Risk Environments of Permanent Supportive Housing for Formerly Incarcerated People with Serious Mental Illnesses: A Protocol for a Novel Mixed Methods Feasibility Study

Abstract

Background: Formerly incarcerated people with serious mental illnesses (SMI) experience the criminal legal system unequally and have elevated rates of recidivism, homelessness, general medical problems, and substance use disorders. Permanent supportive housing (PSH) can be used during reentry, but it has limited resources for addressing community integration, a key component of reentry. PSH are often located in high-poverty environments with increased criminogenic risk. The geography of PSH also includes public spaces, which are associated with positive outcomes. The risk environment framework provides a structure for understanding the geography of PSH through its focus on the physical, social, economic, and policy influences on the micro and macro environments of reentry.

Methods: This is a novel QUAL + QUAN (spatial) concurrent mixed-methods study that will examine how individual, interpersonal, and environmental factors interact with public and private spaces to inform reentry wellbeing. Eighty multimethod interviews (i.e., qualitative, quantitative, and participatory mapping methods) will be conducted with formerly incarcerated clients with SMI. Go-along interviews will be conducted with 20 of these participants. Participatory mapping will be geocoded and sites identified as places of importance, frequent participation, and belonging will be evaluated in relation to objective features of spaces to develop a community resilience index. Findings will ultimately be integrated into an intervention development codesign process with a community advisory board.

Discussion: During reentry, individual, interpersonal, and environmental factors can interact with these environments to produce or reduce risk. If addressed, these factors can contribute to reentry wellbeing, through improved community participation and treatment engagement and reduced psychiatric distress and substance use.

Keywords

risk environment framework permanent supportive housing reentry serious mental illness mixed methods spatial methods

Background

Among people in prison, the prevalence of homelessness is twenty times higher than the general population, resulting in housing insecurity upon release (Bashir et al., 2021). Consequently, housing interventions, like permanent supportive housing (PSH), are increasingly supplementing or even substituting for reentry interventions by focusing on reducing legal involvement among residents (Hunter & Mercier, 2023; Kriegel et al., 2016; LeBel, 2017). PSH is a housing intervention that provides permanent housing through an apartment or unit and case management, medical, mental health, and substance use services (Henwood et al., 2013). Like most reentry interventions for people with serious mental illnesses (SMI), PSH primarily focuses on meeting the clinical and service needs of people with SMI and substance use disorders with largely positive outcomes. There is substantial evidence indicating that PSH improves housing outcomes and reduces hospitalization and ER visits (Rog et al., 2014).

Although PSH independently provides resources that can be essential to the reentry process, they are not necessarily responsive to the unique needs of formerly incarcerated people with SMI (Fontaine, 2013; Hamlin & Purser, 2021; LeBel, 2017). Relative to formerly incarcerated people without SMI, people with SMI have longer incarcerations, higher recidivism rates, lower likelihood of getting and staying on community supervision as a condition of early release, and less time in the community (Cloyes et al., 2010; O’Keefe & Schnell, 2007; Prins, 2011; Skeem et al., 2011). They also have elevated rates of homelessness, medical problems, and substance use disorders (Baillargeon et al., 2010; Baranyi et al., 2022; Bashir et al., 2021; Bronson, 2017; Bronson & Berzofsky, 2017; Cuddeback et al., 2010; Roman & Travis, 2016). Reentry interventions provide a critical anchor in the absence of coordinated discharge planning, often beginning during an incarceration and continuing through the transition and integration of a formerly incarcerated person with SMI.

During reentry, housing interventions like PSH require additional strategies to function as standalone reentry programs aimed at reducing recidivism and improving mental health outcomes during reentry (Leclair et al., 2019). This is in part because PSH are often located in high-risk environments (e.g., high-crime, high-poverty neighborhoods) where its formerly incarcerated residents are more heavily surveilled and also surrounded by spaces they are often mandated to avoid (Tsai et al., 2011). In addition, despite provision of housing and mental health services, PSH does not address the community integration of its residents, a critical component of reentry (Motte-Kerr et al., 2020). Reentry programming is significant not only as a touchpoint for necessary resources and social services, but also in its potential provision of support to people who are re-engaging with their communities following extended incarcerations (Blank Wilson, 2013; Wilson et al., 2017). While PSH do not systematically address this type of community integration, community integration is an inevitable extension of housing. Housing is situated in a larger neighborhood environment with both risks and benefits, unavoidable to its residents, and also, potentially resourceful.

Extant research indicates that community context, like the environments surrounding PSH, influences intervention success (Barrenger & Draine, 2013; Barrenger et al., 2016). If interventions operate in risky environments (e.g., an area with higher crime), their impact may be diminished without supplemental intervention focused on the environment itself. Likewise, resource density plays a role in the types of services and linkages that can ultimately be made accessible to community members if an intervention is case management oriented and reliant on external treatment and service providers. For people with SMI, riskier, resource-poor environments can broadly impact access to treatment and opportunities for support and community participation.

Although PSH are typically situated in high-risk environments, they are also surrounded by public spaces that can be resourceful and supportive for formerly incarcerated people, people with SMI, and people who are homeless (Fields, 2011; Herbert & Beckett, 2010; Kriegel, 2019; Kriegel et al., 2019, 2023). Quality public space is associated with a sense of community and can contribute to improved physical and emotional wellbeing (Cattell et al., 2008; Francis et al., 2012; Gesler, 1992; Snethen et al., 2021). Formerly incarcerated people with SMI face cumulative stigma related to their SMIs and incarceration histories, prior institutionalizations, and policies preventing their full community participation and sustaining a sense of carceral citizenship (Miller & Stuart, 2017; Rowe & Baranoski, 2000). Given this, public space engagement and accessibility is particularly important for this population, though often hampered by restrictive and/or exclusion policies (Herbert & Beckett, 2010; Miller & Stuart, 2017; Mitchell, 1995; Stuart, 2016). Formerly incarcerated people with SMI, particularly those who are or have been homeless, rely on public spaces because they provide social interaction and a sense of belonging (Herbert & Beckett, 2010; Kriegel, 2019; Kriegel et al., 2023). They also use these spaces to build capital, find social support, and meet basic needs through access to computers, internet, outlets, and a safe space to spend the day (Desmond, 2012; Fields, 2011; Kriegel et al., 2023). In a recent study on the use of public space during reentry, formerly incarcerated participants with mental illnesses described seeking out strangers in public spaces, with whom they developed rapport and trust and participated in supportive interactions, with little expectation for reciprocity (Kriegel, 2019). Participants also used non-clinical public spaces for solitude, socialization, and resource acquisition (Kriegel et al., 2023). These spaces collectively demonstrate an opportunity for PSH to not simply address the housing and treatment needs of its residents, but to explore their unique community reentry needs of those residents who are formerly incarcerated and have SMIs.

Theoretical Framework

The risk environment framework provides a structure for understanding community reentry and has been used to inform the development of structural interventions addressing health disparities (Rhodes & Simic, 2005). Risk environments are the spaces where individual, interpersonal, and structural/environmental factors interact to increase the chance of a certain risks like drug use, HIV transmission, or reincarceration (Barrenger et al., 2017; Henwood et al., 2018; Rhodes & Simic, 2005). They have two dimensions (i.e., types and levels) and are shaped by two mechanisms of influence (i.e., susceptibility and vulnerability; see Figure 1). The different types include physical environments (e.g., public spaces, PSH), social environments (e.g., social networks), economic environments (e.g., cost of living), and policy/political environments (e.g., federal housing policies, municipal homelessness policy priorities). Each type of risk environment exists at the macro environmental level (i.e., broader structures and policies informing local practices) and the micro environmental level (i.e., personal decisions or community norms). As an example, a micro social environment might include everyday social interactions while a macro social environment would include both structural stigma and marginalization. While conceptualized through its elements, a risk environment is the totality of these various types, levels, and mechanisms of influence.

Figure 1.

Risk Environment Framework.

During reentry, production of risk in these environments is confounded by the presence of SMI and histories of homelessness, both of which inform susceptibility and vulnerability to different risks. Both populations are, for example, more susceptible to reincarceration and more vulnerable to the consequences of incarcerations (Herbert et al., 2015; Matejkowski & Ostermann, 2015; Rhodes, 2002). Risk environment analyses, conducted using ethnographic methodologies, have examined both reentry for people with SMI and PSH for adults engaged in high risk behaviors (Barrenger et al., 2017; Henwood et al., 2018). In both cases, risk environment analyses situated the role and impact of community context therefore creating opportunities for multilevel intervention targets that are both responsive to community context and optimize the perspectives of impacted community members through participatory methods (Agurs-Collins et al., 2019; Blankenship et al., 2006; Minkler & Wallerstein, 2011).

Risk environment factors interact in both private and public spaces to contribute to community participation, treatment engagement, psychiatric distress, and substance use, which we collectively define as reentry wellbeing in the present study (see Figure 2). PSH are both private and public spaces and are consequently well-situated as a platform for interventions focused on improving reentry wellbeing by targeting the interplay of individual, interpersonal and structural factors. Yet, the current approach to understanding risk environments through ethnographic methods alone limits the reach of these examinations because risk environments are both measurable and meaningful. Ethnographies study the subjectively meaningful relationship that people have to their spatial environments, but require prolonged engagement for often undetermined periods of time and resources (Atkinson et al., 2001; Small, 2009). Short term ethnographic methods, like go-along interviews, provide an alternative, but are neither participatory nor do they capitalize on geospatial analysis through GIS, and therefore benefit from complementary methods to enhance their findings. Triangulating data through participatory and spatial methods is thus critical to optimizing risk environment analyses to examine the broader geography of PSH, including both private and public spaces, and their measurable and meaningful impacts on reentry wellbeing through individual, interpersonal, and structural factors. With this information in hand, it is possible to develop more impactful reentry interventions and better facilitate reentry wellbeing.

Figure 2.

Reentry Wellbeing for People with SMI.

To date, public space and interactions within have not been investigated empirically to understand the reentry outcomes of formerly incarcerated people with SMI, including those in PSH. Extant research often relies on singular methods to examine reentry; however, reentry is both complex and dynamic and informed by both individual and structural processes. Given this research gap, this study incorporates GIS, participatory mapping, semi-structured interviews, and go-along interviews to apply a risk environment framework to PSH for people with SMI during reentry and support the development of a targeted intervention. In doing so, we hope to achieve a multi-angular understanding of what reentry looks like through measurable data (e.g., GIS data) and meaningful experiences captured through qualitative and participatory methods. In the final phase of the study, we will employ participatory methods, through use of a Generative Co-Design Framework, to co-design a multilevel intervention framework with both data collected and the lived experiences of participants and providers. Participatory methods, used during data collection through participatory mapping and in this final phase, are methods that prioritize community voice and action in the research process (Minkler & Wallerstein, 2011; Stringer, 1996). Through participatory methods, we leverage the contribution of people with lived experiences in the data we collect and analyze.

Methods

Study Design

The study uses a rigorous, novel QUAL + QUAN (spatial) concurrent mixed-methods design to examine how individual, interpersonal, and environmental factors situated in the PSH risk environments interact in public and private spaces to inform reentry wellbeing (see Figure 2; Palinkas, 2014). The study will use a convergent parallel design combining quantitative and qualitative data in equal measure to confirm/disconfirm respective findings (i.e., triangulation), enhance and elaborate findings of each method (i.e., complementarity), and to expand on independent findings to create nuance (i.e., expansion; Creswell & Clark, 2017).

Using the risk-environment framework, 80 multi-method interviews (i.e., qualitative, quantitative, and participatory mapping methods) that look at individual, interpersonal, and environmental factors will be conducted with formerly incarcerated participants with SMI (40 people per site). To increase rigor, data collected through these interviews will be triangulated with go-along interviews with up to 20 participants (10 per site; Padgett, 2011). Participatory mapping will be geocoded and sites identified as places of importance, frequent participation, and belonging will be evaluated in relation to features of spaces like resource and treatment availability using GIS to develop a community resilience index. The findings will be presented to a community advisory board (CAB) of two providers and six formerly incarcerated PSH residents with SMI as part of a Generative Co-Design process (Bird et al., 2021; McKercher, 2020).

Recruitment and Eligibility

Eligibility criteria for the study requires participants (1) be an adult (≥18 year of age) residents in PSH in either Los Angeles, California or Spokane, Washington; (2) have a self-reported SMI diagnosis (schizophrenia, major depressive disorder, or bipolar disorder); (3) be released from jail or prison in the recent 5 years following a ≥6 month sentence; (4) be able to speak and understand English. Purposive sampling of formerly incarcerated people with SMI living in PSH will be used through flyers posted in PSH sites as well as word of mouth through program providers. Jail and prison incarcerations are included to ensure inclusion of people serving shorter sentences (less than 1 year) in jails rather than prisons. Those who express interest are screened through a brief 5-question screening tool and eligible participants provide consent to participate. Participants are compensated $40 for the initial multi-method interview and $40 for each subsequent go-along interview.

Study Sites

Data is being collected in Los Angeles, California (n = 40) and Spokane, Washington (n = 40). There are currently over 22,000 PSH units in Los Angeles and 500 in Spokane (Los Angeles Homeless Services Authority, 2024; U.S. Department of Housing and Urban Development, 2023). Access to PSH is based on measures of vulnerability that prioritize people with mental illnesses and consequently, PSH primarily serves people with SMI (Gilmer et al., 2015). Los Angeles and Spokane were selected as a method of critical case sampling, which aims to show variations that emerge through very different samples (Patton, 1990). Los Angeles is a highly urban setting with both racial/ethnic and socioeconomic diversity and is facing a significant housing crisis. Spokane is a mid-sized city that has seen a 46% increase in homelessness in the past 10 years (Spokane City/County Continuum of Care, 2020). It is tucked amid rural counties and American Indian Reservations and has a more racially and ethnically homogenous population. Sample variation derived from the differences between Spokane and Los Angeles will help to resolve feasibility concerns in support of future development and evaluation of an intervention across diverse urban and rural environments.

Measures and Materials

The study is undergirded by the Risk Environment Framework and its analytic axis examines how individual, interpersonal, and environmental characteristics interact in private and public spaces to facilitate reentry wellbeing. To evaluate these characteristics, the measures described below will be used as part of a baseline semi-structured interview. Data collection methods are informed by the ethnographic shadowing template developed by for a prior risk environment analysis of PSH (Henwood et al., 2018). We will examine reentry wellbeing through (1) community participation; (2) treatment engagement; (3) psychiatric distress; and (4) substance use.

Measures of Individual Characteristics

We will use a brief sociodemographic survey, modeled after surveys used in similar studies to identify sociodemographic characteristics. Questions on the survey will cover age, sex, gender, sexual orientation, race and ethnicity, education, homelessness history, mental health history, substance use history, criminal legal history.

Measures of Interpersonal Characteristics

Interpersonal characteristics are defined through social support (i.e., participants’ perceptions of the functions performed by members of their social networks) and social capital (the quality and quantity of resources in participants’ networks; Lin, 2002; Valente, 2010). Social support will be measured using the Medical Outcomes Study (MOS) Social Support Survey, which assesses social support through a 19-item questionnaire (Sherbourne & Stewart, 1991). The survey is divided into four subscales: emotional/information, tangible, affectionate, and positive social interaction. Participants indicate how often each type of support is available to them using a 5-point rating scale (e.g., 1 = none of the time and 5 = all of the time). Social Capital will be measured using the Resource Generator-United States (US), which assesses social capital through a 26-item scale (Foster & Maas, 2016). The survey has three subscales on expert access, problem solving, and personal support. Participants are asked whether they know someone who can provide different listed resources and to then specify relation. The measure captures the human, cultural, financial, political, and physical capital domains of social capital. The Resource Generator-US is adapted from prior scales used in the Netherlands, UK, and Canada for a US audience.

Measures of Environmental Characteristics

Environmental characteristics are defined through community integration (i.e., physical, social, and psychological integration into any defined community), public space engagement (i.e., the use of public spaces to access resources and interpersonal connections), and housing environment. Public space engagement is also defined subjectively through participants’ appraisals of importance, sense of belonging, frequency of contact in each space and objectively using GIS to examine resource and treatment availability/accessibility. To measure public space engagement, geocoded variables derived from participatory mapping data (described in further detail below) will be categorized into three areas: number of places of belonging, number of places of importance, and number of areas of frequent contact. This method of geocoding participatory mapping variables is adapted from Townley and colleagues who used a similar method to understand how homeless youth experience sense of community in their activity spaces (Townley et al., 2016). Community integration will be measured using the 10-item Community Integration Measure (McColl et al., 2001; Townley et al., 2013). The scale looks at feelings of acceptance and opportunities for community participation using a 5-point rating scale (i.e., “always disagree to “always agree”). Housing environment will be measured through the Housing Environment Survey, which assesses the housing environment using a social ecological approach through seven scales and three inventories, including two physical environment scales (i.e., dwelling physical quality and neighborhood physical quality); two social environment scales (i.e., neighborhood safety and neighborhood social climate); three interpersonal relationship scales (i.e., neighbors, roommates, and landlords); and three inventories on housing demographics, housing costs, and residential satisfaction (Kloos & Shah, 2009).

Measures for Reentry Wellbeing Outcomes

For this study, reentry wellbeing is defined through measures of community participation, treatment engagement, psychiatric distress, and substance use. Community participation will be measured using the 31-item Temple University Community Participation (TUCP) Scale (Salzer et al., 2014, 2015). The TUCP asks participants to indicate number of days in the prior 30 they participated in 27 different areas (e.g., going a restaurant or coffee shop), whether they perceive their participation in that area to be important, whether they have participated in that area as much as they desire. Treatment engagement will be measured using the Continuity of Care Measure (CONNECT), which focuses on continuity of care, a critical element of treatment access and provision of people transitioning between systems (e.g., criminal legal, mental health, and housing systems; Ware et al., 2003). The 72-item survey uses a 5-point rating scales to assess a person’s perception of their service providers’ knowledge and availability to client; flexibility; attention to primary care; movement from inpatient to outpatient care; changes in provider; and changes in housing. Psychiatric distress will be measured through the Modified Colorado Symptom Index (MCSI), a 14-item self-report scale that assesses frequency of psychiatric symptoms (Conrad et al., 2001). Responses are rated using a 5-point rating scale (0-not at all to 5-at least every day). Substance use will be assessed using two measures. The Alcohol Use Disorders Identification Test (AUDIT) is a 10-item self-report scale that assesses severity of alcohol use within the recent year, with a total score range of 0 to 40 (Saunders et al., 1993). The Drug Abuse Screening Test-10 (DAST-10) is a brief 10-item scale that assess drug use in the recent year (Skinner, 1982).

Semi-Structured Interview

Semi-structured interviews will focus on attributes of the physical, social, economic, and political environments detailed by the framework and the individual, interpersonal, and environmental/structural factors. Prompts will ask participants to (1) narrate their reentry experiences; (2) describe where they live, including neighborhood and in-house services and security; (3) describe supports and challenges experienced following release as well as since moving into PSH; and (4) describe where they spend time and the people with whom they spend time. Probes will focus on where participants feel most at ease, most welcomed, and most unwelcomed as well as the people with whom people feel at ease, spend time, and provide and receive different types of support. Broad qualitative prompts permit participant-led descriptions of a problem and interviewer probes ensure the expanse of a question is addressed (Creswell & Poth, 2016).

Participatory Mapping Protocol

Participatory mapping will be used to examine place-based interactions and participant-driven descriptions of community spaces and icons (Parker, 2006). Participatory mapping incorporates elements of inclusion, transparency, and empowerment and has been used in previous research with people with SMI (Townley et al., 2009). Participants will be asked to identify (1) spaces they frequent; (2) spaces they feel are important; and (3) spaces of inclusion and exclusion. To understand the social context of these spaces, participants will be prompted to list any interactions that typically take place in the spaces mapped.

Go-Along Interviews

Up to 10 go-along interviews will be conducted at each site (n = 20 participants). The final number of participants will be determined by saturation of data (i.e., congruence between interview data and go-along data). Go-along interviews combine qualitative interviews with ethnographic observations to understand participants’ experiences and practices as they interact with social and physical environments. During these interviews, interviewers accompany individuals on their natural outings. They are usually 90 min long, but participants ultimately determine length of time that makes sense for the outing (Kusenbach, 2003). The interviewer might follow up on specific interactions and spaces mentioned during the prior interview. Following each interview, the interviewer will record field notes using a field note template, developed by Henwood and colleagues (Henwood et al., 2018).

Spatial Data

Spatial data will include public and participatory mapping data. Public data will be gathered from federal and local government agency websites, including Census.gov, the County of Los Angeles Enterprise Geographic Information Systems, the California state geoportal, City of Spokane Open GIS data, and the Washington Geospatial Open Database. Mortality and crime data will be purchased through state agencies. Collectively, these public servers have data on mental health and substance use programs and centers and other public services, including police stations, courthouses, and libraries. Using methods employed by Townley and colleagues, we will geocode participatory maps, which involves the conversion of addresses to longitude and latitude coordinates (Townley et al., 2016). Addresses will be identified using Google maps through information provided by participants. If participants identify private spaces outside PSH sites (e.g., someone’s home), we will ask participants to identify the closest major intersection to ensure privacy is protected. Participatory mapping data will be parsed into variables based on interview data, including: (1) spaces of belonging; (2) spaces of importance; and (3) frequently accessed spaces.

Analytic Approach

Qualitative Analysis

A risk environment analysis will be conducted through qualitative analyses in two phases. The first phase will include inductive grounded theory methods (e.g., open, axial, and selective coding; Glaser & Strauss, 1967; Strauss & Corbin, 1994). Constant comparative analysis through memo writing will be used as part of the coding process to systematically identify similarities and differences across all interview sources, including semi-structured interviews, participatory maps, and go-along interviews, and to organize and synthesize the large amount of data collected (Glaser & Strauss, 1967; Strauss & Corbin, 1994). Initially, two members of the research team will independently code an initial transcript and then compare results to reach consensus on the codebook. They will then independently code four transcripts using developed codebook and compare code assignments by passage or quote. Discrepancies will be resolved through consensus, and themes will be developed from coded material. The second phase of coding will focus on the different types of environments and influences detailed through the risk environment framework using a template analytic approach (Crabtree & Miller, 1992). Template analysis is particularly useful when using an existing conceptual framework because it allows for the use of a priori codes (e.g., macro environment and micro environment, physical, social, economic, and policy environments, susceptibility factors, vulnerability factors, public spaces, and private spaces). The triangulation of interview data, participatory mapping data, and go-along interview data provides an added strategy of rigor to identify weaknesses and strengths in using our proposed measure of risk environments. ATLAS.ti qualitative software will be used to code transcribed data (ATLAS.Ti Scientific Software Development GmbH, 2020).

Quantitative Analysis

We will summarize baseline characteristics, including age, sex, gender, sexual orientation, race and ethnicity, education, homelessness history, mental health history, substance use history and criminal legal history. Quantitative data collected reflects the risk environment framework and seeks to evaluate relationships demonstrated in Figure 2. We will conduct simple linear regression to determine the association between each interpersonal and environmental measure and study outcome. As the data allows, we will investigate multivariable regression models that include all independent variables to assess the independent effect of each variable on each outcome. All modeling will be performed in SPSS (IBM, 2020) or R statistical software (R Core Team, 2020), as applicable. Both qualitative and quantitative analyses will consider the impact of gender and racial/ethnic identities on the experience of study outcomes given women and BIPOC communities navigate risk environments differently because of social and structural inequalities (Nkansah-Amankra et al., 2013; Spjeldnes & Goodkind, 2009). Quantitative and qualitative findings will be integrated and merged during this analytic phase of the study in order to provide a conceptual framework to inform intervention development (Fetters et al., 2013).

Community Resilience Index Development

Spatial data collected from participants will be geocoded and then analyzed in two ways. First, we will use data collected to develop a community resilience index model. Second, we will conduct a cluster analysis. Conventionally, we use vulnerability indices to characterize and quantify susceptibility to illness or public health hazard (Amram et al., 2020). For this study, we aim to develop a model for an index that can characterize and quantify opportunities for community engagement for formerly incarcerated people with SMI who live in high-risk environments, like PSH. These opportunities can support public space engagement and community integration, which, we hypothesize, help to facilitate reentry wellbeing. Because of the novelty of our index and the limited evidence supporting the integration of objective and subjective measures of space, we aim to develop a model for an index first. Once refined, a community resilience index can identify physical and social spaces of risk reduction and production to support a risk environment analysis. Because we are collecting this data in two predominately urban areas, we will focus on density through availability of these spaces (i.e., how many of these types of spaces exist within a 3-mile buffer around the PSH sites). To develop a model for a community resilience index, we will use unadjusted and adjusted multilevel modeling to examine individual, interpersonal, and environmental indicators as well as density of parks, libraries, coffee shops and restaurants, entertainment centers (e.g., movie theaters), retail spaces, food banks/soup kitchens, and treatment providers (e.g., community mental health centers, substance use treatment programs) and their association with spaces of belonging, importance, and frequency of use (as reported by participants during interviews). To further determine how the relationship between these indicators and the three outcomes generalize to unseen data through a community resilience index, we will utilize leave-one-out cross validation to evaluate fitted models via R^2 and root mean square error (RMSE). A cluster (i.e., hotspot) analysis will subsequently be conducted using the Getis G function in ESRI ArcGIS to identify geographic clusters of two evidenced risk factors affecting formerly incarcerated people with SMI: (1) areas of arrest, clustered by crime and (2) drug overdose, clustered by drug type (Jacobs & Skeem, 2021). The Getis G statistic is used to quantify and assess links between spatially distributed variables (Getis & Ord, 2010). This analysis will allow the research team to identify and map ‘hot’ and ‘cold’ spots of arrests and drug overdoses. To support the risk environment analysis, we will then do an overlay with the community resilience index model to examine the relationship between community resilience (i.e., belonging, frequency of use, and importance in relation to treatment and resource access and provision) and significant characteristics of the risk environment. All geocoding and GIS analyses will be conducted using ArcGIS Pro (Esri, 2020).

Mixed Method Integration

Qualitative, quantitative, and spatial data will be merged after independent analyses are completed for the purpose of convergence, complementarity, and expansion (Creswell & Clark, 2017). First, through convergence, we can confirm or disconfirm findings by triangulating each form of data. Secondarily, we will use qualitative data to provide nuance and depth to quantitative and spatial data. Quantitative and spatial data help to establish relationships between variables, but unless we open the black box with qualitative data, we cannot know the mechanisms by which they work. Finally, we will use qualitative, quantitative, and spatial data to expand on individual findings. This final step allows us to create a more comprehensive and inductive understanding of a complicated theory like the Risk Environment framework. Community-based research is often messy and individual data methods provide limited insight into complex problems. Through mixed methods, we aim to create a nuanced portrait that help in the co-design of a multilevel reentry intervention that supports formerly incarcerated people with SMI living in PSH.

Ethics

In consideration of ethical standards, informed consent will be obtained in writing from all potential participants prior to participation. Potential participants will be told about the study’s purpose and all procedures involved. Study staff will administer the University of California, San Diego Brief Assessment for Capacity to Consent (UBACC), a brief screener of capacity to consent to research procedures designed for adults with psychiatric illness (Jeste et al., 2007). Those who cannot demonstrate capacity to provide informed consent on this screener will be administered the MacArthur Competency Assessment Tool for Clinical Research (MacCAT-CR), a comprehensive, gold standard tool for assessing capacity to provide informed consent for adults with SMI (Appelbaum & Grisso, 2001). Only people with a demonstrated capacity to provide informed consent on these measures will be enrolled in the study.

Direct identifiers will be removed from qualitative data, and identifiers maintained for the purpose of follow-up with go-along interview participants (e.g., email addresses, phone number, name) will be stored in a password-protected document on a password-protected computer separate from participant data. Only study staff will have access to survey data, participatory maps, audio recordings, and transcripts. During audio-recorded interviews, we will instruct the interviewee not to mention their name or anyone else’s name. In the case that someone’s name is mentioned, identifiers will be removed or altered in transcription. Completed data forms will be placed in locked file cabinets at every site, and only authorized staff members will have access to the data. All study personnel will participate in extensive training in consenting procedures and will be required to complete Institutional Review Board, HIPAA, and compliance training.

Rigor

To achieve trustworthiness (including credibility, transferability, and confirmability) as well as evidentiary adequacy, we will incorporate multiple strategies of rigor (Erickson, 1986; Lincoln & Guba, 1985). The research team will document context notes following interviews as well as site notes to detail any observations of PSH sites both during recruitment and data collection (i.e., audit trail). Context notes are qualitative and ask interviewers to (1) describe participants’ overall wellbeing during the interview, including whether they are socially engaged or withdrawn, their level of functioning, and any evidence of mental health symptoms; (2) describe the location and place of the interview, including physical descriptions (e.g., layout, people who occupied the space during the interview, location condition, meaningful interactions at the location, and any background on the importance of the location to the participant, if relevant; and (3) describe any notable moments during the interview, including any interactions the participant had with other people, the nature of the relationship, or apparent reactions of others to the researcher. In addition, we have embedded member checking, peer debriefing and support, data and methodological triangulation, as well as prolonged engagement at various phases of the study. Phase 3 includes a CBPR-informed intervention co-design process with people with lived experience as well as providers during which findings from qualitative interviews and any initial analyses are reviewed and considered for future intervention development (i.e., member checking). The use of go-along interviews allows for both prolonged engagement across twenty participants to reduce risk of reactivity or respondent bias. Prolonged engagement can also increase risk of researcher bias, which we hope to offset through triangulation of data (i.e., qualitative data, participatory mapping data, and field notes), which can improve reactivity, respondent bias, as well as researcher bias. The research team will finally meet at regular intervals to discuss both data collection methods and emergent findings (i.e., peer debriefing and support).

Conclusion

Public space and interactions within have not been investigated empirically to understand the reentry outcomes of formerly incarcerated people with SMI, particularly those in PSH. This study uniquely employs multiple methods including GIS, participatory mapping, semi-structured interviews, and go-along interviews, to apply a risk environment framework to PSH for people with SMI during reentry. In doing so, we can co-design a targeted intervention that considers the relationship between public and private spaces and the individual, interpersonal, and environmental factors in risk environments that support reentry wellbeing. The findings from this study are well-positioned to consider the variability of risk environments—through geography and embedded features including rurality—and its impact on reentry wellbeing. Doing so will also resolve feasibility concerns in support of larger studies oriented toward intervention development, implementation, and evaluation as well as replication of this study in a larger sample across urban and rural contexts.

Footnotes

ORCID iD

Liat S. Kriegel

Ethics Approval

The Risk Environments of Permanent Supportive Housing for Formerly Incarcerated People with Serious Mental Illnesses study was approved by Washington State University IRB (#19786).

Informed Consent

Informed consent to participate will be obtained from all participants.

Authors Contributions

LK is the principal investigator of the study. The remaining authors are co-investigators and reviewed the manuscript. The authors read and approved the final manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The current study was funded by an award from the National Institute of Mental Health (NIMH) (K01MH129482, PI: Kriegel). The funding agencies had no role in the design of this study and will not have any role during its execution, analyses, interpretation of the data, or decision to submit results.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Anonymized data from the study will be available from the authors after the study has ended.*

Appendix

References

Agurs-Collins

Persky

Paskett

E. D.

Barkin

S. L.

Meissner

H. I.

Nansel

T. R.

Arteaga

S. S.

Zhang

Das

Farhat

(2019). Designing and assessing multilevel interventions to improve minority health and reduce health disparities. American Journal of Public Health, 109(S1), S86–S93. https://doi.org/10.2105/AJPH.2018.304730

Amram

Amiri

Lutz

R. B.

Rajan

Monsivais

(2020). Development of a vulnerability index for diagnosis with the novel coronavirus, COVID-19, in Washington State, USA. Health & Place, 64, 102377. https://doi.org/10.1016/j.healthplace.2020.102377

Appelbaum

P. S.

Grisso

(2001). MacCAT-CR: MacArthur competence assessment tool for clinical research. Professional Resource Press.

Atkinson

Coffey

Delamont

Lofland

(2001). Handbook of ethnography. Sage.

ATLAS.ti scientific software development GmbH (version 9). (2020). [Computer software].

Baillargeon

Hoge

S. K.

Penn

J. V.

(2010). Addressing the challenge of community reentry among released inmates with serious mental illness. American Journal of Community Psychology, 46(3), 361–375. https://doi.org/10.1007/s10464-010-9345-6

Baranyi

Fazel

Langerfeldt

S. D.

Mundt

A. P.

(2022). The prevalence of comorbid serious mental illnesses and substance use disorders in prison populations: A systematic review and meta-analysis. The Lancet. Public Health, 7(6), e557–e568. https://doi.org/10.1016/S2468-2667(22)00093-7

Barrenger

S. L.

Draine

(2013). “You don’t get no help”: The role of community context in effectiveness of evidence-based treatments for people with mental illness leaving prison for high risk environments. American Journal of Psychiatric Rehabilitation, 16(2), 154–178. https://doi.org/10.1080/15487768.2013.789709

Barrenger

S. L.

Draine

Angell

Herman

(2017). Reincarceration risk among men with mental illnesses leaving prison: A risk environment analysis. Community Mental Health Journal. https://doi.org/10.1007/s10597-017-0113-z

10.

Barrenger

S. L.

Kriegel

L. S.

Angell

Draine

(2016). Role of context, resources, and target population in the fidelity of critical time intervention. Psychiatric Services, 67(1), 115–118. https://doi.org/10.1176/appi.ps.201400108

11.

Bashir

A. Y.

Moloney

Elzain

M. E.

Delaunois

Sheikhi

O’Donnell

Dunne

C. P.

Kelly

B. D.

Gulati

(2021). From nowhere to nowhere. Homelessness and incarceration: A systematic review and meta-analysis. International Journal of Prisoner Health, 17(4), 452–461. https://doi.org/10.1108/IJPH-01-2021-0010

12.

Bird

McGillion

Chambers

E. M.

Dix

Fajardo

C. J.

Gilmour

Levesque

Lim

Mierdel

Ouellette

Polanski

A. N.

Reaume

S. V.

Whitmore

Carter

(2021). A generative co-design framework for healthcare innovation: Development and application of an end-user engagement framework. Research Involvement and Engagement, 7(1), 12. https://doi.org/10.1186/s40900-021-00252-7

13.

Blankenship

K. M.

Friedman

S. R.

Dworkin

Mantell

J. E.

(2006). Structural interventions: Concepts, challenges and opportunities for research. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 83(1), 59–72. https://doi.org/10.1007/s11524-005-9007-4

14.

Blank Wilson

(2013). How people with serious mental illness seek help after leaving jail. Qualitative Health Research, 23(12), 1575–1590. https://doi.org/10.1177/1049732313508476

15.

Bronson

(2017). Drug use, dependence, and abuse among state prisoners and jail inmates, 2007–2009. 27.

16.

Bronson

Berzofsky

(2017). Indicators of mental health problems reported by prisoners and jail inmates, 2011-12 (No. NCJ 250612, pp. 1–16). U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics.

17.

Cattell

Dines

Gesler

Curtis

(2008). Mingling, observing, and lingering: Everyday public spaces and their implications for well-being and social relations. Health & Place, 14(3), 544–561. https://doi.org/10.1016/j.healthplace.2007.10.007

18.

Cloyes

K. G.

Wong

Latimer

Abarca

(2010). Time to prison return for offenders with serious mental illness released from prison: A survival analysis. Criminal Justice and Behavior, 37(2), 175–187. https://doi.org/10.1177/0093854809354370

19.

Conrad

K. J.

Yagelka

J. R.

Matters

M. D.

Rich

A. R.

Williams

Buchanan

(2001). Reliability and validity of a modified Colorado symptom index in a national homeless sample. Mental Health Services Research, 3(3), 141–153. https://doi.org/10.1023/A:1011571531303

20.

Crabtree

B. F.

Miller

W. L.

(1992). A template approach to text analysis: Developing and using codebooks. In Crabtree

B. F.

Miller

W. L.

(Eds.), Doing qualitative research (pp. 93–109). Sage Press.

21.

Creswell

J. W.

Clark

V. L. P.

(2017). Designing and conducting mixed methods research. Sage Publications.

22.

Creswell

J. W.

Poth

C. N.

(2016). Qualitative inquiry and research design: Choosing among five approaches. Sage Publications.

23.

Cuddeback

G. S.

Scheyett

Pettus-Davis

Morrissey

J. P.

(2010). General medical problems of incarcerated persons with severe and persistent mental illness: A population-based study. Psychiatric Services (Washington, D.C.), 61(1), 45–49. https://doi.org/10.1176/ps.2010.61.1.45

24.

Desmond

(2012). Disposable ties and the urban poor. American Journal of Sociology, 117(5), 1295–1335. https://doi.org/10.1086/663574

25.

Erickson

(1986). Qualitative methods in research on teaching. In Handbook of research on teaching (3rd ed., pp. 119–161). Macmillan.

26.

Esri . (2020). ArcGIS (Version 10.8.1) [Computer software].

27.

Fetters

M. D.

Curry

L. A.

Creswell

J. W.

(2013). Achieving integration in mixed methods designs—principles and practices. Health Services Research, 48(6 Pt 2), 2134–2156. https://doi.org/10.1111/1475-6773.12117

28.

Fields

(2011). Emotional refuge? Dynamics of place and belonging among formerly homeless individuals with mental illness. Emotion, Space and Society, 4(4), 258–267. https://doi.org/10.1016/j.emospa.2010.09.002

29.

Fontaine

(2013). The role of supportive housing in successful reentry outcomes for disabled prisoners. Cityscape, 15(3), 53–76.

30.

Foster

K. A.

Maas

C. D.

(2016). An exploratory factor analysis of the resource generator-United States: A social capital measure. British Journal of Social Work, 46(1), 8–26. https://doi.org/10.1093/bjsw/bcu111

31.

Francis

Giles-Corti

Wood

Knuiman

(2012). Creating a sense of community: The role of public space. Journal of Environmental Psychology, 32(4), 401–409. https://doi.org/10.1016/j.jenvp.2012.07.002

32.

Gesler

W. M.

(1992). Therapeutic landscapes: Medical issues in light of the new cultural geography. Social Science & Medicine, 34(7), 735–746. https://doi.org/10.1016/0277-9536(92)90360-3

33.

Getis

Ord

J. K.

(2010). The analysis of spatial association by use of distance statistics. In Anselin

Rey

S. J.

(Eds.), Perspectives on spatial data analysis (pp. 127–145). Springer. https://doi.org/10.1007/978-3-642-01976-0_10

34.

Gilmer

T. P.

Stefancic

Henwood

B. F.

Ettner

S. L.

(2015). Fidelity to the housing first model and variation in health service use within permanent supportive housing. Psychiatric Services, 66(12), 1283–1289. https://doi.org/10.1176/appi.ps.201400564

35.

Glaser

B. G.

Strauss

A. L.

(1967). The discovery of grounded theory: Strategies for qualitative research.

36.

Hamlin

Purser

(2021). “A program, not the projects”: Reentry in the post-public housing era. Journal of Contemporary Ethnography, 50(6), 806–834. https://doi.org/10.1177/08912416211017270

37.

Henwood

B. F.

Cabassa

L. J.

Craig

C. M.

Padgett

D. K.

(2013). Permanent supportive housing: Addressing homelessness and health disparities? American Journal of Public Health, 103(Suppl 2), S188–S192. https://doi.org/10.2105/AJPH.2013.301490

38.

Henwood

B. F.

Lahey

Harris

Rhoades

Wenzel

S. L.

(2018). Understanding risk environments in permanent supportive housing for formerly homeless adults. Qualitative Health Research, 28(13), 2011–2019. https://doi.org/10.1177/1049732318785355

39.

Herbert

C. W.

Morenoff

J. D.

Harding

D. J.

(2015). Homelessness and housing insecurity among former prisoners. The Russell Sage Foundation Journal of the Social Sciences: RSF, 1(2), 44–79. https://doi.org/10.7758/RSF.2015.1.2.04

40.

Herbert

Beckett

(2010). ‘This is home for us’: Questioning banishment from the ground up. Social & Cultural Geography, 11(3), 231–245. https://doi.org/10.1080/14649361003637661

41.

Hunter

S. B.

Mercier

(2023). Addressing homelessness among people with justice involvement: Los Angeles county’s just in reach pay for success demonstration project. Cityscape, 25(2), 57–72.

42.

IBM . (2020). SPSS statistics (Version 26).

43.

Jacobs

L. A.

Skeem

J. L.

(2021). Neighborhood risk factors for recidivism: For whom do they matter? American Journal of Community Psychology, 67(1–2), 103–115. https://doi.org/10.1002/ajcp.12463

44.

Jeste

D. V.

Palmer

B. W.

Appelbaum

P. S.

Golshan

Glorioso

Dunn

L. B.

Kim

Meeks

Kraemer

H. C.

(2007). A new brief instrument for assessing decisional capacity for clinical research. Archives of General Psychiatry, 64(8), 966–974. https://doi.org/10.1001/archpsyc.64.8.966

45.

Kloos

Shah

(2009). A social ecological approach to investigating relationships between housing and adaptive functioning for persons with serious mental illness. American Journal of Community Psychology, 44(3–4), 316–326. https://doi.org/10.1007/s10464-009-9277-1

46.

Kriegel

L. S.

(2019). Stranger support: How former prisoners with mental illnesses navigate the public landscape of reentry. Health & Place, 56, 155–164. https://doi.org/10.1016/j.healthplace.2019.01.021

47.

Kriegel

L. S.

Barrenger

S. L.

Henwood

B. F.

(2023). Public space as a potential resource during reentry for formerly incarcerated people with serious mental illnesses. Health and Place, 84, 103–133. https://doi.org/10.1016/j.healthplace.2023.103133

48.

Kriegel

L. S.

Henwood

B. F.

Gilmer

T. P.

(2016). Implementation and outcomes of forensic housing first programs. Community Mental Health Journal, 52(1), 46–55. https://doi.org/10.1007/s10597-015-9946-5

49.

Kriegel

L. S.

Townley

Brusilovskiy

Salzer

M. S.

(2019). Neighbors as distal support for individuals with serious mental illnesses. American Journal of Orthopsychiatry.

50.

Kusenbach

(2003). Street phenomenology: The go-along as ethnographic research tool. Ethnography, 4(3), 455–485. https://doi.org/10.1177/146613810343007

51.

LeBel

T. P.

(2017). Housing as the tip of the iceberg in successfully navigating prisoner reentry. Criminology & Public Policy, 16(3), 889–908. https://doi.org/10.1111/1745-9133.12313

52.

Leclair

M. C.

Deveaux

Roy

Goulet

M.-H.

Latimer

E. A.

Crocker

A. G.

(2019). The impact of housing first on criminal justice outcomes among homeless people with mental illness: A systematic review. Canadian journal of psychiatry. Revue canadienne de psychiatrie, 64(8), 525–530. https://doi.org/10.1177/0706743718815902

53.

Lin

(2002). Social capital: A theory of social structure and action. Cambridge University Press.

54.

Lincoln

Y. S.

Guba

E. G.

(1985). Naturalistic inquiry. Sage.

55.

Los Angeles Homeless Services Authority . (2024). 2024 Housing Inventory Count (HIC)- Los Angeles Continuum of Care (CoC). https://www.lahsa.org/documents?id=8162-2024-housing-inventory-count.xlsx

56.

Matejkowski

Ostermann

(2015). Serious mental illness, criminal risk, parole supervision, and recidivism: Testing of conditional effects. Law and Human Behavior, 39(1), 75–86. https://doi.org/10.1037/lhb0000094

57.

McColl

M. A.

Davies

Carlson

Johnston

Minnes

(2001). The community integration measure: Development and preliminary validation. Archives of Physical Medicine and Rehabilitation, 82(4), 429–434. https://doi.org/10.1053/apmr.2001.22195

58.

McKercher

K. A.

(2020). Beyond sticky notes: Co-design for real: Mindsets, methods and movements. Beyond Sticky Notes.

59.

Miller

R. J.

Stuart

(2017). Carceral citizenship: Race, rights and responsibility in the age of mass supervision. Theoretical Criminology, 21(4), 532–548. https://doi.org/10.1177/1362480617731203

60.

Minkler

Wallerstein

(2011). Community-based participatory research for health: From process to outcomes. John Wiley & Sons.

61.

Mitchell

(1995). The end of public Space?People’s park, definitions of the public, and democracy. Annals of the Association of American Geographers, 85(1), 108–133. https://doi.org/10.1111/j.1467-8306.1995.tb01797.xa

62.

Motte‐Kerr

W. L.

Rhoades

Henwood

Rice

Wenzel

(2020). Exploring the association of community integration in mental health among formerly homeless individuals living in permanent supportive housing. American Journal of Community Psychology, 66(1–2), 3–13. https://doi.org/10.1002/ajcp.12459

63.

Nkansah-Amankra

Agbanu

S. K.

Miller

R. J.

(2013). Disparities in health, poverty, incarceration, and social justice among racial groups in the United States: A critical review of evidence of close links with neoliberalism. International Journal of Health Services: Planning, Administration, Evaluation, 43(2), 217–240. https://doi.org/10.2190/HS.43.2.c

64.

O’Keefe

M. L.

Schnell

M. J.

(2007). Offenders with mental illness in the correctional system. Journal of Offender Rehabilitation, 45(1–2), 81–104. https://doi.org/10.1300/J076v45n01_08

65.

Padgett

D. K.

(2011). Qualitative and mixed methods in public health. Sage publications.

66.

Palinkas

L. A.

(2014). Qualitative and mixed methods in mental health services and implementation research. Journal of Clinical Child and Adolescent Psychology, 43(6), 851–861. https://doi.org/10.1080/15374416.2014.910791

67.

Parker

(2006). Constructing community through maps? Power and praxis in community mapping. The Professional Geographer, 58(4), 470–484. https://doi.org/10.1111/j.1467-9272.2006.00583.x

68.

Patton

M. Q.

(1990). Qualitative evaluation and research methods (2nd ed, p. 532). Sage Publications, Inc.

69.

Prins

S. J.

(2011). Does transinstitutionalization explain the overrepresentation of people with serious mental illnesses in the criminal justice system? Community Mental Health Journal, 47(6), 716–722. https://doi.org/10.1007/s10597-011-9420-y

70.

R Core Team . (2020). A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/

71.

Rhodes

(2002). The “risk environment”: A framework for understanding and reducing drug-related harm. International Journal of Drug Policy, 13(2), 85–94. https://doi.org/10.1016/S0955-3959(02)00007-5

72.

Rhodes

Simic

(2005). Transition and the HIV risk environment. BMJ, 331(7510), 220–223. https://doi.org/10.1136/bmj.331.7510.220

73.

Rog

D. J.

Marshall

Dougherty

R. H.

George

Daniels

A. S.

Ghose

S. S.

Delphin-Rittmon

M. E.

(2014). Permanent supportive housing: Assessing the evidence. Psychiatric Services, 65(3), 287–294. https://doi.org/10.1176/appi.ps.201300261

74.

Roman

C. G.

Travis

(2016). Taking stock: Housing, homelessness, and prisoner reentry. https://www.urban.org/research/publication/taking-stock

75.

Rowe

Baranoski

(2000). Mental illness, criminality, and citizenship. The Journal of the American Academy of Psychiatry and the Law, 28(3), 262–264.

76.

Salzer

M. S.

Brusilovskiy

Prvu-Bettger

Kottsieper

(2014). Measuring community participation of adults with psychiatric disabilities: Reliability of two modes of data collection. Rehabilitation Psychology, 59(2), 211–219. https://doi.org/10.1037/a0036002

77.

Salzer

M. S.

Kottsieper

Brusilovskiy

(2015). Intermethod reliability and factors affecting recall with the Temple University Community Participation measure. Journal of Mental Health, 24(4), 189–195. https://doi.org/10.3109/09638237.2015.1036976

78.

Saunders

J. B.

Aasland

O. G.

Babor

T. F.

Fuente

J. R. D. L.

Grant

(1993). Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction, 88(6), 791–804. https://doi.org/10.1111/j.1360-0443.1993.tb02093.x

79.

Sherbourne

C. D.

Stewart

A. L.

(1991). The MOS social support survey. Social Science & Medicine, 32(6), 705–714. https://doi.org/10.1016/0277-9536(91)90150-B

80.

Skeem

J. L.

Manchak

Peterson

J. K.

(2011). Correctional policy for offenders with mental illness: Creating a new paradigm for recidivism reduction. Law and Human Behavior, 35(2), 110–126. https://doi.org/10.1007/s10979-010-9223-7

81.

Skinner

H. A.

(1982). The drug abuse screening test. Addictive Behaviors, 7(4), 363–371. https://doi.org/10.1016/0306-4603(82)90005-3

82.

Small

M. L.

(2009). ‘How many cases do I need?’: On science and the logic of case selection in field-based research. Ethnography, 10(1), 5–38. https://doi.org/10.1177/1466138108099586

83.

Snethen

Jeffries

Thomas

Salzer

(2021). Welcoming places: Perspectives of individuals with mental illnesses. The American Journal of Orthopsychiatry, 91(1), 76–85. https://doi.org/10.1037/ort0000519

84.

Spjeldnes

Goodkind

(2009). Gender differences and offender reentry: A review of the literature. Journal of Offender Rehabilitation, 48(4), 314–335. https://doi.org/10.1080/10509670902850812

85.

Spokane City/County Continuum of Care . (2020). 5-Year strategic plan to prevent and end homelessness 2020–2025 (p. 49). https://static.spokanecity.org/documents/ending-homelessness/about/2020-2025-strategic-plan-to-prevent-and-end-homelessness.pdf

86.

Strauss

Corbin

(1994). Grounded theory methodology: An overview. In Handbook of qualitative research (pp. 273–285). Sage Publications, Inc.

87.

Stringer

E. T.

(1996). Action research: A handbook for practitioners (pp. xxi, 169). Sage Publications, Inc.

88.

Stuart

(2016). Down, out, and under arrest: Policing and everyday life in skid row. University of Chicago Press.

89.

Townley

Kloos

Wright

P. A.

(2009). Understanding the experience of place: Expanding methods to conceptualize and measure community integration of persons with serious mental illness. Health & Place, 15(2), 520–531. https://doi.org/10.1016/j.healthplace.2008.08.011

90.

Townley

Miller

Kloos

(2013). A little goes a long way: The impact of distal social support on community integration and recovery of individuals with psychiatric disabilities. American Journal of Community Psychology, 52(1–2), 84–96. https://doi.org/10.1007/s10464-013-9578-2

91.

Townley

Pearson

Lehrwyn

J. M.

Prophet

N. T.

Trauernicht

(2016). Utilizing participatory mapping and GIS to examine the activity spaces of homeless youth. American Journal of Community Psychology, 57(3–4), 404–414. https://doi.org/10.1002/ajcp.12060

92.

Tsai

Mares

A. S.

Rosenheck

R. A.

(2011). A geographic analysis of chronically homeless adults before and after enrollment in a multi-site supported housing initiative: Community characteristics and migration. American Journal of Community Psychology, 48(3–4), 341–351. https://doi.org/10.1007/s10464-010-9363-4

93.

U.S. Department of Housing and Urban Development . (2023). HUD 2023 Continuum of care homeless assistance programs housing inventory count report. https://files.hudexchange.info/reports/published/CoC_HIC_CoC_WA-502-2023_WA_2023.pdf

94.

Valente

T. W.

(2010). Social networks and health: Models, methods, and applications. Oxford University Press.

95.

Ware

N. C.

Dickey

Tugenberg

McHorney

C. A.

(2003). Connect: A measure of continuity of care in mental health services. Mental Health Services Research, 5(4), 209–221. https://doi.org/10.1023/A:1026276918081

96.

Wilson

A. B.

Barrenger

S. L.

Brusilovskiy

Draine

Salzer

M. S.

(2017). Community participation among individuals with serious mental illnesses leaving jail. Journal of Psychosocial Rehabilitation and Mental Health, 4(1), 45–51. https://doi.org/10.1007/s40737-016-0074-5