Officials and Providers Perspectives on ERP services were collected through qualitative interviews with people served through the ERP and affiliated City representatives and service providers. The reflections in this section offer insight into ERP operations from those targeted for services and front-line staff involved in every aspect of the initiative. Anonymity was maintained to facilitate frank responses, yet many of the views and suggestions proffered by respondents were similar and can be used to direct continued and future efforts. 5.2.1- People Served Through ERP Perspectives from people served through the ERP were elicited during qualitative interviews with individuals staying in the PPP (n=9) and ODAAT respites (n=2) and utilizing the PPP drop-in (n=5), as well as direct observation by members of the evaluation team at these locations. Interviews were organized around the domains used to structure the Outreach Encampment Survey and previous interviews that were presented in Chapter 4. (See Appendix F for Interview Guide.) Interviewees receiving services provided through the ERP described conditions at the respites and navigation center as well as the impact of the ERP on their daily activities and service use and their perspectives on the encampment closures. While these accommodations were seen as an improvement over staying in encampments, and some did receive assistance connecting with needed services, many lived a life not dissimilar from how they had lived prior to the encampment closures. Living Situation. Individuals staying at the respites and navigation center who participated in interviews described a variety of rules that they were expected to follow. For those who had previously used City homeless shelters, the rules for staying were considerably looser though respondents also noted similarities: respites and the navigation center closed during the day, so they had to leave each morning and couldn’t return until the early evening; no fighting; and no drug use or drug paraphernalia was permitted (note: although people could bring their “works” into the navigation center, the expectation was that they be checked with staff). Overall, while individuals who participated in interviews reported that living under these conditions was not ideal, they recognized that it was better than staying on the streets. During operating hours, single individuals as well as couples spent their time sleeping, watching television, and sharing in communal meals provided by the facilities. For many, staying in the respites and navigation center provided a chance to shower and clean their clothing for the first time since leaving the encampments. Improvements in living conditions enabled some individuals to address their sobriety and mental health issues as well as reestablish connections, mostly of a professional nature. However, while there were residents committed to their sobriety and addressing their treatment and housing needs, there were also many who continued a lifestyle not dissimilar from their lives in the underpass encampments, often leaving during operating hours to purchase and use drugs. When asked about challenges to staying in these low-barrier shelters, some people expressed frustration about others who left for extended periods of time, which they felt created a backlog for others who would like to access housing to get off of the street. In addition, given the communal nature of these accommodations, unattended property was often stolen or misplaced, and people reported constant concern about the safety of their personal belongings. There was not enough space at these facilities for people to safely store their belongings. Among the individuals interviewed at the PPP drop-in center, several lived at the PPP respite, but most lived on the street, often in the remaining two underpass encampments, or in some other living arrangement. The lives of those who didn’t have a stable place to live were filled with more uncertainty than those who stayed at the respites or navigation center or had a place of residence elsewhere. In 57 contrast, those at the PPP drop-in who had received shelter through the ERP were more likely to be trying to get sober or seeking treatment. Typical Day. Strikingly, given the ubiquity of heroin use among the people staying in encampments interviewed prior to the encampment closures, fewer than half of those interviewed at the respites and navigation center reported time spent acquiring and injecting opioids. Of those who transitioned to the PPP navigation center, many made attempts to get and remain sober. For some, regular work or activities (e.g., going to the PPP drop-in to sleep and watch television, spending time at the public library) supported sobriety. Others made efforts to leave the neighborhood during the day for communities that didn’t offer as many temptations to start using drugs again. Despite noting frustrations over the need to leave the respites during the day, this was described as an advantage by those who tried to get on a more regular schedule of being awake during the day and asleep at night. One commonality among the daily activities of those who received services through the ERP was the need to earn money to take care of their needs. Many relied on practical skills like painting, flooring, cleaning, and other forms of construction-related labor. Otherwise, they relied on survival labor—or “hustling,” in the words of interviewees—including theft, acting as look-outs for drug dealers, sex work, selling condoms, and picking up used needles to exchange for clean ones that could be sold. Interestingly, the substance use of the individuals staying at the PPP and ODAAT facilities was markedly different. While most of the individuals interviewed through PPP had opiate use issues or were in recovery, those at ODAAT did not report heroin use, but rather engaged in smoking crack-cocaine or K2 or abusing alcohol. However, their daily behavioral patterns were not dissimilar; the differences primarily lay in their substance of choice and how pressing their need for substances were. For those who continued to use, a typical day was not much different from when they stayed in the encampments; their daily schedule continued to revolve around getting money to purchase drugs and doing drugs. What had noticeably changed for substance users staying in the housing offered through the ERP was where they slept during the day and where they used drugs. Fewer people reported injecting drugs in the open, opting for alleys (i.e., “duck spots”) or other clandestine places to decrease their exposure to law enforcement, who were seen as more likely to intervene compared to when individuals lived in the encampments. In this sense, whatever protective factors that helped those who injected drugs whilst in the encampments were no longer as robust. Among those interviewed at the PPP drop-in, many were actively using opiates multiple times per day, going to the drop-in for clean needles, to watch television, or to rest since they often couldn’t sleep at night. For them, their day revolved around getting money to purchase drugs, finding someone to sell them drugs, using, and after their high has worn off, moving on with their day before repeating the pattern at night. Though these individuals were vulnerable in many ways, the PPP drop-in offered a sense of safety since staff kept watch to ensure that no harm came to them. Service Use. Those who engaged in services through PPP and ODAAT found them useful, and social workers were overwhelmingly described as helpful and engaging. Those receiving services through the ERP and others reported receiving a variety of assistance, including: help identifying housing options; gathering paperwork for identification and establishing residency; accessing substance abuse, mental health, and/or medical treatment, as well as help getting health insurance; and linkages with other social service agencies that were able to meet the diverse needs of this population. However, the need for social services outpaced the availability of assistance. Available assistance also varied by location and service provider. For example, social workers at the PPP drop-in provided more 58 linkages to opiate abuse resources than individuals at ODAAT, which was likely reflective of the location and needs of those staying there as well as the service use philosophy of the provider (note: while it wasn’t expressly forbidden for individuals staying at ODAAT to be enrolled in MAT, it was not encouraged). Moreover, regardless of their living situation, individuals who received day-time services at PPP generally noted that staff were overwhelmed by the sheer volume of those seeking assistance. People engaged through the ERP also pointed out multiple, ongoing barriers to treatment and services despite the City’s efforts to alleviate them. Photo identification continued to be a challenge. This was further complicated by a variety of factors: lacking other forms of identification (e.g. a social security card, passport, birth certificate); inability to prove Pennsylvania residency; and being born in and needing documents from a different state. Residency issues continued to plague individuals seeking services, and while some churches and religious organizations allowed individuals to use their places of worship as their place of residence, this was not always a viable option for people living unstable lives. Transportation to treatment was also expressed as a barrier for those without income or means to pay for travel fare. This was particularly problematic for individuals on MAT, which required a daily commute to where the medications were dispensed. Finally, a handful of individuals were removed from the PPP respite due to rule infractions, which returned them to a state of street homelessness. In addition to logistical concerns, there were a variety of behavioral patterns among those served through the ERP and others that prohibited them from taking full advantage of available services. At times, this manifested itself in conflict with staff. At other times, it led to missed appointments and a lack of follow-through. The unpredictable lifestyle associated with unstable housing and addiction was a challenge to service engagement as the cycle in which these individuals lived prevented them taking steps towards achieving their goals. Lifetime experiences of trauma also plagued many of those interviewed. One woman, who was sleeping on the streets, stated that getting substance abuse and mental health treatment was a source of emotional stress given that it would put her back in contact with a past that she was trying to avoid by using substances to numb herself to the trauma. In addition, though people served through respite and navigation center housing were encouraged to access case management services during the day, they often did seek assistance; they suggested that if services were offered on-site in the evening when they settled in for the evening, they would be more likely to take advantage of them. People served through the ERP also suggested that more housing and social services are needed to address the issues and needs that are rampant in Kensington, particularly the need for temporary housing. Interviewees suggested that, should more respite beds be made available, the beds should be located in Kensington. Though ODAAT continued to have openings throughout our evaluation, many people struggling with substance use disorder and housing instability were reticent to go there because of the respite’s distance to the opiate market. They feared being too far from the market to purchase drugs before “dope sickness” set in. The discomfort and pain felt by those suffering with addiction trumped a warm, safe bed with three square meals in a different part of the city. According to one, if the goal of the City is to get people into housing, the housing should be where the people are, not in another place. At the same time, they suggested that offering long-term housing in different parts of the city may be helpful for those in recovery who may be triggered by staying in or around Kensington. Perspectives on Encampment Closure. When asked about their perspectives on the closure of the Kensington and Tulip Street encampments, people served through the ERP expressed that the initiative could have been better thought out. One noted that by closing the Kensington and Tulip encampments, the problems inherent to the camps spread to different parts of Kensington, while another suggested that the problems experienced in Kensington in the wake of the closures was that housing options were 59 insufficient to assist all those in need. This issue continued once people were housed temporarily through PPP and ODAAT with those staying at ERP-associated facilities lamenting their inability to move on to other housing. They also felt this further contributed to the ongoing backlog of people in need of shelter. Interviewees reflected that many of the problems experienced following the initial encampment closures will reoccur when the City closes the two remaining encampments. Many of those staying at the respites and navigation centers as well as those who received services at the PPP drop-in said that as long as there is insufficient housing, poor social services coordination, and poor resource provision for those in need, Kensington’s problems with homelessness, substance abuse, and other social ills will continue to be magnified. 5.2.2- ERP Officials and Providers Perspectives from City officials and service providers were collected during qualitative interviews with representatives affiliated with the ERP (n=13), including OHS, DBH, PPP, ODAAT, and NET, as well as attendance at ERP and community meetings. During interviews, representatives were asked about their background and experiences related to the Kensington encampments; perspectives on available services, including barriers and facilitators to service use; the closure of the targeted encampments and suggestion for improvement; and ideal outcomes for the initiative and needed resources. (See Appendix H for Interview Guide.) City officials and service providers reflected on the efforts made to address services offered during the ERP and suggested other opportunities for service improvement that they identified throughout the course of the ERP. Engagement Issues. Overall, better engagement of people staying in encampments was emphasized as a need moving forward. Poor attendance and interaction of people staying in encampments during ERP “town halls” was seen as a missed opportunity; future efforts could include providing chairs, tables, or even a bullhorn within the encampments to reach residents. Incentives of care packages with soap, socks, basic toiletries, etc., was also suggested as a means of connecting with residents and building rapport. At the same time, those with prior experience offering incentives for participation stressed the need to ensure incentives reached the target population of the initiative and services; community residents interested in incentives could serve to distract from efforts and end up taking resources away from those they are intended to serve. Transportation services were heavily utilized during the ERP. Outreach teams appreciated having an on- call van to transport individuals interested in treatment to the NET for evaluation. Additionally, weekly transportation to the Pennsylvania Department of Transportation (PennDOT) offices to assist people in getting their IDs was described as a useful engagement tool, allowing staff to meet individuals where they are and begin to build a trusting relationship, while also positively impacting their ability to access treatment, food stamps, and employment opportunities. Not only did staff feel this option should continue, they suggested that it could be improved by offering care packages at PennDOT, where accessing services can often take time. It was suggested that some people may begin to suffer from withdrawal systems while waiting, and offering water and/or food could help ensure their comfort. Staff suggested that future efforts should include protocols and expectations for tent removal. Individuals who left the encampments to enter respites and the navigation center often left their tents behind and asked that they not be taken down so that they could return to them during the day when the respites were closed or leave them for someone else. Outreach staff struggled with how to handle tent removal in these situations, as they complicated the engagement dynamic while also making closing the camps more difficult. Because of this, suggestions were made that if someone accesses housing or treatment, that individual should not be able to leave their tent outside and should instead 60 be offered a storage solution for future retrieval, and outreach workers should be trained on engaging residents on this issue. Finally, outreach workers reported that they don’t have the time or capacity to serve the entire area. The perception of some outreach workers was that following the closure of the underpass encampments, individuals moved to more hidden areas in the woods or abandoned houses, making identification and engagement more difficult. People staying in isolated locations are at a higher risk of overdosing and not receiving help as those nearby often lack phones or may be afraid to contact police because they fear being kicked off of private property. Respite and Navigation Housing. The respite and navigation beds made available through the ERP offered benefits in quickly serving those on the BNL, but these housing options also posed challenges for providers. Offering housing with low barriers to entry was seen as crucial for getting people to accept options, though explaining the various options and differences between housing choices to people staying in encampments was seen as difficult at times. Overall, housing without curfews that allowed people to come and go was attractive to people staying in encampments and those entering respite and navigation housing from the underpass encampments benefited from existing relationships when multiple people from the camps accessed housing at the same time. However, service providers acknowledged the difficulty of attempting to house so many individuals simultaneously. Given that respite housing had been in operation prior to the camp closures, it enabled staff to both prioritize people from the encampments and engage only a few people at a time. When the navigation center opened, it offered more of a logistical challenge when staff tried to house 40 people at a new location all at once. While additional housing was helpful, some City representatives and service providers didn’t feel that the resources brought online for the ERP were sufficient. In particular, more options were needed for partners, females, and people with pets. Respites allowed partners, but other providers who offered housing for couples were outside of Kensington and considered too far away by people staying in encampments, limiting options. In addition, though the ODAAT shelter had available beds in the month following the closure announcement, its location further from drug sources similarly remained a challenge in getting people to accept these beds, leaving some to question why more shelters weren’t opened where people would use them. Conversely, one service provider pointed out that changing neighborhoods can be seen as a commitment to change for those interested in moving past their experiences with homelessness and addiction and appreciated having housing options above the fray. Continued engagement also posed a challenge. Some individuals accessed housing early in the closure process, but it proved difficult to get them to stay, leading to a fair amount of turnover, with more people coming in and staying in towards the end of the month following the closure announcement. During this period, staff made it a point to be clear about their discharge policy; people could not “reserve” a bed in advance and then not use it only to return at the end of the month when the camps were closed. Typically, if people did not come to the shelters for 72 hours, they were discharged; exceptions were made for those who were hospitalized, in detox, or incarcerated. In these cases, staff tried to wait to find out how long the individual would be away and attempted to be flexible to help them keep their bed. If the person wouldn’t return for 2 weeks, they were usually be discharged, but staff indicated that they would work reengage them when their circumstances changed. Resources available on-site differed by housing option. All provided clean facilities, meals, and often necessities like clothing and toiletries. The PPP and ODAAT respites also offered programming that could be accessed during the day, including substance abuse meetings and programs, wellness services, 61 spiritual services, and case management. Individuals staying at the navigation center were encouraged to access day services at the PPP respite, three blocks away, which started offering extended daytime hours at about the same time as the ERP closure announcement. However, offering 24-hour facilities was suggested to provide a safe place for people during the day and improve engagement with programming. The need for housing to be better equipped with medical staff and capabilities to respond to the needs of this population was also reported. Though individuals staying in respite and navigation housing were offered opportunities for engagement, staff stressed the need for more integrated case management services. Stays in respites and the navigation center were not initially considered time-limited, and few people moved on to more permanent housing in the months following the encampment closures. City representatives and service providers underscored the need to free up space at existing facilities by helping those staying at the respites and navigation center access needed resources and create long-term housing plans. Long-Term Housing. City representatives and service providers identified an urgent need for more long- term housing options for those on the BNL for several reasons. First, outreach reported difficulty getting people to engage in services when they were unsure what would happen once they left the street (e.g., get own apartment); this sense of uncertainty is a barrier to engagement. Second, staying in respite and navigation housing can make recovery difficult. Though some people may try to stay sober in Kensington, others want or need to access housing outside of the neighborhood, away from known drug dealers and triggers. Third, housing resources were needed to create flow from respite and navigation center beds to more permanent housing options. Finally, while resources were available—Pathways to Housing, Journey of Hope, Safe Havens, rapid rehousing, shallow rental subsidies—these resources were insufficient and individuals often face challenges accessing those that are available. There were also limited options for individuals who weren’t interested in recovery-oriented housing with sobriety requirements. Even for those who wanted to be sober, some did not want their housing stability to be dependent on sobriety. These stakeholders emphasized the need to increase access to Housing First programs promoting housing as a human right that should not be tied to engagement in treatment. Substance Use Treatment. Service providers pointed out that addiction is a reoccurring illness and relapse is common. Most people have multiple treatment episodes, which is not necessarily an indictment of the treatment system, but rather an indication of their need for care. Given this, an objective of the ERP was to increase access to care for those on the BNL to be more responsive to the opioid crisis in Kensington and led the system to reevaluate its identification requirements and processes for denials and pre-authorizations for certain levels of care. The ERP offered a way to test policy changes to eliminate barriers to treatment access, and virtually all of the changes that were made as part of the initiative quickly became adopted across the services network. Despite this, service providers described a continuing need for leniency on identification (or possibly use of an alternate ID system) and more education for providers who consistently still deny patients without ID. They envisioned a day when pre-authorizations weren’t needed for any level of care, allowing providers to accept patients directly from the street. Providers offered several additional suggestions to improve access, including: offering a peer support specialist at the NET to support people awaiting assessment and treatment; extensions to periods of stay; and increased access to medically assisted detox, withdrawal management, followed by treatment.76 Additionally, some opined that it 76 The common perception of there being a lack of available treatment slots voiced here contrasts with reports of ample supply. See Aubrey Whelan (1/18/2019). “As Philly pushes for more medication-assisted drug treatment, 62 might be helpful for individuals seeking treatment to “be away from all this,” meaning the neighborhoods where they know where to get drugs, making it easier to quickly leave a program to purchase and use drugs. Overall, providers noted general improvements to the treatment system over time and increased use of evidence-based best practices, while stressing the need for more. They reported that standards of care in treatment are undergoing a period of transformation with growing recognition that the standards that have been in use for decades don’t reflect the needs of the population immersed in the current opioid crisis. Providers are improving continuity with MAT while people move through levels of care (e.g., inpatient to ongoing outpatient). However, while MAT has become the gold standard, some providers nonetheless felt that MAT was not conducive to communal living (i.e., creates a dynamic where the physical appearance of those using MAT is similar to drug abuse) and may pose challenges to sobriety. In addition to reflecting on efforts to address services offered during the ERP, City officials and service providers suggested opportunities for service improvements. Specifically, they offered perspectives on integrating data, increasing collaboration, and expanding the services continuum, and offered suggestions for other creative alternatives. Data Collection. Creating the BNL and tracking the individuals from the Kensington and Tulip Street encampments represented a huge amount of effort from all those involved in the ERP. Outreach workers use WebFocus. Though WebFocus is connected to the City’s HMIS system, it can be difficult to track outreach contacts unless people provide their name and date of birth (note: tracking of contacts is necessary to verify homelessness). However, the program is not formatted for cell phones and the required information is extensive and the format is difficult to use on-the-go, so teams often entered data at a later time. Similarly, though staff at respites and the navigation center were tasked with completing coordinated entry forms for all of the individuals entering housing, this task was difficult to complete when so many people entered housing at the same time, leaving staff to play catch up. This was further complicated by staff scheduling and limited operating hours. However, while some people were connected to services through coordinated entry, a requirement for some PSH and transitional housing programs, PPP can also make direct referrals for many programs. Services Collaboration. The ERP benefited from the cross-collaboration of multiple entities, but City representatives and service providers recommended further collaboration in the future. Service providers pointed out that the issues at play in Kensington denote a public health issue, suggesting that the Office of Emergency Management and the Department of Health provide robust teams to assist continuing outreach and enforcement efforts. In addition, outreach staff identified opportunities for increased collaboration with local hospitals. They reported that some people staying in encampments expressed interest in treatment, but had medical issues that took priority over their substance use issues (e.g., abscesses) and were taken to the hospital. Unfortunately, a number of these individuals were lost to follow-up when emergency department and hospital social workers unfamiliar with the ERP discharged them to the street. Despite a willingness to engage in treatment, these individuals often ended back at the encampments, with an increased risk for an overdose given their recent hospitalization. Service providers noted an opportunity to do more networking with hospitals to coordinate discharges. thousands of treatment slots are still empty.” Philadelphia Inquirer (https://www.philly.com/health/philadelphia- opioids-medication-assisted-treatment-capacity-overdose-crisis-20190118.html). 63 Other areas for increased collaboration described by City representatives and service providers included: working with additional assessment agencies, rather than relying on one NET, and improving coordination of care for people from outside of Philadelphia county. People from the community also promoted the idea of partnerships with support social service organizations (e.g., Project Safe, Sex Workers Outreach Project, Women Against Abuse, Women Organized Against Rape) and respite centers for sex workers and victims of domestic violence who require service coordination or even simply need a safe place to stay. Expanded Services Continuum. For individuals entering treatment from the Kensington and Tulip Street encampments, the ideal service model at present flows from the NET for assessment to treatment (generally at Kirkbride) to DBH case management and, finally, Journey of Hope. This “marriage of services” offers stability while people work on their recovery. Yet, while the goal is for people to access treatment and then move onto recovery-oriented housing, outcomes have been variable. Some have left treatment against medical advice and been lost to the system. Others have gone back and forth to treatment multiple times. Some have moved into recovery houses or opted for long-term treatment through programs including Journey of Hope and others. Safe Havens have also been useful following treatment; however, most Safe Havens offer wet and dry beds which may jeopardize sobriety. Overall, providers feared that, without adequate housing and support options in place, people would return to the street following treatment. Certainly, providers recognized the complexity of factors (e.g., mental health, social capacity, employment) that can promote or impede individuals’ recovery. One provider urged programs to consider the “human aspect” of this crisis and follow a recovery model that attends to aftercare planning and support for those in treatment. As this provider described, when people have been living in chaos with their addiction and then have a period of abstinence where they can finally begin to think logically, they need support afterward instead of being thrown back into chaos. Providers felt that current treatment is too often disconnected from life skills and housing related issues, all but ensuring that some people still have nothing when they exit. While some treatment providers offer family reunification, vocational rehab, and education services, among others, there is variability in the inclusion of other modalities of care. Most often, people are linked with services following 28/30 days of treatment, but there is an opportunity to offer more services while in treatment. Providers recommend a more holistic approach to support successful housing and recovery outcomes, one where housing, treatment, and supportive services “talk to each other” to address the multiple needs of these individuals since treating one aspect won’t resolve issues within the other areas. City representatives suggested that housing, treatment, and long-term supports should be better integrated and optimized to meet people with where they are at in the process (i.e., treatment to housing or vice versa). One dreamed of an integrated DBH and CBH department that could combine housing and treatment options to help people live independently and be successful in their recovery. Creative Alternatives. City representatives and service providers with expansive knowledge of the issues as well as the local Kensington context supported the ERP and suggested that the City consider other creative strategies in response to the homelessness and opioid crises. One proposed that, rather than trying to make people fit into the existing model, the City give people as many options as possible since different things will work for different people. Alternate strategies included: - OPSs were overwhelming supported. Most believed a OPS would reduce fatal overdoses, disease, and hazardous waste; ensure people use clean supplies; and serve as an access point for people who regularly use heroin to engage with and get to know service providers, so they know who they can connect with if/when they decide to seek services or treatment. They also 64 felt that a SIS could help reduce the stigma that can be a barrier to service use. However, one provider did express concern that a OPS would just serve as a “patch job” without addressing deeper issues of homelessness and addiction. - Mobile treatment options were also heavily favored. A mobile suboxone pilot program started in Kensington at about the same time of the closure announcement, and providers agreed that “as much as we can meet people where they are, it’s fantastic” and helps reduce the stigma of seeking treatment as well as the logistics of travel. - Prevention services were also touted as paramount to combating the epidemic moving forward. Providers fear that the cycle will continue unless there are better services to eliminate poverty, provide mental health support, improve neighborhoods, and educate children and the public about addiction. - A sanctioned encampment was proposed by service providers who were concerned about people living outdoors in isolated areas where they were more at-risk of experiencing overdose or assault. They felt a sanctioned camp could provide bathrooms and the opportunity to promote harm reduction while also providing access to water, food, and electricity. - An incentivized housing strategy was posited by one City representative interested in considering creative housing options such as making use of Philadelphia’s unused housing stock (e.g., provide home ownership opportunities for people to maintain vacant properties over 10 years). 5.3- Impact of the Encampment Closures on the Surrounding Area A second general set of outcomes concern the immediate ecological impacts of the Kensington Avenue and Tulip Street encampment closures. Even before these encampments were cleared, there were concerns about the aftermath. Outreach services facilitated placing 126 persons from the encampments into some type of housing placement or treatment service during the implementation phase. However, others took up the vacated spots to where the numbers of persons staying at the two encampments on a given night remained steady at between 90 to 100. This led to questions about whether the sites of the erstwhile encampments would remain clear, and what effects the dispersal of those formerly staying in the two encampments would have on the area around the encampment sites. There were no specific, officially sanctioned outcomes measures for this area beyond keeping the two underpasses clear of any further camping activity. However, one underlying rationale for the ERP was that the removal of the encampments would improve the quality of life for those living in conventional housing or doing business in the area. Quality of life improvements could be measured subjectively, such as when residents perceive there to be less discarded syringes strewn around the area, or objectively, such as by the number of persons in the area who continued to sleep in unsheltered settings. Both subjective and objective assessments are important, as they do not necessarily align and both contribute to assessing the success of the ERP and the feasibility of subsequently replicating this pilot. In this section we report a range of both subjective and objective outcomes related to possible neighborhood impacts of the encampment closures during the four-month period after May 30, 2018 and based upon several data sources. The perceptions of community impact (i.e., the perceived outcomes) are based upon observations of community meetings and interviews with stakeholders who largely comprise of area residents, persons experiencing homelessness and persons advocating for those living in unsheltered circumstances. More quantifiable outcomes are based upon data collected by 65 police on the number of unsheltered persons experiencing homelessness counted throughout the police’s East Division; a limited range of arrest data; and emergency response incidents that involved overdoses. 5.3.1- Perceived Impact As reported in Chapter 4, community members (i.e., residents and business owners in the area directly impacted by the presence of the underpass encampments and the widespread drug-related activity in the area), advocates (i.e., people who saw themselves as taking up the concerns of those staying in the encampments), and individuals experiencing homelessness in Kensington often found themselves on different sides of the issues related to the encampment closures. Ongoing issues between these constituencies continued to play out following the closing of the Kensington and Tulip underpass encampments. Many community members were supportive of the ERP process and initially expressed satisfaction with the encampment closures in that they felt that it became safer for children to walk to school and ridded the neighborhood of an eyesore. However, they felt differently as the summer progressed and problems related to homelessness seemed to increase and occurred in broader swaths of the neighborhood, which escalated tensions expressed by community members. Encounters between the housed residents and the unsheltered homeless were regular and contentious. Ensuing complaints manifested themselves in public health and law enforcement concerns. Some residents noted that, however disagreeable, the encampments functioned to contain many of the issues that became increasingly problematic in the neighborhood. Individuals who continued to experience homelessness felt like they were caught between competing forces. Community members who did not live in the encampments would hurl insults and, on occasion, would throw projectiles or shoot pellet guns from moving cars, similar to what occurred while they were in the encampments. Members of law enforcement would keep them on the move to prevent them from establishing a semi-permanent presence on sidewalks and on private property. In addition, people who remained on the street continued to experience violence from their unsheltered peers. They highlighted these continued stressors while continuing to deal with unsheltered conditions and addiction. Specific issues that came up most often were drug use, with unconcealed drug-related behaviors (injection, drug purchasing, etc.) and omnipresent needle litter; and sanitation, where public defecation and piles of human waste underscored the absence of adequate toilet and washing facilities. Other issues that were most consistently raised included panhandling, sex work, local transit authority (Southeastern Pennsylvania Transportation Authority [SEPTA]) security, and law enforcement. These areas, which acted as flashpoints in the generally contentious relationship between housed and unsheltered contingents in the areas around the erstwhile encampments, will be addressed here in further detail. Sanitation. Many of the sanitation issues that were present in the underpass encampments continued with the makeshift sleeping arrangements that proliferated in the wake of the closures. In response, community members complained about having to clean up human waste, used food containers, old clothes and other trash and refuse. Where city sanitation crews did regular cleanups of the encampments, there were few concerted cleanup efforts outside of the encampment and community members were left to clean their properties and surrounding areas themselves. Sanitation problems, as 66 they became more acute, raised fears of a Hepatitis A outbreak like one that emanated from a San Diego homeless camp earlier in 2018.77 Homeless individuals recognized the problems of spreading waste in their communities and expressed a desire to have portable toilets placed in multiple locations along Kensington Avenue, as well as more regular trash pick-ups by city sanitation workers. Some community members also offered concrete suggestions for responses to sanitation issues, including: supplying portable public toilets; increasing trash pick-ups; offering mobile showering facilities; and opening community centers that provide showers and laundry services, which could further serve as a point of linkage for social services. However, other community members also expressed opposition to the idea of restroom facilities in the neighborhood, fearing that it would encourage people to remain on the street rather than moving on to other housing options. Advocates also stressed the importance of having a place for people experiencing homelessness to properly dispose of waste, and went farther and underscored the need for individuals to have a place to shower, eat, and sleep—offering them an opportunity to “feel human” as a prerequisite to considering lifestyle changes. According to City officials, the Streets Department increased the number of trash pick-ups and other sanitation efforts along Kensington Avenue. Moreover, they have promoted efforts by non-profit, private organizations, like the Kensington Community Food Program and AiM, that wish to work with existing organizations that provide food to those in need, following recognition that the uncoordinated distribution of goods to those in the encampments made sanitation problems worse, leading to increases in waste production and the amount of vermin in the area. While the extent to which these efforts made noticeable overall differences in sanitation issues was unclear, sanitation efforts continued to be part of the ERP initiative. Drug-related Litter. The seemingly ubiquitous proliferation of discarded used syringes and other related items (e.g. used cookers, used cotton, needle caps) was another major community concern. Community members described how these items have impacted their ability to move within their neighborhoods, causing them to avoid public spaces and warn their children not to pick things up off the ground, and leading to challenging commutes on SEPTA. They reported that clean-up efforts quickly go unnoticed, and community members don’t feel their concerns are sufficiently addressed by the City. Advocates noted that those living on the street empathize with the community. To address this issue, one advocate group proposed hiring members of the community, regardless of experience or circumstance, to collect needles and properly dispose of them. They also suggested placing sharps containers on telephone poles in places where children could not reach them. Given the prevalence of this issue, PPP altered its needle exchange policy to a one-to-one exchange rate. Additionally, law enforcement, in concert with SEPTA, has increased its surveillance of station stops in an effort to arrest those selling drug paraphernalia; however, possession of needles, by itself, is not a crime, meaning that they must observe an individual in the act of injecting him/herself with a needle that contains a substance in order to issue a citation for possession of a controlled substance. In a 77 See Joel Wolfram, “Philadelphia wary of Hepatitis A outbreak as Kensington homeless population grows,” WHYY, October 1, 2018, https://whyy.org/articles/philadelphia-wary-of-hepatitis-a-outbreak-as-kensington-homeless- population-grows/ 67 related suggestion, one community member stated there should be financial incentives to those willing to collect used needles and dispose of them in appropriate containers. Panhandling. Community members frequently mentioned an increase in panhandling, to where it occurs outside of people’s homes. They expressed that, since the initial encampment closures, panhandlers had acted more aggressively, pressing people to provide anything of value that can be of use or sold. Business owners lamented panhandlers near their businesses and said it drove customers away. Several community members described how panhandling has changed how they and their children go about their lives outside of their homes. Advocates attributed the increases in panhandling to the poverty of those living on the street. While individuals experiencing homelessness said they were not proud of nor did they wish to engage in such behavior, they felt panhandling was one of their few options for getting money. Moreover, many, who themselves panhandle, expressed frustration with those who get overly aggressive with community members because they believed it further contributes to the discrimination encountered by those living on the street. Law enforcement acknowledged panhandling is a protected form of free speech. However, should panhandling increase in aggression to the point of harassment, they suggested community members report it to 311. Reports of harassment could then justify additional resources to respond to this problem. Sex Work. Following the closure of the Kensington and Tulip encampments, community members noted the continued presence of those working in the street-based sex industry. While they recognized that sex work and trafficking has been taking place in Kensington for decades, there had been a recent uptick that they linked to the encampments. Residents encountered sex workers while taking care of daily errands and taking their children to school. Business owners have had to deal with sex workers soliciting their customers. To address sex crimes, the number of vice officers working in Kensington increased, and, again, law enforcement stressed to community members the importance of reporting incidence to 311 to justify increased resources to address the problem. Those who work in the street-based sex industry reported dangerous conditions related to soliciting strangers, often in cars. Such problems have been longstanding and, they feel, largely unaddressed to where sex workers feel they are largely on their own to protect themselves. At least four respondents reported interactions with law enforcement that, while unsubstantiated, they considered to be harassment and coercion. SEPTA Security. The two SEPTA stations in the area along Kensington Avenue are at two intersections, Allegheny Avenue and Somerset Street, and have been notorious for drug activity. Community members reported that taking SEPTA has become more stressful as more and more people took up sleeping outside of these elevated train stops. They also expressed concern over the sale of needles and other drug paraphernalia at these stations. The City and PPD reported working closely with SEPTA to address these challenges. SEPTA has installed additional security cameras and increased the number of transit officers in these stations. They have also tried to ask people who are not actively using the transit lines to move on from the stations, in hopes of discouraging individuals from establishing themselves at a particular station stop. Conversely, advocates have criticized the ways law enforcement interacts with those staying at or close to SEPTA stations. According to advocates, SEPTA officers speak harshly to many of the homeless and at times would remove them from the stops, while people experiencing homelessness typically seek to be treated with decency and respect. 68 Law Enforcement. Police officially viewed the co-occurring homelessness and opioid problem in Kensington as a public health problem that the City could not “arrest our way out of.”78 This police response tacitly acknowledged that arresting and citing people for minor homeless and drug possession charges makes little difference in the situation around the camps. Stephen Clark, Captain of the 24th Police District (in which the encampments were located), described in an interview79 that opioid use in Kensington was a public health crisis “like we’ve never seen.” In assessing the situation, he parsed situations that police handle in the area around the encampments as mostly (with exceptions) quality of life issues, which are geographically apart from where most of the drug dealing takes place, and thus where most of the violence and other more serious situations occur. Clark placed a higher priority on directing limited police resources on the latter issues, and prefers to have other entities, such as outreach services, take the lead in handling the quality of life issues. Despite this, the 24th District has police teams trained to address homeless individuals and their needs, who will often respond to situations around the encampments and provide support to social service and medical efforts in this area. Clark, in an interview, maintained that the police did not arrest or cite anyone on homeless related offenses, although he acknowledged that the threat of citation or arrest was a tool for dispersing makeshift outdoor sleeping arrangements. Other PPD officials agreed that PPD cannot “arrest their way out” of the problem, in no small part because the sheer number of individuals who purchase, use, and sell drugs would overwhelm the system. They echoed Clark’s view that the scale of the problems is unprecedented and unconventional, and calls for more untraditional approaches, such as working in concert with housing, treatment, and public health providers. In contrast, some service providers expressed opinions that things have gotten to the level that they have in Kensington, at least in part, because of lack of police enforcement. In this view, while it is illegal to sleep on the sidewalk, openly use drugs, or participate in sex work, the PPD fails to respond to such offenses. Current law enforcement responses of waking people up who are sleeping on the street and telling them to move on and ensuring that tents and mattresses are not set up on the street are insufficient for the magnitude of the problem. Though they recognized that strict enforcement could lead to increased violence in the community, they felt that those working in Kensington should not have to feel concerned for their own safety while working to engage people staying in encampments. One felt that law enforcement should be stern, yet fair, with both users and sellers of opiates and more willing to make arrests and/or issue citations. Another suggested that while we can’t “arrest our way out” of the situation, we cannot continue to let people to stay on the street, suggesting that consequences could include a treatment court strategy that is minimally punitive and encourages treatment. Community members often expressed frustration at this policing approach. A common sentiment among area residents was the desire to remove homeless, substance using persons from the area, and law enforcement seemed the most expedient means toward achieving this end. The problem was simple from their perspective, make more arrests related to what they saw as illegal behaviors carried out in open view. Some community members went as far as to suggest involuntary commitment for those who were visibly under the influence of opiates. These opposing viewpoints led to highly charged public meetings where residents shouted down city officials. Finally, people from the community suggested 78 A variant of this quote was heard on numerous occasions, including by Raymond Convery, Inspector in command of the Philadelphia Police Department’s (PPD) East Division, and Stephen Clark, Captain of the PPD’s 24th Division, which included the homeless camps. 79 Interview with Captain Stephen Clark was conducted on September 21, 2018. 69  that security issues could be improved by offering subsidized security camera installation for community members who want them, but could not otherwise afford them. While community members don’t want to become overly surveilled, cameras could prove helpful in reporting illicit activities while also offering additional data points to PPD. According to advocates, the relationship between people living on the streets and in the remaining Kensington encampments was fraught with frustration and tension. This included instances in which members of law enforcement took a heavy-handed approach to moving people along on sidewalks (to reduce loitering), while law enforcement, on the other hand, believed that this approach has prevented camps from reopening. Advocates have also detailed how members of law enforcement are slower to address concerns from people who remain on the street, particularly crimes that are committed against them. In light of bottles thrown, pellet guns shot, and insults hurled at people living on the street, they felt law enforcement did not address these concerns in a timely or compassionate manner. One advocacy organization noted that one way to help officers gain some empathy for the struggles of those on the streets is to offer policing opportunities to those with a vested interest in policing Kensington. On the other hand, some officers expressed a lack of desire to work in this area, given the constraints to policing and the issues that Kensington presents. Discussion of law enforcement issues also led, in some instances, to OPSs. When asked about their opinions on OPSs, which could potentially limit some of the drug paraphernalia found on the streets of Kensington, many community residents were not in favor as they believe it would further promote addiction. Moreover, many were uneasy with the idea that people would be able to go to a sanctioned place to use drugs without legal consequences, suggesting that it was not fair to those who don’t break the law. However, some were supportive of the idea, but were not necessarily in favor of having a OPS in their neighborhoods. Advocates, in contrast, fully supported the ideas of a OPS and tended to be the most familiar with the associated research and potential benefits. However, they expressed concerns that those using illegal substance may not be as trusting of the idea, instead seeing it as a way of entrapping people in possession of narcotics. 5.3.2- Data-based Measures The Kensington Avenue and Tulip Street encampments stayed closed, and returned to being non- descript segments of street that dipped under railroad tracks to provide through routes for vehicular and pedestrian traffic. There were no encampments that were established in the wake of these closures that might be seen as replacing these camps. From those basic measures, the closures were successful. Data collected and provided by PPD80 provided the basis for further empirical measures that could be useful for examining potential impacts that the closures may have had on surrounding area. The results from the PPD data that are reported here provide a means to validate some of the perceived impacts that we just reported, and provide further context on how these camp closures changed neighborhood conditions and quality of life. Census Counts. One regularly collected data source were police counts of people who were apparently unsheltered that were taken in the late night or early morning and covered the more general sector that included the former encampment sites. We reported results from similar counts, which only included the immediate encampment areas, in Chapter 4. Police also continued these more limited counts. One expected change was that at least some of the persons displaced from the Kensington and Tulip encampments would move to the still-standing Emerald and Frankford encampments. The Emerald 80 We acknowledge Captain Stephen Clark and Analyst James Carrion of the 24th District of the PPD for compiling these data and corresponding maps. 70 encampment had nightly census counts in the 40s and 50s through late April and upwards of 70 (with a high of 87 people) in the weeks leading up to the camp closure.81 In June, the Emerald census climbed to around 100 and then remained largely steady, with temporary spikes reaching as high as 130 on Labor Day weekend. The Frankford encampment census also expanded, but more modestly, from counts in the 30s in April to the 40s in May and then into the 50s during the summer. The maximum numbers of people at these two encampments appeared bounded by the space available under the underpasses. There was concern in June that the camp would encroach the sidewalk in each encampment that was left passable for pedestrians, or to areas outside of the underpass, but these expansions did not come to pass. Nonetheless, while the links between the closing of the Kensington and Tulip encampments and the expansion of the Emerald and Frankford encampments cannot be definitively proven, a substantial expansion of the latter two camps occurred immediately after the resolution of the former two encampments on May 30. Beyond the Emerald and Frankford encampments, police enforced strictures against setting up tents, laying down mattresses, and any other structures that could be precursors to new encampments. What camps emerged over the summer were small and in secluded areas. The largest such camp was set up along the Conrail tracks--one near and above the Tulip Street underpass and another further east where the track area bisects Trenton Avenue. Both were dismantled by Conrail employees, with the help of Philadelphia Police, at the request of Conrail. Police monitored the cleared Tulip Street and Kensington Avenue underpass sites, and did not report any attempts at resettling. Police did permit sleeping outdoors, with the understanding that the sleepers would move on, with their belongings, in the early morning. The 1.5-mile stretch of Kensington Avenue from the underpass north to Erie St. became the site of as many as 150 persons nightly who slept individually or in small groups on the sidewalk, against storefronts, and in the elevated train stops. Somerset Avenue, which runs perpendicular (east-west) to Kensington Avenue along the northern side of the railroad tracks (see Appendix B), also had clusters of persons sleeping overnight at various locations. Residents and police reported further sleeping activity in a variety of locations, including Harrowgate Park, under the I-95 underpass, McVeigh Playground, and numerous street locations. There were doubtlessly others sleeping more inconspicuously in more secluded areas among the abandoned railroad spurs and vacant industrial buildings throughout the area. Where the census had tallied 300 unsheltered persons in late May (before the encampments were cleared), a month later, on June 22, police enumerated 454 people, mainly in the Kensington area. In late July this headcount was at 500 unsheltered homeless. Then in August the police enumerated an unprecedented 703 persons. This drew press attention that featured quotes by OSH Director Liz Hersh, pointing out that “more than half of the city's [unsheltered] homeless population is now concentrated in Kensington,” and Brian Abernathy, the city's First Deputy Managing Director, stating “that the neighborhood is under siege."82 81 The census numbers that follow were reported at the regular ERP planning meetings by Philadelphia Police Department and DBH officials over the course of the summer (June through September) 2018. 82 Aubrey Whelan, “Philadelphia’s Kensington ‘under siege’ as opioid-linked homelessness soars” philly.com, http://www2.philly.com/philly/health/addiction/opioid-epidemic-heroin-homeless-population-kensington- philadelphia-20180918.html. See also, Tom MacDonald, “Homeless increase in Kensington has city councilwoman angrily calling for answers,” WHYY, September 19, 2018, https://whyy.org/articles/homeless-increase-in- kensington-has-city-councilwoman-angrily-calling-for-answers/. 71 Encampment Arrests. While the police did not, per policy, prioritize punitive and law enforcement-based approaches in the area around the encampments, there was a substantial police presence in the area both before and after the encampment closures. Data to quantify this observation was difficult to obtain. In one telling statistic, police reported, in July 2018, there to have been 1,100 police calls at the Emerald Street encampment alone during the first half of 2018. Most of these calls were from area residents or people passing by, and involved assaults, fights, and disorderly conduct. Among these calls were two shootings at the Emerald encampment; with one over the summer being fatal. PPD Captain Clark stated that there was almost certainly drug trafficking going in the Emerald encampment, and there were reports of human trafficking occurring from this encampment as well. While there was no substantiation of the nature of these trafficking activities, the presence of such activities was widely accepted by officials and workers involved in the ERP, along with the belief that these activities would complicate efforts to close the Emerald encampment. Figure 5a. Monthly numbers of arrests (January 2018 through September 2018) in two parts of Kensington impacted by opioid use. Police also provided data on narcotics arrests for the areas immediately surrounding three of the four underpass encampments (Kensington, Frankford and Emerald), as well as on the Kensington Avenue corridor between Somerset Street and Allegheny Avenue. Figure 5a shows the tallies of monthly arrests for the three camps (combined) and the Kensington Avenue corridor covering the first nine months of 2018. Two apparent conclusions based upon this figure are, first, that there were no clear trends in these arrests, and that the number of arrests was modest compared to the level of drug use and 72 trafficking that was described as typically happening in these areas. There was little apparent correlation between the fluctuations among the two areas. For example, narcotics arrest numbers increased over the first three months of 2018 for the Kensington Avenue corridor, while they declined during that same period around the encampments. The July spike around the encampment areas corresponded to a spate of narcotics arrests that all occurred at the Emerald encampment. Crime data for the first nine months of 2018 for the area immediately surrounding the Tulip camp was also provided, and involved crime reports (instead of arrests) over a broader variety of offenses. Because of this, these data were not comparable with the data used for Figure 5a. The narrative accompanying the 2018 police data describes the Tulip encampment as: the scion of the criminal activity increase in the surrounding neighborhood, much to the dismay of the residents and people who work in the area. The crime in this area includes a Robbery point of handgun, four (4) Aggravated Assaults by Handgun, one (1) Rape, nineteen (19) Thefts, six (6) calls for Gunshots, eleven (11) Vandalisms, ten (10) Thefts from Auto, forty-eight (48) calls for a Person with a Gun, nine (9) Narcotics Arrest and a VUFA (Violation of the Uniform Firearms Act).83 Data were not clear on how many arrests were made in association with these crime reports, or the extent to which the perpetrators of these crimes had ties to the Tulip Street (or any other) encampment. Emergency Responses to Overdoses. Figure 5b shows the number of overdoses where emergency responders were present, broken down by month, from July 2017 through August 2018. The number of overdoses around the four encampments includes those that occurred in the immediate vicinity of the encampments (between Lehigh Avenue and Somerset Street), while those on Kensington Avenue occurred directly on the 0.6 mile-long commercial corridor from Somerset Street going north to Allegheny Avenue. The results show a clear seasonal trend where the number of emergency calls dropped during the colder months and peaked in the summer months. Comparing the summer months of 2017 and 2018, the Kensington corridor saw a substantial increase in overdoses and the encampment areas saw a decreased number. The seasonal variation in overdoses is consistent with the increased numbers of unsheltered persons experiencing homelessness counted during the summer months. The summer 2018 findings showed a sharp increase in overdoses along the Kensington Avenue corridor that was consistent with the influx of unsheltered persons using opioids into the area during this time, while the net decrease in overdoses in encampment areas was consistent with observations of drug activity shifting toward Kensington Avenue as the areas around the Kensington Avenue and Tulip Street encampments became less hospitable. The limited nature of this overdose data has implications for their interpretation. The persons who overdosed were not necessarily homeless or affiliated with the encampments. A majority of persons who overdosed presumably did not come to the attention of emergency responders, although the greater police and social service presence around the encampments in the spring and summer months in 2018 would likely have led to a higher proportion of overdoses in that area that involved emergency responders. Thus the divergence in trends in the summer months of 2018 between the Kensington Avenue corridor and the encampment indicates some shift in opioid use away from the encampment areas, although there is still a substantial amount of such activity in the these areas. 83 Quoted from a report provided to ERP evaluators by Analyst James Carrion of the 24th District of the Philadelphia Police Department. 73 The other overdose data that were available from the 24th District involved fatalities. Overdose fatalities are more complete than more general overdose data, as fatalities would have been much more likely to involve emergency responders than non-fatal overdoses. The data included fifteen fatal overdoses (out of the 578 total overdoses involving emergency responders) recorded during the time period covered (July 2017 through August 2018). Of these fatalities, one-third (n=5) occurred in the areas around the encampments. Two of the fatalities, both in the Kensington corridor, occurred during the period in which the ERP was active (one in July and one in August). Based on this, there was no indication that the camp clearances led to an increase in fatal overdoses as advocates had feared, although the limited areal coverage limits making any definitive conclusions on this. Figure 5b. Monthly emergency response calls for overdoses (July 2017 through August 2018) in two parts of Kensington impacted by opioid use. 5.3.3- Impact of the Encampment Closures on the Surrounding Area Summary The closing of the Kensington and Tulip encampments preceded summer months where there were large increases in the numbers of sheltered persons experiencing homelessness. The camp closures would not have contributed to this increase as those displaced by the encampment closures were already in this area. However, the influx of incoming people experiencing homelessness more than replaced the persons who left the area in conjunction as a result of ERP. Despite this influx, the erstwhile encampment sites stayed clear, and no new encampments emerged. In the wake of the encampment closures, emergency calls for overdoses went down in the encampment areas, while they increased in 74 the adjoining Kensington Avenue commercial corridor. There was, however, no corresponding pattern with arrests. More generally, it was a particularly difficult summer for the area surrounding the encampment sites. This was reflected in stakeholder comments that typically took little solace in the encampment closures and instead focused on correlates of opioid use and outdoor sleeping that, by their accounts, persisted during the months following the closures. In other words, local residents appreciated not having an encampment in view of an elementary school, but the law enforcement and public health concerns related to the encampments did not abate. At the extreme, residents pondered whether it was preferable to have these problems concentrated in several encampments, as it was prior to the closures, or more dispersed in the surrounding community as it was over the subsequent summer. We present these conclusions not as a critique of ERP as much as to underscore the difference in relative size between the ERP initiative and the twin opioid and homelessness problems in the area around the encampments. This mismatch relegates the encampment closures to a tactical gain which were forgotten in light of the larger problems which, if anything, increased to where by the end of the summer they became even more difficult to ignore than the encampments were six months prior. A subsequent chain of events followed in tacit recognition of this situation. In August, the council persons in whose districts cover Kensington, released a letter calling for more assertive actions in response to the drug use and homelessness. It stated that the City has, in the past year, “normalized a situation that has become intolerable.”84 Subsequently, in early October, Mayor James Kenney took the dramatic move of declaring an opioid response emergency order, targeting the encampments in Kensington and conditions city-wide. This provided extra resources and facilitated an enhanced level of coordination between City departments that were already working in Kensington. Efforts to close the remaining two encampments at the Emerald and Frankford underpasses would be part of a more general effort that also included reducing trash and litter, including discarded hypodermic needles; reducing levels of unsheltered homelessness; reducing overdoses; increasing the number of people in substance use treatment; reducing open-air drug use and sales; and engaging long-term residents. This initiative, which cast subsequent encampment closures as part of a larger effort, would be known as the Philadelphia Resilience Project.85 84 Maria Quinones and Mark Squilla, “Memo to Mayor on Encampments and Related.” August 13, 2018, https://www.scribd.com/document/389012298/Memo-to-Mayor-Re-Encampments-August-2018#from_embed. 85 See https://www.phila.gov/2018-10-19-the-philadelphia-resilience-project-our-response-to-the-opioid-crisis/ for more information on this initiative. 75 Chapter 6- Conclusion and Lessons Learned This report has chronicled the ERP, including its planning phase in early 2018; its 30-day implementation phase that culminated with the closure of the underpass encampments on Kensington Avenue and Tulip Street on May 30; and ongoing efforts through October 15 to serve those displaced from the encampments and remain responsive to the concerns of the surrounding neighborhoods. As noted at the end of the previous chapter, although the ERP activities have largely concluded,86 the Philadelphia Resilience Project has continued and expanded the work of the ERP, including closing the two remaining encampments on Frankford Avenue (in November 2018) and Emerald Street (in January 2019). With the access we were given to people, meetings, documentation, and data for this process evaluation, we have been able to undertake an evaluation that is unprecedented in the depth of its examination into how resolving a large-scale encampment was planned and carried out. This has enabled us to report on the characteristics, services use, and housing and treatment outcomes for those displaced by the camp closures. Additionally, we have been able to document how the ERP actions impacted the people of staying in and living in the vicinity of the encampments and the larger community. Here, we distill this material, which we have reported in detail in the previous chapters, to what we consider to be the key findings from our assessment of the ERP. We present these findings in the form of 22 “lessons learned” that we find insightful, instructive, or corrective. 6.1- Lessons Learned in the Planning Phase 6.1.1- Don’t reinvent the wheel. The initial task of the ERP planning committee was to look at other cities for a model to follow in closing the encampments. San Francisco’s approach, which featured procedures for engaging people experiencing homelessness with services as part of closing the encampment, was chosen. In Philadelphia, the ERP had to augment the original San Francisco model by accommodating the need for substance use treatment services, making the Philadelphia effort more complex as it involved including an entirely separate service delivery system. Despite such challenges, the ERP has demonstrated how this model, combining services with encampment closure, can be successfully adapted and replicated to clearing encampments while engaging people with needed services. As such, this approach can be further applied to addressing other encampments in Philadelphia and elsewhere. 6.1.2- Effective coordination between participating entities is essential to overall success. Key organizations involved in the ERP began working together with the Gurney Street closure, and built upon their collaboration in the context of this pilot. As demonstrated throughout this report, implementing the ERP required coordinating multiple facets of City government with non-profit agencies and taking into consideration the needs of multiple constituencies. Based upon our access to many planning meetings and field settings related to implementing the ERP, we observed a high and, in our experience, unusual degree of coordination and cooperation among an array of services. This included cooperation among such diverse entities as police, homeless outreach, sanitation, and behavioral health services. This was noteworthy and critical for the successful development and carrying 86 As confirmed by OHS Director Liz Hersh, there has not been an official end date to the ERP and persons with names on the BNL could continue to receive temporary housing and substance use services. However, by mid- October 2018, the ERP BNL referred to in this report was no longer the basis for placing individuals into these beds. 76 out of the ERP, and formed a basis for the continued collaboration and coordination of these entities in the subsequent, and more comprehensive, Philadelphia Resilience Project. 6.1.3- Keep expectations in perspective. The ERP was a small project charged with addressing a specific situation involving two encampments that emanated from the much larger contexts of homelessness and opioid use. ERP organizers emphasized the heroic nature of pitting the limited resources of a pilot project against these daunting public health crises, and with the underlying imperative of taking action, even if the measures taken were new and untested. This provided a justification for setting aside bureaucratic caution, and also offered an opportunity to implement novel approaches as well as to evaluate and learn from the measures taken. 6.2- Lessons Learned in the Implementation Phase 6.2.1- Encampments are an opioid-related problem. The Outreach Encampment Survey found near ubiquitous substance use among those staying in the encampments, with 94 percent of respondents reporting current substance use and 73 percent of all respondents (79 percent of those reporting substance use) reporting opioids as their drug of choice. Nearly three-quarters of respondents (73 percent) reported that they had previously been to substance use treatment, and an equal proportion expressed interest in getting treatment for their drug use (74 percent). A match between 189 people on the BNL and the City’s behavioral health services records gave similar results using more reliable data: with almost two-thirds (65.1 percent) having received at least one substance dependency diagnosis and 60.8 percent (i.e., almost all of those diagnosed with a substance dependency) having received an opioid dependency diagnosis. While perhaps evoking little surprise, these findings show extremely high rates of substance use and diagnosed dependency, even when considering the typically high rates that are usually found in surveys of homeless populations. These findings further underscore how the ready availability of substance use services was a critical component of the ERP. 6.2.2- Encampments are a homelessness-related problem. All people staying in the Kensington Avenue and Tulip Street encampments were, by virtue of their sleeping in these encampments, homeless. For most, their encampment stay was part of a more extended pattern of homelessness. Findings from the Outreach Encampment Survey and CARES records matches found that over half (57 percent) of survey respondents reported having spent time in a homeless shelter and an even higher number (63 percent) of those on the BNL had a record of having stayed at a homeless shelter. For survey respondents, the median time spent homeless was 6 to 9 months, and 40 percent reported being homeless for over a year. Given that virtually all respondents likely had substance use issues (among other conditions) that may have been considered disabling, a large proportion of this 40 percent would likely have fit the federal definition of "chronically homeless." This is clearly a population with substantial interactions with the local homeless services system, and with substantial housing needs, and not simply a population who was homeless “by choice” in order to facilitate their substance use. 6.2.3- Homelessness in Kensington is a Philadelphia problem. On the Outreach Encampment Survey, 84 percent of survey respondents were Philadelphia residents, with 65 percent stating they were from Philadelphia and 19 percent coming from elsewhere but had lived in Philadelphia for over a year. This indicates the prominence of local origins among the encampment population and the need for services to treat unsheltered homelessness in Kensington as a problem with Philadelphia origins and requiring Philadelphia-based solutions. Furthermore, this casts 77 the oft-repeated assumption that the encampment population is largely the result of people migrating into Philadelphia in pursuit of high-quality heroin as an unsubstantiated myth. 6.2.4- Temporary housing availability is limited by local resistance. Throughout the Kensington Avenue and Tulip Street encampment closure processes, the City stated that there were ample temporary housing beds to accommodate demand. However, this availability was uneven: the popularity of the local navigation center beds ensured that they remained full, while the more distant respite beds remained available throughout the implementation phase. Additional, local temporary housing was blocked by community interests, which resisted the local siting of temporary housing and other services, and viewed Kensington as an area that already had a disproportionate amount of service facilities. The resulting temporary housing supply appears to have been an inadvertent compromise: more than many residents wanted, but less than providers and advocates maintained was needed. More proximal services would likely have led to higher levels of engagement by those staying in the encampments. 6.2.5- Closing encampments means balancing competing interests. The standoff over housing was one example of contrasting views that divide community members and advocates. This report examined the contrasting perspectives of these two key stakeholders, where those living in the area around the two encampments held the predominant view that removing the homeless and opioid using populations was a prerequisite to improving the local quality of life, while the more diffuse group that supported those staying in the encampments viewed the closures as inflicting further harm to an extremely marginalized population. Both of these groups criticized aspects of the closure process, though from different perspectives. The City, for its part, has tried to maintain a middle position and be responsive to concerns from both sides, but maintaining such a balance was challenging. We contrasted differing levels of involvement in the ERP planning and implementation processes that were afforded to community residents, by virtue of their political representation and access to community groups, and advocates, who expressed concern about what they saw as their exclusion and expressed their views by alternative means that could have been disruptive to the ERP process. 6.2.6- Individual placements do little to relieve population pressures at the encampments. Data presented in this evaluation demonstrated something already intuitively apparent to outreach workers: the substantial number of individual placements to housing and substance use treatment services did not lead to collateral reductions in the encampment populations. As people left, others took their places and the encampments maintained a rough population equilibrium. This does not suggest that providing services to get people out of the encampments is futile, as individuals stood to benefit greatly from such services. However, it begs for a better understanding of the latent demand and degree of natural turnover inherent to encampment populations. This can lead to services that can address this apparent “churning” dynamic. 6.2.7- Involving people experiencing homelessness in resolving encampments is difficult. One objective of the planning and implementation phases of the ERP that was not realized was soliciting meaningful input from persons who were staying in the encampments targeted for closure. A series of four “town hall” meetings were planned so that there was maximum opportunity for the target population to participate (one meeting was held at the Kensington encampment), but the level of participation was minimal at the two meetings that were actually held. Furthermore, no one from any of the encampments stepped forward to lay claim on any leadership role or to speak on behalf of others in any of the camps. Despite the initial failure in this objective, the goal remains important. This challenges future closure efforts to develop innovative ways to involve the target population in the closure process. 78 6.2.8- Effective services require removing access barriers. The necessity of providing attractive and effective housing and substance use treatment services under ERP resulted in streamlined approaches for providing low demand temporary housing and treatment on demand to those in the encampments. On the housing side, this required relaxing rules and being more flexible than shelters typically are on such topics as housing couples together. With substance use services, many of the approval processes were shifted to follow admission, and identification requirements were relaxed. Such changes made these services more amenable and reduced or eliminated the lag between asking for services and receiving them. These newly adopted best practices were instrumental in engaging 126 persons in housing and/or treatment at some point during the implementation phase. 6.2.9- Closing the encampments is the easy part. Closing the encampments was “easy” in that, as described at the end of Chapter 4, the actual removal of the encampments happened largely without incident and looked deceptively simple as the various City departments and other entities carried out their parts in what was, by consensus, a well-coordinated and executed effort. Beyond that, the closure represented a concrete action with a tangible outcome, and the whole process was attainable with the resources allocated to the ERP. After the camp closure, meeting ERP’s objectives became more challenging and necessitated bringing the limited pilot resources to bear against the larger, more diffuse problems associated with homelessness and substance use in the Kensington area. 6.3- Lessons Learned in the Sustainment Phase 6.3.1- Most of the encampment population is involved with municipal services systems. Among those on the by-name list (BNL), who comprised those most targeted for services through ERP, 55 percent were actively enrolled in Medicaid and 90 percent were matched with some type of record in the CARES data repository. This means that most of the individuals in the encampments were not only known to the various City services systems, and made substantial use of healthcare, housing, treatment and criminal justice services. Providing effective, coordinated services would not only mean better individual outcomes, but would also mean more efficient and cost-effective services delivery. 6.3.2- Use of a by-name list (BNL) is essential to coordinating individuals’ services. Creating and maintaining a BNL was the centerpiece of engaging with and managing services for 189 persons targeted for services. Six months after the encampments were cleared, the BNL was instrumental in enabling outreach staff and caseworkers to maintain ongoing contact with 62 percent of the target population, and engaging 41 percent with housing or substance use treatment services. BNLs in homeless services are a relatively recent development, and their use permits clearer targeting of services, better coordination between different agencies and services providers, and data for assessing outcomes. Based upon our observations, use of a BNL provided a model for other efforts. However, ongoing improvements are also being implemented to address limitations with how the BNL was structured and used for the ERP. This includes ways to better identify people who should be on the BNL; collecting and entering services data from various sources onto the BNL, and consolidating these data for targeted individuals to provide meaningful profiles for services planning and outcomes measures. 6.3.3- Having data makes a difference. The ERP was initiated with a survey of people in the encampments; had access to data from CARES, a state-of-the-art integrated data system for tracking municipal social, mental health, and corrections services; and further collected data based upon a BNL containing names of individuals from the encampments that were targeted for services. This enabled a detailed, data-rich examination of various 79 aspects of the initiative in documenting the process and some basic outcomes, all reported here. The use of data and evaluation in future projects can learn from and build upon what was accomplished in this pilot. Established data sharing arrangements, decreased time lags between services provided and records being available, and better data integration protocols are examples of improvements that would facilitate even more effective uses of the available data. 6.3.4- Be deliberate in identifying outcomes and setting benchmarks. This evaluation began when the initiation phase commenced. This meant there was limited opportunity to structure the evaluation beyond existing programmatic structures, and left the evaluators to adapt the evaluation to existing data, rather than to design data collection to better fit with existing evaluation questions. As mentioned in Chapter 5, what constituted a successful program outcome, and the benchmarks by which to measure such outcomes, were also unclear. For example, “success” in providing services could be operationalized as outreach workers maintaining engagement with the target population; getting people to accept short-term housing and substance use services; or placements in permanent housing and long-term recovery programs. We provide measures for each of these outcomes levels, but were unable to set specific benchmarks for what degree of engagement, initial services participation, or ultimate placements constitutes success. Establishing such benchmarks is difficult; the findings reported here should help future evaluations meet this challenge. 6.3.5- There is no model for policing an area overwhelmed by homelessness and opioid use. The role of the police in the ERP had, by the admission of the police captain of the 24th District, no apparent precedent for formulating a response to the circumstances around the encampments. Police recognized the futility of a heavy-handed law enforcement approach, repeating frequently that they could not arrest their way out of the situation. Police officials instead took a public health approach in which they balanced the need to provide support for outreach and other engagement efforts with the need to ensure some basic quality of life standards for the surrounding neighborhoods. This put them in an unfamiliar position which none of the community stakeholders appeared to appreciate. Police officials were often excoriated by persons at community meetings and other public functions for what they perceived as police inactivity in the face of open and illegal behaviors. Many aspects of the police’s role in this situation were not in the bounds of traditional policing, and more support is needed to further formulate, implement, and communicate a clearer and more proactive role for the police. 6.3.6- Availability of short-term resources contrasts with scarcity of long-term resources. In October 2018, half of the 72 persons on the BNL who received either housing or substance use services were in long-term or permanent placements. This proportion would be higher, and the ERP outcomes would have improved, had more permanent housing and long-term recovery housing been available citywide. The ERP helped substantial numbers of persons access short-term housing and treatment services, in part through the availability of dedicated resources to facilitate timely placements that met recipients’ needs and preferences. However, after individuals stabilized in these short-term programs, they joined others city-wide seeking placements in the limited supply of permanent housing and long-term recovery housing. Ultimately, sustaining the gains made by pilot projects such as ERP will be upon the more general availability of long-term housing and recovery housing. 6.3.7- Consolidating gains made by pilot program requires routinizing pilot services. Many of the ERP services required either diversions of existing resources or additional resource allocations. Police, housing and outreach services were prominent examples of this. Administrative attention to the ERP was also considerable. All of this was essential to the accomplishments achieved under the ERP, but leaves questions about the sustainability of this effort. Converting the level of services expended in these encampment closures into ongoing services is necessary for continued 80 access to housing and substance use treatment for the unsheltered homeless population targeted in this pilot. 6.3.8- Summer 2018 was long and difficult in Kensington. In the summer following the encampment closures, the number of persons counted as unsheltered and homeless increased to an unprecedented 700. This increase was unrelated to the encampment closures, but efforts to consolidate the gains from the closures were strained from this influx. Police struggled to prevent new encampments, and more people slept in makeshift outdoor locations in the area near the former encampments. This increase in homelessness muted the impact of the encampment closures for the surrounding neighborhoods, and stakeholders typically looked beyond the encampment closures to the continuing correlates of opioid use and outdoor sleeping. This underscores the limited overall impact of a targeted initiative such as ERP. 6.3.9- Crisis creates opportunity. This modest initiative to close two encampments underscores the massive scale of the affordable housing and opioid addiction problems in Philadelphia and nationally. One response to this mismatch of resources to need is incorporating innovative and alternative means to resolve the encampments and provide services. Among the lasting and most widely adopted innovations of the ERP was changing intake procedures to facilitate and expedite access to substance use treatment. Changes in how temporary housing was provided led to people using these services who would otherwise have remained outdoors. Implementing such changes were more feasible in crisis conditions like those faced in Kensington, and the ERP was able to capitalize on local circumstances. This report contains further ideas from ERP officials and providers, stakeholders, and those using (or declining) services about measures that might supplement or improve upon ERP efforts. Other jurisdictions have taken various approaches to addressing encampments that bear further exploration. Taken together, these and other sources can continue the innovation that has been central to the ERP process. 6.3.10- Pilots should lead to larger initiatives. We closed our examination of the ERP at the point at which the City implemented the Philadelphia Resilience Project, which proposes a broader effort to address the homelessness and substance use issues that continued to proliferate in Kensington and citywide during Summer 2018. The Philadelphia Resilience Project will continue the efforts of the ERP in the sense that closing the two remaining encampments, on Emerald Street and Frankford Avenue, are among the objectives it will address. Beyond that, many of the Philadelphia Resilience Project initiatives would not be possible without the groundwork in logistics, inter-departmental and agency coordination, and services provision that was developed through the ERP. This illustrates the value of a pilot project such as the ERP, but efforts to further apply approaches and lessons learned through this initiative should continue- in Kensington, other parts of Philadelphia, and in other areas faced with similar situations. 81 Appendix A- Methods This evaluation of the City’s Kensington Encampment Resolution Pilot comprised four primary activities, in a modified application of the RARE method (Rapid Assessment, Response and Evaluation), to provide a timely, multifaceted assessment of the closure and resettlement process, as well as of the impact on the displaced residents and surrounding communities. The first activity consisted of an initial profile of the residents in the encampments, based upon survey data collected by outreach workers. The second activity built upon this initial profile by linking survey data with the City’s CARES database to provide additional service history context, and a baseline for assessing changes in services use that may have resulted from the intervention. The third activity was an ethnographic study of the impact of the closure and resettlement based upon direct observations, and interviews with people staying in encampments, community members, service providers and City representatives, along with use of the information from the City’s community services data system. Finally, a process evaluation documented the history of the intervention development, the implementation process, and the relative efficacy of the project. Four primary strategies form the core of the evaluation: 1. Outreach Encampment Survey. People staying in encampments were surveyed by social service providers to obtain identifying information, current living conditions and homelessness history, place of origin, OHS intake history, service needs and interests, substance use, treatment status, and barriers to treatment. The client survey form was developed by the City, and modified based on feedback from the evaluation team. 2. Philadelphia Health and Human Services Utilization Profile. Encampment surveys were merged with CARES database maintained by the City’s Data Management Office to develop a baseline assessment of residents’ services utilization history prior to encampment closure, including use of mental health and substance use treatment, homeless services, city jail, child welfare, and other services tracked in the CARES database. Use of those services post encampment closure was also assessed. 3. Qualitative and Ethnographic Data. An ethnographic team collected data on the closure process that encompassed the array of perspectives held by various stakeholders. Data sources included interviews with people staying in encampments, community residents, City representatives (e.g., OHS, CBH, DBHiDS, PPD), and service providers (e.g., PP, ODAAT, NET). Additionally, the team gathered field notes during direct observation of the encampment and closure, and at the PP and ODAAT shelters, and at ERP and community meetings. See Appendices F through H for Interview Guides. 4. Process Evaluation. This strategy provided an overall narrative of the closure and resettlement process, and a critical assessment of the key elements of this process. Data for the process evaluation came from a review and summary of relevant planning documents, meeting minutes, policies and procedures, and implementation plans and decisions. Direct observation of meetings among stakeholders informed this assessment, and were supplemented by qualitative interview data. Program documents recording residents’ placements were also be gathered and summarized. 82 Appendix B- Map of Area Containing the Targeted Encampments Sources: Inset map was adapted from a map on the Spirit News website (https://spiritnews.org/wp- content/uploads/2015/07/New-Map.png) and the Philadelphia map is from the Philadelphia City Planning Commission. 83 Appendix C- Outreach Encampment Survey 84 Appendix D- Outreach Encampment Survey Results Responses are based off of an unduplicated number of individuals surveyed, which account for most recent response given from each individual. Outreach Encampment Survey Results: Demographics (N=169) Item % Gender (n=165) Male 76 Female 24 Age (n=162) 18-24 6 25-34 31 35-44 31 45-54 20 55-64 9 65+ 2 Race (n=147) Black/African American 31 White 57 Asian 0 Latino/a 12 Veteran Status (n=161) Veteran 6 Not Veteran 94 Discharge Type (n=10) Honorable 90 Dishonorable 10 First Language (n=131) English 94 Spanish 6 85 Outreach Encampment Survey Results: Current Living Conditions and Place of Origin (N=169) Item % Are you with a partner? (n=157) 33 Where did you sleep last night? (n=155) Street/bridge 92 Emergency housing/Safe Haven/Journey of Hope 1 Other 8 Do you currently live on the street? (n=161) 94 How long have you been living on the street? (n=159) Not homeless 3 Less than 1 month 3 1-3 months 17 3-6 months 14 6-9 months 14 9-12 months 9 Over 1 year 40 Are you from Philadelphia? (n=162) 65 If not, how long have you been in Philadelphia? (n=53) Less than 1 month 4 1-3 months 11 3-6 months 9 6-9 months 15 9-12 months 6 Over 1 year 55 Do you still have family, friends, or other support where you are from? (n=59) 63 If able, are you willing to return? (n=56) 39 Are you interested in receiving assistance reaching family members and possibly mediation services to help address family conflicts? (n=133) 41 Note. Results reflect an affirmative response. 86 Outreach Encampment Survey Results: Homeless Service Use and Interest, Employment History, and Service Needs (N=169) Item % Have you ever gone to a shelter in Philadelphia? (n=165) 57 Are you willing to go into shelter? (n=160) 56 Do you know about Safe Haven/Journey of Hope? (n=154) 50 If there was a residence with minimal rules, would you go? (n=139) 91 Are you interested in housing? (n=167) 94 Would you be willing to go to any of the following? (n=132) Journey of Hope 65 Safe Haven 71 Respite    58 Overnight café 43 Shelter 39 Other 4 Do you have an employment history? (n=159)    82 Are you interested in employment assistance? (n=156) 76 Are you looking for help with any of the following? (n=162) Housing 91 Work/job 62 Applying for entitlements/benefits    33 Substance use treatment 56 HIV treatment 3 Identification/birth certificate/social security card    42 Mental health treatment 30 Medical treatment    21 If not, what is preventing you from accessing assistance (n=84) No identification No/out-of-county insurance Do not want to leave community Want to remain with partner/relative Criminal history Substance use Lack of work history/job training Ongoing mental health issues Medical issues Trouble navigating the system independently Other Note. Results reflect an affirmative response. 64 12 8 10 14 33 5 15 14 19 7 87 Outreach Encampment Survey Results: Substance Use, Treatment and Mental Health (N=169) Item % Do you currently use alcohol? (n=162) Do you currently use drugs? (n=166) What is your drug of choice? (n=154) 25 93 Opioids Marijuana K2 9 PCP 3 Cocaine 43 Benzodiazepines 8 Other 0 Have you used in the last 6 months? (n=149)    94 Route of administration? (n=133) Oral 11 Inhalation 35 Subcutaneous injection 11 Intravenous injection 71 Are you interested in getting treatment for you drug use? (n=155) 74 Have you ever been on any form of medically assisted treatment? (n=133) Methadone treatment 27 Suboxone treatment    31 Vivitrol/Naltrexone 4 None 48 Are you interested in medically assisted treatment? (n=142)    61 Have you ever been to treatment? (n=162) 73 Are you currently enrolled in treatment? (n=149) 12 Would you be interested in long-term housing that may lead to treatment? (n=148)      82 What is keeping you from getting treatment? (n=117) No insurance/cost 31 Not interested in or ready for treatment at this time 52 No program nearby/transportation problems 17 Other 61 Have you had any challenges with getting into treatment recently? (n=136) 28 Do you struggle with mental health challenges? (n=161) 65 If so, what mental health challenges? (n=103) Anxiety 50 Depression 68 Bipolar Disorder 44 Schizophrenia 19 Post-Traumatic Street Disorder 19 Other trauma disorder 3 Other 5 If so, are you receiving help for any mental health challenges? (n=95) 32 Are you interested in receiving help for any mental health challenges? (n=97) 71 Note. Results reflect an affirmative response. 79 18 88 Appendix E- Sample Weekly Outreach Schedule 89 Appendix F- Interview Guide for People Staying in Encampments and Served through the ERP INTRODUCTION Introduction/Description of evaluation and reason for interview Confidentiality Expected duration: 15-20 minutes Questions Permission to begin LIVING SITUATION Tell me about coming here. • First, how do you refer to this place? • Where had you been staying before coming here? For how long? What type of place was it (e.g., apartment, recovery house)? • How did you hear about the camp? o Howdidyoucometobestayinghere?Areyoustayingherewithanyoneelse? o Howlonghaveyoubeenhere? o Whatareyourreasonsforcontinuingtostayhere? o Whataresomegoodthingsaboutstayinghere? o Whataresomedifficultthingsaboutstayinghere? • Are you currently staying anywhere else? TYPICAL DAY • What does a typical day look like to you? • Where else do you hang out? • Can you tell me about any drugs you currently use (e.g., alcohol, opioids [heroine, fentanyl, oxys, percs], marijuana, benzos [Xanax, klonopin, valium], cocaine, PCP, K2)? o Howoldwereyouwhenyoufirstused?What’syourcurrentuselike?Howdoesyour current use compare to a year ago? o Wheredoyouusuallybuy?Wheredoyouusuallyuse? 90 BACKGROUND AND SERVICE USE Next, I’d like to talk more about any services you’ve accessed. • Tell me about any services you currently use or have used in the past. o [Dependingonhowparticipantresponds,promptwithadditionalservicetypes]Whatabout ... substance use treatment, mental health, emergency, shelter/housing, HIV, employment, case management ... services? • Tell me about any plans you have to access services or get into treatment. • Can you tell me more about any barriers or challenges (e.g., personal [mental health, trauma] or programmatic [eligibility, requirements]) you’ve faced getting your needs met? • What, if anything, has been helpful in getting your needs met? CAMP CLOSURE Now, I’d like to talk more about your thoughts on the closure of the camp and what you plan to do next. • How did you first learn that the camp would be closing? What were your first thoughts? What are your thoughts now? • Where do you plan to go when the camp closes? What are some of your concerns? • Have you spoken with anyone from the City helping with the initiative (i.e., “orange shirts”)? How did that go? • What do you think should be done about the encampments? What are your suggestions for what the City should do to improve conditions and access to services? • What do you think about the respite centers that have opened to connect camp residents to services? • What do you think about the possibility of a supervised injection facility with services on-site? Would you be likely to go to a supervised injection facility? Why or why not? • What do you think about mobile treatment options (i.e., mobile suboxone van)? WRAP UP Finally, I’d like to talk about your goals, any help you might need reaching them, and anything else you’d like to share about your experiences that you think are important. • What are your goals? What help do you need reaching your goals? • Is there anything else you’d like to share about your experiences here? Appreciation for participating Follow-up period/Contact card Good place(s) for a meet-up in a couple weeks for follow-up Other residents who might be interested in participating 91 Appendix G- Interview Guide for Community Members INTRODUCTION Introduction/Description of evaluation and reason for interview Confidentiality Expected duration: 15-20 minutes Questions Permission to begin NEIGHBORHOOD RESIDENT BACKGROUND Tell me a little about yourself. • Where do you live? For how long? • Does your family live in the area? • Did you grow up here? • What perspective(s) do you feel you best represent when it comes to your community? • What’s your understanding of the drug problem in this area? o Whatisyourexperiencewiththedrugprobleminthisarea? o Wheredopeopleusingbuy?Wheredopeopleusuallyuse? o Whatkindofdrugsaremostcommon? o Howdoesthecurrentdrugsituationcomparetoayearago?Fiveyearsago? • Can you tell me more about how the camps operate? o Howlonghavepeoplebeenhere? o Howdoestheencampmentaffecttheneighborhood? o Whataresomeofthereasonsyouthinkpeoplecontinuetostayinthecamps? o Whataresomegoodthingsaboutstayingthere? o Whataresomedifficultthingsaboutstayingthere? AVAILABLE SERVICES Next, I’d like to talk more about available services. • What types of services are available in this area? How do people feel about these services? o [Dependingonhowparticipantresponds,promptwithadditionalservicetypes]Whatabout ... substance use treatment, mental health, emergency, shelter/housing, HIV, employment, case management ... services? 92 CAMP CLOSURE Now, I’d like to talk more about your thoughts on the closure of the camps. • How did you first learn that the camps would be closing? What were your first thoughts? What are your thoughts now? • What are your expectations for the closure of the encampments? Where do you think people will go when the camps close? What are some of your concerns? • Can you tell me more about city outreach efforts? How are they being received by the neighborhood? • What do you think should be done about the encampments? What are your suggestions for what the city should do to improve conditions and access to services? • What do you think about the respite centers that have opened to connect camp residents to services? • What do you think about the possibility of a supervised injection facility with services on-site? • What do you think about mobile treatment options (i.e., mobile suboxone van)? WRAP UP Finally, I’d like to talk about what an ideal outcome for these camps might be, what would be needed to achieve this outcome, and anything else you’d like to share about your experiences that you think are important. • What would be an ideal outcome for this initiative? What would be needed to achieve this outcome? • Is there anything else you’d like to share about your experiences here? Appreciation for participating Identification of other community members or people staying in encampments who might be interested in participating 93 Appendix H- Interview Guide for City Representatives and Service Providers INTRODUCTION Introduction/Description of evaluation and reason for interview Confidentiality Expected duration: 15-20 minutes Questions Permission to begin SERVICE PROVIDER BACKGROUND First, I’d like to learn more about your organization. • Can you tell me about the organization you work for? • Can you tell me more about your role at the organization? How long have you worked there? How did you come to be in this position? • What’s your understanding of the drug problem in this area? o Whatisyourexperiencewiththedrugprobleminthisarea? o Wheredopeopleusingbuy?Wheredopeopleusuallyuse? o Whatkindofdrugsaremostcommon? o Howdoesthecurrentdrugsituationcomparetoayearago?Fiveyearsago? • Can you tell me more about how the camps operate? o Howlonghavepeoplebeenhere? o Whataresomeofthereasonsyouthinktheycontinuetostayhere? o Whataresomegoodthingsaboutstayinghere? o Whataresomedifficultthingsaboutstayinghere? AVAILABLE SERVICES Next, I’d like to talk more about available services. • What types of services are available in this area? Which services do people staying in encampments most frequently access? How do people feel about these services? o [Dependingonhowparticipantresponds,promptwithadditionalservicetypes]Whatabout ... substance use treatment, mental health, emergency, shelter/housing, HIV, employment, case management ... services? • Can you tell me more about any barriers or challenges (e.g., personal [mental health, trauma] or programmatic [eligibility, requirements]) residents face in getting their needs met? 94 • What has been helpful in getting residents’ needs met? CAMP CLOSURE Now, I’d like to talk more about your thoughts on the closure of the camps. • Can you describe your organization’s role in the city’s initiative to close the encampments? • How did you first learn that the camps would be closing? What were your first thoughts? What are your thoughts now? • Where do you think people will go when the camps close? What are some of your concerns? • Can you tell me more about city outreach efforts? How are they being received by residents? • What do you think should be done about the encampments? What are your suggestions for what the city should do to improve conditions and access to services? • What do you think about the respite centers that have opened to connect camp residents to services? • What do you think about the possibility of a supervised injection facility with services on-site? Would you be likely to go to a supervised injection facility? Why or why not? • What do you think about mobile treatment options (i.e., mobile suboxone van)? WRAP UP Finally, I’d like to talk about what an ideal outcome for these camps might be, any resources you might need to reach this outcome, and anything else you’d like to share about your experiences that you think are important. • What would be an ideal outcome for this initiative? What resources do you need to reach this outcome? • Is there anything else you’d like to share about your experiences here? Appreciation for participating Identification of other service providers, people staying in encampments, or community members who might be interested in participating 95 Appendix I- Glossary Angels in Motion (AiM): Philadelphia advocacy organization; details at aimangelsinmotion.org By-Name List (BNL): List of all persons who were sleeping at the Kensington and Tulip encampments who were prioritized for housing and services CARES Integrated Data System: Consists of a common data model, dedicated data warehouse, a highly complex probabilistic matching algorithm, and Extract Transfer Load procedures that retrieve key data from source system databases; integrates and stores data about clients from 1997 to the present, including: demographics, client/family relationships, case history, case manager(s), service provider(s), services, and date of services; used for internal analytics, research, and integrated case management Community Behavioral Health (CBH): Not-for-profit 501c (3) corporation contracted by the City to provide mental health and substance abuse services for Philadelphia County Medicaid recipients Crisis Response Center (CRC): Emergency service center for individuals experiencing crises related to substance abuse and/or mental health issues Data Management Office (DMO): Collects, analyzes, and reports on social services data, including: mental health, substance abuse, child welfare, juvenile justice, lead exposure, emergency shelter, and local incarceration Department of Behavioral Health and Intellectual Disability Services (DBHIDS): One of five departments under the City’s Office of Health and Human Services; offers behavioral health care, intellectual disability supports and early intervention services through a comprehensive integrated system Department of Human Services (DHS): Public child welfare and juvenile justice agency Encampment Coordinating Team (ECT): ERP that provided operations leadership from DBHIDS and OHS Encampment Engagement Team (EET): ERP team that engaged individuals on the BNL with the aim of facilitating their exit from the camps and acceptance of services; coordinated outreach staff with police patrols to provide alternatives to enforcement actions Encampment Oversight Team (EOT): ERP team that provided logistical support, securing needed resources, and removing barriers to housing and recovery services Encampment Resolution Pilot (ERP): An initiative of the City of Philadelphia with the goal of shutting down two outdoor homeless encampments after actively reaching out to and providing assistance with housing, substance abuse and other services to people sleeping in the encampments Homeless Management Information System (HMIS): Data management system used to collect information about persons and families experiencing homelessness and those at risk of homelessness Journey of Hope: Collaboration between several long-term residential treatment programs designed to serve individuals experiencing prolonged homelessness, substance use disorders, and co-occurring mental health challenges; details at dbhids.org/about/organization/office-of-addiction- services/recovery-house-initiative/the-journey-of-hope-project/ Medically Assisted Treatment (MAT): Treatment option for individuals with opiate addiction that involves taking a regular dose of medication (e.g., Methadone, Suboxone, Vivitrol) to alleviate symptoms of withdrawal and help the individuals wean themselves off opiates 96 Managing Director’s Office (MDO): Responsible for directing the resolution group, media relations and other communications functions Mental Health Partnership (MHP): Philadelphia advocacy organization; details at mentalhealthpartnerships.org Navigation Center (also known as the Nav Center): Low-barrier temporary shelter operated by Prevention Point Philadelphia (PPP) as part of the Enclosure Resolution Pilot (ERP) NorthEast Treatment Centers (NET): Provides a continuum of trauma informed behavioral health and social services, including assessmets for substance use treatment; details at netcenters.org Office of Homeless Services (OHS): Provides emergency housing and other services to people who are homeless and to those at risk of homelessness Office of Mental Health (OMH): An integral component of DBHIDS, contracts with treatment providers to provide mental health services for people in Philadelphia One Day at a Time (ODAAT): Social services agency located in North Philadelphia that provides temporary shelter and case management services Overdose Prevention Sites (OPS): Also known as Safe Injection Sites (OPS), Supervised Injections Facilities (SIFs) and Comprehensive User Engagement Sites (CUES), these are locations where individuals can inject narcotics under the supervision of skilled nurses while simultaneously serving as a gateway to service engagement for housing, substance abuse and mental health treatment and others services utilizing a Harm Reduction approach Pennsylvania Department of Transportation (PennDOT): Oversees transportation issues across the state of Pennsylvania Philadelphia Department of Prisons (PDP): Local jail system holds people who are incarcerated during the pre-trial period if they have not been released on bail, for sentences under two years duration, and for sentences of longer duration until they can be transferred to the state corrections system Philadelphia Police Department (PPD): Provided security, coordinate with outreach, and lead role in shutting down the encampments and preventing (with outreach services) resettlement Philadelphia Resilience Project: The City’s unified approach to tackling the opioid crisis, triggered by the signing of Executive Order 3-18 by Mayor Jim Kenney, which declared a citywide emergency and empowered City agencies to come together to immediately solve the problem. The order prompted officials to gather together for an intensive, two-week period to meet daily, share ideas and perspectives, assess available resources, and identify needs and gaps in services. Prevention Point Philadelphia (PPP): Social services agency located in Kensington that offers a needle exchange program, temporary shelter, and other services Project HOME: Philadelphia non-profit organization that provides affordable housing, employment, health care, and education; participated in outreach activities during the ERP Project SAFE: Advocacy organization that provides services and information that helps street-based sex workers in Philadelphia 97 Safe Haven: Long-term, low-demand housing that targets people experiencing homelessness who are resistant to staying in conventional shelter facilities and unwilling to participate in supportive services. Although these accommodations are long-term, its residents are still considered homeless Southeastern Pennsylvania Transportation Authority (SEPTA): Regional public transportation authority that operates bus, rapid transit, commuter rail, light rail, and electric trolleybus services in and around Philadelphia Sex Workers Outreach Project (SWOP): National advocacy organization to help prevent violence against sex workers TANF/HB/MAGI—is a Medicaid eligibility category based upon eligibility for the Temporary Assistance for Needy Families (TANF) or Health Beginnings (HB) program, or qualifying based on income falling below the modified adjusted gross income (MAGI) criteria; households qualifying for Medicaid under this category include low-income families, children, pregnant women, and childless adults without disabilities that qualify them for Supplemental Security Income (SSI) Women Against Abuse (WAA): Domestic violence shelter and advocacy organization for women who’ve been subject to multiple forms of domestic violence in Philadelphia Women Organized Against Rape (WOAR): Non-profit organization dedicated to preventing all forms of sexual violence and supporting women who’ve been the victims of sexualized forms of violence through treatment services, special programs, and advocacy for those who’ve experienced sex-based violence 98