Scoping review of interventions to link individuals to substance use services at discharge from jail

Abstract

Introduction:

Individuals with substance use disorders (SUD) must be linked to community-based SUD treatment and other services upon their release from jail, given their high service needs and risks for relapse, recidivism, and opioid-related overdose following release.

Method:

This scoping review identified 14 studies (28 affiliated publications) that used experimental or quasi-experimental designs to evaluate jail re-entry interventions for individuals with SUD. The team coded intervention components, study characteristics, and study outcomes based on a service continuum for treatment linkage and retention and for post-release substance use and criminal justice outcomes.

Results:

This review included 4 randomized controlled trials (RCT) for linkage to treatment with medications for opioid use disorder (MOUD); 4 RCTs and 4 quasi-experimental studies for linkage to non-specific SUD treatment; and 2 RCTs for linkage to HIV + SUD services. Most studies (9/14) used case management and/or peer or patient navigation as the core intervention; 2 studies provided medical management for MOUD induction and/or facilitated referral to MOUD in the community; and 3 studies used motivational-based linkage interventions. A qualitative analysis of study outcomes found evidence to support the effectiveness of a diverse range of interventions to link individuals to community-based SUD treatment, MOUD, and other services at re-entry, but limited support for intervention effects on longer-term outcomes, including treatment retention, medication adherence, recidivism, and substance use.

Conclusion:

Future controlled trials and implementation studies should help to unpack and examine core components of jail re-entry interventions and their successful implementation to enhance treatment retention and improve post-release outcomes.

1. Introduction

Individuals with substance use disorders (SUD) have a high likelihood of experiencing legal problems and involvement with the criminal justice system (Moore, Oberleitner, et al., 2020). In 2019, 10.3 million admissions to jails occurred in the United States, with an average daily population of 734,500 individuals, including individuals who were sentenced or awaiting trial or disposition (Zeng & Minton, 2021). A majority (63%) of sentenced individuals in jail met criteria for a drug use disorder from 2007 to 2009; apart from marijuana, rates of use in the month prior to arrest were highest for cocaine/crack, methamphetamine, and heroin/opioids (Bronson, Stroop, Zimmer, & Berzofsky, 2017). Moreover, only about 10% of individuals in jail receive any kind of treatment, counseling, or self-help participation prior to their release (Bronson et al., 2017). Given the high volume of individuals that are detained in jails, jails represent a critical venue in which to intervene with individuals with SUD by increasing their engagement in treatment services while confined and helping to connect them to services in the community following their release, potentially strengthening continuity of care.

The surge in opioid-related overdose has increased exponentially the need to intervene with individuals with opioid use disorders (OUD), who have high rates of criminal-justice involvement (Winkelman, Chamg, & Binswanger, 2018) and extremely high risk of overdose following their discharge to the community. Individuals with OUD typically have contact with community service providers prior to or after their discharge from jail to the community, but they lack access to sustained interventions that may avert relapse and overdose. Alex et al. (2017) examined the rate of death among detainees within 6 weeks following their release from New York City jails in 2011–2012. Opioid overdose accounted for the highest share of deaths among the 59 decedents (37%), exceeding deaths due to other drug overdose, chronic disease, and assaultive or other forms of trauma. Moreover, 77% of those who died from opioid overdose had a history of prior overdose or opioid detoxification, 14% were released with a methadone dose, and half (50%) were referred to opioid treatment within the community at release. Thus, low rates of in-jail methadone dosing as well as having received a referral to methadone treatment in the community did not protect individuals from opioid overdose. Similarly, Hacker et al. (2018) examined service system contacts preceding opioid overdose fatality among 1399 individuals in Alleghany, PA, from 2008 to 2014. A majority of the sample (68%) had a public human service encounter preceding their overdose death; more than half of these (55.5%) had been incarcerated in the county jail, and 26% of those incarcerated in the year prior to their death overdosed within 30 days of their last release from jail. Using data from the Massachusetts Public Data Warehouse, Larochelle et al. (2019) determined that individuals who had been released from prison or jail had 30 times the risk of death from opioid overdose compared to those in the general population, controlling for age and gender. Finally, Victor, Zettner, et al. (2021) analyzed factors associated with fatal opioid overdose among all individuals who were released from jail in one county in Indiana in 2017–2019; each prior booking increased the risk of mortality by approximately 20% and a syringe-related charge at most recent booking prior to release more than tripled the risk. Thus, much more needs to be done to reduce the risks of relapse, overdose, and death by intervening with individuals prior to, during, and following their discharge from jail.

Yet individuals leaving jail face numerous barriers to accessing medications for opioid use disorder (MOUD), overdose prevention, and other vital behavioral health, physical health, and social services. These barriers include a pervasive lack of inter-organizational relationships between jails and community providers (Grella, Ostlie, Scott, Dennis, & Carnevale, 2020), lack of insurance to pay for treatment, logistical challenges, and resistance to referring individuals to SUD treatment, particularly with MOUD, among community corrections (Mitchell et al., 2016), all of which hinder continuity of care. Several studies have found that individuals with OUD who are treated with MOUD while incarcerated are more likely to engage in treatment following release (Moore et al., 2019). In a recent retrospective study, about 40% of a sample of men with OUD admitted to two jails in Connecticut continued methadone treatment while incarcerated. Moreover, those who continued methadone treatment in jail were three times as likely as those who discontinued treatment of resuming methadone treatment in the community after release, which in turn reduced their odds of overdose and reincarceration (Hass et al., 2021). Hence, continuity of treatment from jail to community may be essential for reducing relapse, overdose, and recidivism.

Research has proposed and/or tested several mechanisms for increasing access to SUD treatment and other needed services following discharge from jail. Such mechanisms include initiation of treatment in jail, linkage to designated community providers, and team-based interventions to help individuals navigate community service systems (Hu et al., 2020). Some interventions utilize peers who share similar backgrounds and experiences and may variously function as peer advocates, peer mentors, health navigators, or recovery coaches to support individuals in sustaining recovery following their release (Bassuk, Hanson, Greene, Richard, & Laudet, 2016; Satinsky et al., 2020; Stanojlović & Davidson, 2021). Others use professional or paraprofessional case managers or patient navigators to assist individuals in negotiating complex health and social service systems, or combine elements of these different intervention models (Mitchell et al., 2021).

Currently, a need exists to better understand the types of jail re-entry interventions for individuals with SUD that have been studied. Areas of interest include the target of the intervention; training and qualifications of individuals who deliver the interventions; intervention components and duration; their success in linking individuals to SUD treatment and other services; and outcomes regarding substance use, treatment retention, recidivism, among others. While a well-developed body of research exists on interventions to link people living with HIV/AIDS from jail to community services (Laufer, Arriola, Dawson-Rose, Kumaravelu, & Rapposelli, 2002), with some focus on linkage to ancillary SUD services due to high rates of substance use among this population (Molitor et al., 2005; Woznica et al., 2021), the opioid overdose crisis has shifted the focus to linking individuals with substance use problems at discharge to community-based SUD treatment, including treatment for OUD. This paper reports findings from a scoping review of the literature from the past 20 years on interventions that aim to link individuals to community SUD services upon discharge from jail to evaluate the state of evidence in this area of research, gaps in knowledge, and areas that are in need of future research.

1.1. Study aim and rationale

This study uses the framework of a service cascade to evaluate outcomes related to jail re-entry interventions to link individuals to SUD/MOUD treatment and other services. The HIV care continuum originally provided a useful framework for evaluating interventions that “assist individuals in accessing needed care (linkage), facilitate continuing involvement in care (retention), promote compliance with medication regimens (medication adherence), and for individuals who drop out of care, encourage them to reenter care (reengagement)” (Rapp et al., 2013, p. S90). More recently, research has used the OUD treatment cascade to evaluate the sequential process by which individuals engage in treatment for OUD, particularly MOUD, and to identify gaps in service delivery (Socias et al., 2018; Williams, Nunes, Bisaga, & Olfson, 2018). Similarly, research has applied the service cascade framework to evaluate access to substance use services for justice-involved youth and incorporates a focus on re-entry services to link youth to behavioral health services in the community (Belenko et al., 2017; Scott, Dennis, Grella, Funk, & Lurigio, 2019). This review evaluates studies that reported on any of the following aspects of a re-entry service cascade and associated outcomes: treatment linkage, retention, medication adherence, and re-engagement; criminal behavior or justice involvement; or substance use outcomes.

The aim of this study is to conduct a scoping review of the published literature to identify studies that evaluated interventions to link individuals to SUD services at the point of community re-entry from jail and to describe these interventions, associated study characteristics, and outcomes. A scoping review utilizes the same search procedures as those used in systematic reviews, but does not include a quantitative synthesis of data across studies. Instead, a scoping review is appropriate to assess a heterogeneous set of studies that used different methodological approaches to address a common theme (Tricco et al., 2018), to “provide an overview or map of the evidence” (Munn et al., 2018). A qualitative synthesis is used to assess the range of studies and nature of their findings, and to identify common themes, areas of concurrence, and research gaps. This approach was considered appropriate to the current review given the diverse goals of the linkage interventions, including the target populations and intervention components, and the range of outcomes reported. The current scoping review was guided by the following research questions:

1. What are characteristics of jail re-entry interventions that aim to link individuals to substance use treatment and other services at release from jail?

2. What are postrelease outcomes for jail re-entry interventions regarding: a) treatment linkage, retention, medication adherence, or re-engagement; b) criminal justice; and/or c) substance use?

2. Methods

2.1. Study design

The review was informed by established methods for conducting and reporting systematic and scoping reviews, as articulated in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Liberati et al., 2009; Moher, Liberati, Tetzlaff, Altman, & the PRISMA Group, 2009) and the PRISMA Extension for Scoping Reviews (PRISMA ScR; Tricco et al., 2018). The team used a qualitative synthesis to ascertain what current literature says about the nature and extent of interventions that aim to link jail re-entry populations to SUD services within the community as it pertains to research gaps, common themes, and intervention strategy typology.

2.2. Eligibility criteria

This review included peer-reviewed publications of studies conducted in the United States. The study team limited the search to studies published between 2001 and 2021. We selected this window because it is a feasible period of time, and also extended the time frame of another recent systematic review of SUD interventions for individuals leaving either jail or prison (Moore, Hacker et al., 2020). This prior review examined publications from 2007 to 2017, which predates many of the recent interventions that focus on linking individuals with OUD to community-based MOUD treatment following their release from jail. Moreover, the current review includes studies of re-entry interventions for individuals with HIV/AIDS who had high rates of SUD and included linkage to SUD services. The review included studies that:

• Included an intervention that aimed to assist individuals in their re-entry to the community at release through linkage or active referral to community SUD providers, inclusive of both non-specific SUD treatment and treatment with MOUD. The review defined “active referral” as more than usual care or a simple referral (i.e., giving people a list of local SUD treatment providers). The study team included interventions that focus exclusively on SUD services or include them as part of a broader array of services.

• Included a measure of substance use among the sample or the sample is selected on the basis of substance use.

• Used an experimental or quasi-experimental design to compare groups on postrelease outcomes regarding treatment linkage or retention, medication adherence, criminal justice status, or substance use following release.

The study team excluded articles based on the following criteria:

• Not in English;

• Published prior to 2001;

• Not in a peer-reviewed journal;

• Does not focus specifically on a jail re-entry population or combined jail and prison re-entry population;

• Does not describe or evaluate an intervention designed to help individuals transition from jail to community-based SUD services or report a measure of baseline substance use or postrelease outcomes;

• Focuses only on a youth or juvenile population¹;

• Did not collect primary data from study sample; and/or

• Is a systematic review, literature review, commentary, opinion piece, clinical trial protocol, or methodology paper.

2.3. Search strategy

The research team conducted an electronic literature search of published papers in PubMed and PsycInfo. Two reviewers worked on the search that took place the week of January 25, 2021. The lead project manager then reviewed all results across reviewers and provided feedback for consistency. The team used two sets of search terms: one set pertaining to interventions (e.g., linkage, re-entry; 22 terms) and one set pertaining to criminal justice terms (3 terms [PubMed]/1 terms [PsychInfo]). We searched in combination each of the intervention terms with each of the criminal justice terms for a total of 66 search term pairs searched across each of the databases identified.

The team imported all initial search results into Zotero and an Excel spreadsheet for review. Two reviewers used a four-tiered search and review process: 1) they initially screened search results for duplication across databases and results were unduplicated; 2) then they screened all records for inclusion based on title and abstract; 3) then they conducted full-text review based on inclusion criteria on remaining articles; and 4) once they completed the full-text review, they reviewed article reference lists to determine if any additional articles existed that met inclusion criteria. Using steps 1–3 described above, they reviewed articles found in the reference list. See Fig. 1 for the flow chart of the search results, based on the PRISMA criteria (Tricco et al., 2018); the PRISMA checklist is in Appendix A and the sample search terms are in Appendix B.

Fig. 1.

Re-entry interventions to link individuals to substance use services at discharge from jail scoping review flow chart.

2.4. Information collected

Two reviewers abstracted data on article and study characteristics and entered these into a centralized Excel database using the following parameters; 1) study identification (e.g., author[s], year of publication, full citation); 2) study characteristics (e.g., aim, research design, setting); 3) sample characteristics (e.g., socio-demographics, criminal justice status, substance use); 4) results (e.g., study findings on treatment participation, criminal justice involvement, and/or substance use); and 5) intervention characteristics (e.g., intervention type, components, and delivery). The team completed the abstraction review on April 23, 2021, and it included articles published online prior to in-print publication at that time. We completed abstraction of articles found during the reference list review on July 13, 2021.

2.5. Analysis

The analysis for this paper is based on findings from 14 unique studies (with 28 affiliated publications) that met the inclusion criteria. We use study as the unit of analysis in the description of interventions, rather than individual papers, because multiple papers may have been published from one “parent” study that report on different outcomes or time frames or a separate qualitative component that provided more in-depth information on the intervention and its implementation. First, we summarize characteristics of the intervention in each study, including: 1) primary target of the intervention: MOUD-specific treatment; SUD treatment, which may include MOUD, but not focused specifically on MOUD; and HIV services that include an SUD treatment component; 2) intervention components, including other services provided or linkage targets; 3) who delivers the intervention; and 4) frequency and duration of contacts with participants. Second, we summarized study characteristics, including study design (i.e., quasi-experimental or randomized control trial), sample size and characteristics, study location/setting, outcome measures and follow-up duration, and main study outcomes regarding SUD/MOUD treatment linkage, retention, or adherence; criminal behavior or justice system involvement; and substance use outcomes. The study team reviewed and discussed all codes to arrive at consensus.

3. Findings

3.1. Characteristics of included studies and re-entry interventions

The 14 included studies were all conducted in the United States. The majority (n = 8) included both males and females, ranging from 51% to 89% male, with a mean of 75.2%. Three studies had all-male samples, two had all-female samples, and one included men and transgender women.

Table 1 displays characteristics of the 14 studies by the primary intervention target, primary intervention mechanism and other components, staff who conducted the intervention, and frequency and duration of contacts.

Table 1.

Characteristics of jail re-entry linkage interventions.

Study/papers	Primary linkage target	Primary intervention mechanism	Intervention components	Who conducts intervention	Frequency of contacts/duration
• Bahr, Harris, Strobell, & Taylor, 2012	SUD	Case mgt, CBT	Employment, GED, life skills	Certified drug tx counselors/CM	Pre-release: 100 h/4 weeks; Post-release: transfer to community tx or work release
• Braga, Piehl, & Hureau, 2009	SUD	Mentoring/case mgt	Transition accountability plan; health, MH, social, employment, education, health insurance/benefits, housing, transportation, ID, clothing	Faith-based mentors from “same neighborhoods”/jail case workers/community providers	Pre-release: meet with jail-staff caseworkers and community mentors; met at jail discharge; Post-release: ave 7.3 contacts/39.7 h contact w/mentors over 12–18 mos
• Cunningham et al., 2018 • Goodman-Meza et al., 2019 • Harawa, Amani, Bowers, Sayles, & Cunningham, 2017	HIV+ SUD	Peer navigation	Goal-setting, skill-training to address stigma & discrimination, health, MH, social support, accompany participant to medical visits & communicate with clinicians, transportation, subsistence needs	Peer navigators	In-jail meeting and met at release; 12 PN sessions sessions/24 weeks
• Farabee, Condon, Hallgren, & McGrady, 2020 • Farabee et al., 2016	MOUD	Medical mgt/patient navigation	Referrals to community services (not specified)	Physician; patient navigator	Pre-release: initiation of XT-NTX, PN visit in jail; Post-release: PN contact in person or by phone at least once per week; less frequent contact in mos 2–3, goal of at least once per 2 weeks; Med mgt visits: 2 times/mo in mos 1–3 and monthly in mos 4–6
• Lee et al., 2015	MOUD	Medical mgt	Brief motivational counseling, relapse prevention, referrals to aftercare & 12-step/recovery support	Physician	Pre-release: screening & initial XR-NTX within one week of release
• McKenzie et al., 2012 • McKenzie, Macalino, McClung, Shield, & Rich, 2005	MOUD	In-jail MMT initiation; facilitate MMT continuation after release	Financial assistance for MMT for 24 weeks, individualized tx plan, ID & documents for tx, transportation assistance, HIV risk reduction, overdose prevention	Project staff	Pre-release counseling session; initiation of MMT pre-release or post-release referral
• Miller & Miller, 2010 • Miller, 2014	SUD	Case mgt/moral reconation therapy/CBT	Health, MH, 12-step/recovery, employment, GED, life skills, anger management	Pre-release: CM, SUD counselors; Post-release: parole/probation staff	Pre-release: jail in-reach, reentry accountability plan; Post-release: aftercare services
• Miller, Miller, & Barnes, 2016	SUD	Case mgt/cognitive therapy/family intervention (CRAFT)	Individual, group & family counseling, individualized treatment plan, reentry accountability plan, crisis intervention	Certified addiction specialist, CM, behavioral therapist, family counselor	Phase I medical stabilization; Phase II primary tx: 90 days; Phase III aftercare maintenance: ~60 days
• Myers et al., 2018 • Koester et al., 2014	HIV+ SUD	Patient navigation/enhanced case mgt	Motivational interviewing, HIV risk reduction, health, MH, social, employment, housing, food, legal, benefits, coaching, transportation, social support	Patient peer navigators; CM	Pre-release: discharge planning & court liaison; Meet at discharge; Post-release: monitor care plan
• Needels, James-Burdumy, & Burghardt, 2005 • Freudenberg, Daniels, Crum, Perkins, & Richie, 2005	SUD	Case mgt	Intensive discharge plan, family, health, crisis intervention, group & individual counseling, housing referrals	CM	Pre-release: group and individual sessions, discharge planning Post-discharge: community case mgt: ave 6.5 h contact with female clients, family, & others; ave 9.5 h contact with male clients in 12 mos post-release
• Prendergast, McCollister, & Warda, 2017 • Prendergast & Cartier, 2013	SUD	Screening, brief intervention, and referral to treatment	Motivational interviewing, CBT, HIV risk reduction	Screening & brief intervention (BI): health educators; Brief treatment (BT): counselors	Pre-release BI: 15–20 min; Post-release BT: 8 sessions –ave 23 min/each
• Schwartz et al., 2020 • Schwartz et al., 2021 • Kelly et al., 2020 • Mitchell et al., 2021	MOUD	Strengths-based motivational patient navigation/case mgt, with or without interim methadone	HIV risk reduction, social, ID, transportation assistance	Patient navigator/case managers	Pre-release: interim methadone, in-person meeting to develop post-release plan; Post-release: 8 or more case mgt sessions over 3 mos
• Scott & Dennis, 2012 • Scott, Dennis, & Lurigio, 2017 • Scott, Grella, Dennis, & Funk, 2016	SUD	Recovery management check-ups	Motivational interviewing with personalized feedback, skills training in self-management, transportation assistance, 12-step referral/recovery support, gender-specific HIV risk reduction	Linkage manager	Pre-release: assessment; Check-ups at release; 30, 60, 90 days post-release and quarterly to 36 mos; accompany participant to intake; engagement & retention protocol, i.e., phone or visit participant in 14 days after tx admission
• Stein, Caviness, Anderson, Hebert, & Clarke, 2010	SUD	Brief MI intervention	Feedback report & change plan, list of community resources & treatment programs, condoms	Clinical psychologist/MA-level counselor	Pre-release: one MI session; Post-release: one MI session at 1–3 mos

Legend: ave = average; CBT = cognitive behavioral therapy; case mgt = case management; CM = case manager; HIV = human immunodeficiency virus; medical mgt = medical management; MAT = medication-assisted treatment; MH = mental health; MI = motivational interviewing; MMT = methadone maintenance treatment; mos = months; MOUD = medication for opioid use disorder; OUD = opioid use disorder; PN = patient or peer navigator; SUD = substance use disorder; tx = treatment; XR-NTX = extended-release naltrexone.

3.1.1. Intervention target

Four studies examined re-entry interventions that were specific to MOUD, 8 examined re-entry interventions for SUD treatment, and 2 examined re-entry interventions primarily for HIV services that also included an SUD treatment component.

3.1.2. Primary intervention mechanism

Five of the studies utilized case management as a core intervention component; 4 utilized peer or patient navigation, 2 of these were combined with case management and one with medical management; 2 studies provided medical management for MOUD induction and/or facilitated referral to MOUD treatment in the community; and 3 studies used motivational-based interventions to refer or link individuals to SUD treatment. All of the studies initiated contact with participants in jail prior to release.

Although varying definitions and models of case management exist, which may depend on the context and population target, case management’s core components include referral and service coordination and, to varying degrees, assessment, planning, linkage, monitoring, and advocacy (Center for Substance Abuse Treatment, 2000; Osher, Steadman, & Barr, 2002). Patient navigation (PN) interventions similarly focus on helping patients to engage in services. Navigators may assist patients with scheduling medical appointments; accompany participants to medical appointments and facilitate communication with clinicians; address logistical problems, such as child care and transportation; and provide emotional support (Farabee et al., 2016). Elements of case management are also incorporated into PN interventions; for example, in Schwartz et al. (2021, 2020) patient navigators used strengths-based case management with motivational techniques. Among the HIV− focused interventions, peer-patient navigators met with HIV+ individuals while in jail, conducted an intensive assessment, psychoeducation, and discharge/case management plan in Myers et al. (2018). The patient navigators used a motivational interviewing harm-reduction approach to link individuals to a wide range of other services at discharge and accompanied study participants to medical visits (Koester et al., 2014). In the Cunningham et al. (2018) study, peer navigators provided a structured, 12-session intervention for HIV+ participants that was initiated in jail prior to discharge; the focus was on fostering social support and self-efficacy, serving as role models, and helping individuals to engage with HIV and other needed services.

Two studies using navigation-based interventions included qualitative interviews that provided more in-depth descriptions of the experiences of participants and staff conducting the interventions in these studies. Participants in the Schwartz et al. (2021, 2020) study described the complexity of interactions between participants and patient navigators. In particular, participants valued the patient navigator’s nonjudgmental caring and persistence, advocacy, help in brokering resources and managing interactions with the criminal justice system, and genuine support; yet several also described the tension between attempts to motivate and push them and, alternatively, to foster their self-determination (Mitchell et al., 2021). Similarly, qualitative interviews conducted in Myers et al. (2018) illustrate the importance of building connection and trust between the navigators and participants, based on their shared experiences related to HIV, incarceration, or drug use. In addition to helping to address clients’ basic needs, such as food and clothing, patient navigators served as role models for recovery and social functioning and helped to resolve problems that arise in attempts to access services or interact with the legal system (Koester et al., 2014).

Linkage interventions focused on assisting participants to access treatment in the community following their discharge by making contact with treatment programs; addressing logistical barriers, such as transportation, and motivational challenges; and assisting with the treatment admission process, such as obtaining needed documentation (McKenzie et al., 2005; Prendergast & Cartier, 2013; Scott & Dennis, 2012).

3.1.3. Other intervention components

Nearly all studies combined treatment linkage with an array of other intervention components. These typically included referrals to a wide range of services, including health services (5 studies; not including MOUD treatment or HIV-specific services), mental health services (4 studies), life skills or social services (5 studies), referral to shelters or other housing assistance (3 studies), employment/vocational assistance (4 studies), and GED or other education services (2 studies). Several helped participants to obtain identification cards or other documents needed to enter treatment or to obtain public assistance or health insurance. Specific therapeutic approaches included motivational interviewing or enhancement (5 studies), crisis intervention (2 studies), cognitive therapies (3 studies), anger management (1 study), Community Reinforcement and Family Training (CRAFT; 1 study), individual or group counseling (2 studies), and self-management training (1 study). Half (7) of the studies included HIV education or risk reduction as part of the intervention model, and 3 studies included overdose prevention/education (all with MOUD-related interventions) in the intervention and usual care. Four studies incorporated 12-step referral or recovery support.

In most studies (10/14) the re-entry process was guided by individualized treatment plans or discharge/postrelease transition plans from jail to the community. In 4 studies, individuals were met at the time of jail discharge and linked or escorted to community-based services. Transportation assistance to community services was provided in 6 studies and 2 studies covered the cost of medication for participants receiving MOUD.

3.1.4. Staff conducting the intervention

Considerable variation existed regarding the individual(s) who provided various intervention components and their level of training/expertise. Among studies that targeted MOUD treatment, medical staff administered MOUD and provided medical management, embedded with other intervention components (Lee et al., 2015); one used both physicians and patient navigators (Farabee et al., 2020); one combined patient navigators and case managers (Schwartz et al., 2020); and one used nonspecific project staff (McKenzie et al., 2012).

Staff delivering interventions focused on linkage to SUD treatment included drug treatment counselors and case managers (Bahr et al., 2012; Miller & Miller, 2010, 2015; Needels et al., 2005); peers or faith-based mentors from the “same neighborhoods” as study participants (Braga et al., 2009); paraprofessional linkage managers (Scott & Dennis, 2012); health educators or clinicians to provide brief motivational interventions (Prendergast & Cartier, 2013; Stein et al., 2010); and a team of case managers, behavioral therapists, and family counselors (Miller et al., 2016).

In studies that targeted HIV + SUD services, peer navigators were used in Cunningham et al. (2018) and peer-patient navigators, who were similar to the target population (i.e., HIV+ and had past histories of incarceration and substance use disorder), worked together with professional case managers in Myers et al. (2018). In contrast to patient navigators, who are not necessarily from the targeted population, peer navigators leverage their shared lived experience with participants to provide recovery support and serve as role models (Bassuk et al., 2016; Gonzalez, Rana, Jetelina, & Roberts, 2019).

3.2. Intervention outcomes

Table 2 summarizes findings using the service cascade framework for the following domains: 1) linkage, engagement, and retention in SUD treatment or MOUD adherence; 2) criminal justice outcomes; and 3) substance use outcomes.

Table 2.

Study designs and outcomes of jail re-entry linkage interventions.

Study/papers	Study design	Sample/setting	Outcome measures/follow-up duration	SUD/MOUD treatment linkage, retention, adherence	Criminal justice outcomes	Substance use outcomes
• Bahr et al., 2012	QE	Individuals who: 1) completed short-term intensive drug tx in jail (n = 70; 88.6% male) or 2) matched comparisons (n = 70, 88.6% male) in Utah, 2005	Recidivism/14–21 mos		Lower recidivism and fewer arrests in treatment vs. control group; Completion of in-jail treatment reduced risk of recidivism by half
• Braga et al., 2009	QE	Men in jail at high-risk for violent crime & recidivism in: 1) tx group (n = 108) or 2) comparison group matched with propensity score (n = 309) in Suffolk County, Boston, MA, 2002	Any arrests, violent arrests/up to 3 years		Treatment group had significantly fewer arrests for any new crime & violent arrests at 1, 2, & 3 years post-release
• Cunningham et al., 2018 • Goodman-Meza et al., 2019	RCT	HIV+ men (n = 303, 85%) and transgender women (n = 53, 15%) released from jail in: 1) PN group (n = 180) or 2) control group (n = 176) in Los Angeles County, CA, 12/2012–10/2016	HIV viral suppression, MAT and other health care visits, days of substance use/3, 6, 12 mos	Participants in PN group had a greater increase in number of MAT visits at 6 mos; control group had a greater increase at 12 mos		No significant group effect on substance use, although use declined substantially in both arms, particularly at 3 mos, and remained lower than baseline at 12 mos
• Farabee et al., 2020 • Farabee et al., 2016	RCT	Individuals with OUD (n = 135; 73% male) in: 1) XR-NTX (n = 46), 2) XR-NTX + PN (n = 45), or 3) ETAU (referral list of MOUD providers & fact sheet on opioid overdose; n = 44), in Metropolitan Detention Center, Bernalillo Co, New Mexico, 2017–2019	Opioid use, OUD treatment adherence, HIV risk, re-arrests (any, drug, felony)/1, 3, 6, 12 mos	Mean no. of injections received (out of 7): 2.26 for XR-NTX vs. 2.93 for XR-NTX + PN; 8% of participants received all 7 injections; no difference by group on medical mgt visits or no. of injections	No differences across groups in rates of rearrest; No difference across groups on a combined measured of no opioid use or incarceration	No differences across groups for any selfreported opioid use, % days using opioids, or OUD at 1, 3, 6, 12 mos; one group difference in opioid use at 12 mos was not significant when including missing cases as positive
• Lee et al., 2015	RCT	Men with OUD within one week prior to jail release in: 1) XR-NTX group (n = 17) or 2) enhanced usual care control: counseling & referral (no medication) (n = 17) in New York City, ½010–4/2013	Opioid relapse, urine tests negative for opioids, opioid abstinence, IDU, cocaine use, SUD tx participation, reincarceration, overdose/2, 3, 4, 8 weeks	15 of 17 initiated XR-NTX and 12/17 (75%) received second injection at week 4; No group differences in other community drug tx participation	No difference between groups in re-incarceration	Significantly better outcomes for XR-NTX vs. controls on: opioid relapse, opioid abstinence, & opioid negative urine tests at 1–4 & 1 –8 weeks No differences in rates of IDU, cocaine use, and overdose
• McKenzie et al., 2012 • McKenzie et al., 2005	RCT	Individuals with OUD and prior MMT tx (n = 90 at baseline, 71% male) in: 1) MMT initiation prerelease, continued MMT in community, & short-term financial assistance; 2) referral to MMT at release & short-term financial assistance; 3) referral to MMT at release without financial assistance in Rhode Island Dept of Corrections, 10/2006–2/2009	MOUD initiation within 30 days post-release, opioid and other drug use, IDU, other SUD tx, overdose, arrests or incarceration/6 mos	Initiation of MMT prior to jail release significantly increased treatment entry and reduced time-to-treatment entry postrelease No significant group differences in current MMT or other SUD treatment at 6 mos	No significant group differences in arrest or incarceration at 6 mos	Relapse to heroin use at 6 mos was significantly lower among pre-release initiation group vs. postrelease MMT referral + payment group & vs. MMT referral-only group; Cocaine use was lower in MMT pre-release group vs. other groups
• Miller & Miller, 2010, 2015	QE	Individuals w/SUD in jail with <18 mos to release in: 1) Re-entry program (cohort 1 n = 73, 80.8% male; cohort 2 n = 62, 79% male) or 2) matched comparison groups (cohort 1 n = 72, 80.8% male; cohort 2 n = 15, 50% male) in Auglaize County, OH in 2004; 2011–2013	Rearrest, probation violations, inmate-on-inmate altercation, inmate-on-staff altercation/12 mos		Recidivism rates were significantly lower among treatment vs. matched comparison groups in both study cohorts; In multivariate models, program participation significantly reduced rearrests (2010); program participation only reduced probation violations (2015)
• Miller et al., 2016	QE	Men who participated in jail SUD tx program who have minor children in: 1) treatment group (n = 34) or 2) comparison group (n = 32), in Delaware County, OH, 4/2012–12/2013	Probation revocation, new charge recidivism, time to ecidivism, compliance with child support orders/12 mos		Lower odds of new charge and any recidivism and increased time to recidivism in tx group vs. comparison, with stronger effects among program completers; No difference between groups for probation revocations.
• Myers et al., 2018	RCT	HIV+ individuals (n = 270; 81.5% male) with prior or current drug use in: 1) PN group (n = 137) or 2) TAU: discharge planning & 90 days case mgt (n = 133) in county jail in San Francisco, CA, 2010–2013	Linkage to and retention in HIV care, substance use, risky sexual behavior/2, 6, & 12 mos			No differences between groups in any alcohol or drug use or weekly any drug use in past 30 days at each follow-up
• Needels et al., 2005 • Freudenberg et al., 2005	RCT	Adult females (n = 704) and adolescent males (n = 706) who received: 1) jail-only services or 2) in-jail plus community services in New York City, 7/1997–5/2000	Substance use (by selfreport & hair analysis), criminal behavior, HIV risk behavior/12 mos	Women in experimental group had higher rates of post-release drug treatment vs. jail-only group; no group difference for adolescent males	Post-release serious arrests (i.e., murder, assault, robbery, burglary) were significantly lower at 12 mos among women in the experimental group vs. jail-only group; No difference between groups for adolescent males	No difference between groups in SUD outcomes among women; No difference between groups in SUD outcomes among adolescent males
• Prendergast et al., 2017; • Prendergast & Cartier, 2013	RCT	Individuals within 4 weeks of release from jail (n = 732, 73% male) in: 1) SBIRT group (n = 369) or 2) control group (n = 363) in Los Angeles County, CA, 12/2012–10/2014	Drug and alcohol use, SUD tx participation, rearrest, HIV risk behaviors, quality of life/12 mos	No differences between groups on SUD treatment or self-help participation	No differences between groups on rearrest	No differences between groups on substance use, including days after release to first use
• Schwartz et al., 2020 • Schwartz et al., 2021 • Mitchell et al., 2021 • Kelly et al., 2020	RCT	Recently arrested adults (n = 225, 80% male) with OUD in: 1) ETAU: 7-day medically supervised withdrawal, psychoeducation, referral helpline number, overdose prevention; 2) IM-only: methadone dosing while in jail and day after release, selection of community program; or 3) IM + PN in Baltimore City, MD Detention Center, 12/2014–10/2017	Enrollment in OUD tx 30 days post-release; opioid use, OUD, cocaine use, cocaine use disorder, illegal activity; cost effectiveness/1, 3, 6, 12, 24 mos	IM + PN group had higher initial rate of methadone tx entry vs. IM-only group, and both were higher than ETAU; no difference between IM groups in tx participation at 1 mo, although both were higher than ETAU; No difference across groups in any type of OUD tx or any SUD treatment at 12 mos	No significant group differences in likelihood of arrest, time to first arrest post-release, or severity of arrest charges at 12 mos; Being older and female were associated with lower arrest rates, longer time to first arrest, & less severe arrest at 12 mos; No significant group differences in self-reported criminal behavior or number of official arrests at 24 mos	No differences for either contrast in opioid-positive tests, although all groups reported a decrease in heroin at 12 mos; No group differences over the post-release period for opioid- or cocainepositive urine test results, self-reported opioid or cocaine use, or diagnosis of opioid or cocaine use disorders
• Scott & Dennis, 2012 • Scott et al., 2017 • Scott et al., 2016	RCT	Women who participated in SUD treatment in jail prior to release in: 1) RMC group (n = 238) or 2) control group (n = 242) in Cook County, Chicago, IL, 8/2008–4/2010	Time to SUD tx entry, days in SUD tx, substance use, illegal activity, arrests, incarceration, HIV risk behaviors/1, 2, 3 mos and quarterly up to 36 mos	RMC group vs. control group returned to treatment sooner and was more likely to participate in any SUD tx 90 days postdischarge; Probation status had stronger effects on SUD treatment participation over time	More SUD treatment & self-help participation, and less substance use were associated with less recidivism (arrests and days in jail); Probation status had negative effect on criminal behavior, rearrest & reincarceration	No group differences in relapse rates at 30, 60, or 90 days post-discharge; Receipt of any SUD treatment increased the odds of abstinence from alcohol and any drugs, marijuana, & cocaine, and opioids
• Stein et al., 2010	RCT	Women in jail/prison in: 1) MI group (n = 125) or 2) control (n = 120) in Rhode Island Dept of Corrections, 2/2004–6/2007	Alcohol use, adverse alcohol consequences, SUD tx entry or AA meetings attended/1, 3, 6 mos	Increase over time (3 & 6 mos post-release) in alcohol tx entry, but no difference between groups		More days abstinent and fewer alcohol-related problems in MI group vs. control at 3 mos; No group difference in number of drinks/drinking day at 3 mos or any alcohol outcomes at 1 & 6 mos

Legend: ave. = average; ETAU = enhanced treatment as usual; HIV = human immunodeficiency virus; IDU = injection drug use; IM = interim methadone; medical mgt = medical management; MAT = medication-assisted treatment; MI = motivational interviewing; MMT = methadone maintenance treatment; mos = months; MOUD = medication for opioid use disorder; mos = months; OUD = opioid use disorder; PN = patient or peer navigator; QE = quasi-experimental; RCT = randomized control trial; RMC = Recovery Management Check-up; SBIRT = screening, brief intervention, and referral to treatment; SUD = substance use disorder; TAU = treatment as usual; tx = treatment; XR-NTX = extended-release naltrexone.

3.2.1. Treatment linkage, engagement, and retention

Ten studies examined outcomes related to treatment linkage, engagement, retention, or adherence.

3.2.1.1. Linkage to MOUD.

Four RCTs examined treatment linkage among individuals who received pre-release interventions designed to increase MOUD initiation and engagement relative to control conditions. Farabee et al. (2020) and Schwartz et al. (2021, 2020) used similar interventions that embedded pre-release MOUD induction with PN that provided access to a range of services as well as continuing support for treatment engagement following release. Both studies found no significant differences between groups that received either extended-release naltrexone (XR-NTX) (Farabee et al., 2020) or interim methadone prior to release, either with or without PN (Schwartz et al., 2020), relative to individuals who received an enhanced version of usual care (i.e., treatment referral, drug education, and overdose prevention information). Although the Schwartz study found higher initial rates of methadone treatment engagement for the IM + PN group versus IM only (80.3% vs. 57.1%), no differences occurred in treatment participation for the IM groups (with or without PN) at 30 days, and no differences across IM groups (with or without PN) and ETAU in OUD-specific treatment or any kind of SUD treatment at 12 months (Schwartz et al., 2020).

McKenzie et al. (2012) examined the effects of pre-release methadone maintenance treatment (MMT). The study randomly assigned individuals to receive either 1) initiation of MMT pre-release and short-term financial assistance for treatment; 2) referral to MMT at release and short-term financial assistance; or 3) referral to MMT at release without financial assistance. Initiation of MMT prior to jail release significantly increased treatment entry and reduced time-to-treatment entry post-release; however, no significant group differences occurred in current MMT or other SUD treatment at 6 months following discharge.

Last, a pilot proof-of-concept study examining an experimental group that received XR-NTX within one week prior to release from jail and referral to community treatment versus those who did not found high rates of treatment initiation, with 15/17 initiating treatment and 12/17 (75%) receiving a second injection at week 4. No significant differences occurred by group in participation in other forms of community-based drug treatment (Lee et al., 2015).

3.2.1.2. Linkage to SUD treatment.

Four experimental studies examined linkage to community-based SUD treatment using motivational, linkage, and/or case management-based interventions. In Needels et al. (2005), women in the jail-to-community case management group had higher rates of participation in SUD treatment postrelease (66.0% vs. 56.6%, p < 0.05) compared to the jail-only group; however, no intervention effect occurred among adolescent males who received a similar intervention. Two studies evaluated motivational-based interventions delivered both prior to and following jail discharge. No significant differences existed in treatment received at 12-months postdischarge between individuals who did and did not receive a motivational intervention combined with brief treatment and treatment referral in Prendergast et al. (2017). Similarly, no significant differences occurred in treatment entry at 3- and 6-months postdischarge among women with alcohol use disorders who did and did not receive the motivational intervention in Stein, et al. (2020).

In the Recovery Management Check-ups (RMC) study, women in the RMC group met with linkage managers at release from jail, monthly in the first 90 days, and quarterly thereafter up to 36 months. During the first 90 days postrelease, women in the RMC condition returned to treatment significantly sooner (60 vs. 90 days; p < 0.05) and were more likely to participate in any SUD treatment (55% vs. 45%; p < 0.05; Scott & Dennis, 2012). In subsequent longitudinal analyses up to 36 months, RMC was associated with higher rates of treatment participation, but only in the absence of probation, which exerted a stronger effect on treatment and self-help participation (Scott et al., 2017).

3.2.1.3. Linkage to HIV + SUD services.

In the Cunningham et al. (2018) study, peer navigators worked with HIV+ men and transgender women to increase their capacity to navigate health service systems and self-efficacy to address stigma and discrimination within service systems. Findings on linkage to medication for addiction treatment (MAT) for alcohol or drug use disorders were mixed. Participants in PN reported a significantly greater increase in number of MAT visits at 6 months, whereas control participants, who received transitional case management services without PN, had a greater increase in number of MAT visits at 12 months; however, the number receiving MAT overall was very small at any time point (< 10% of the study sample). Qualitative interviews conducted with the target population for this study illustrated the lack of integration across SUD treatment, HIV medical care, and criminal justice systems, which created challenges to accessing needed services. Participants described how entry into residential SUD treatment upon release provided them with structure, housing, and stabilization that facilitated their engagement in HIV care and helped them to avert relapse at discharge (Harawa, Amani, Bowers, Sayles, & Cunningham, 2017).

3.2.2. Criminal justice outcomes

Overall, 11/14 studies examined criminal justice outcomes; finding were mixed, with 5 studies showing beneficial effects associated with the re-entry intervention, 6 showing no significant effects.

Four quasi-experimental studies used administrative records to examine criminal justice-related outcomes among individuals who received interventions that included both in-jail treatment and linkage to community services at release. Findings from these studies are mixed, with generally favorable outcomes regarding recidivism in the initial follow-up period; however, positive intervention effects in some studies weakened over time or did not hold up in multivariate models. Bahr et al. (2012) compared arrest outcomes of 70 participants who completed a short-term intensive drug treatment program in jail and then transitioned to community-based treatment with a comparison group of 70 nonparticipants using propensity score matching. Using arrest records, the study found program completers were about half as likely to recidivate as the matched controls (27% vs. 46%, HR = 0.54) up to 20 months following release. Similarly, Braga et al. (2009) evaluated an intensive case management and mentoring program for individuals at risk of violent crime and recidivism versus a nontreatment comparison group using propensity score matching. Survival analyses found the risks of any re-arrest and arrest for violent crime were reduced by about one-third in the intervention group, controlling for participants’ characteristics. At one year postrelease, the re-arrest rates for the intervention and comparison groups were 36.1% versus 51.1%, respectively, although the difference between the groups narrowed to about 10% at years 2 and 3.

Miller and Miller (2010, 2015) compared one-year postrelease recidivism outcomes for two study cohorts that participated in a comprehensive case management re-entry program situated in a rural jail relative to comparison groups matched on demographic and legal status variables. The first cohort evaluation (Miller & Miller, 2010) found that re-arrests were significantly lower in the intervention versus comparison group (12.3% vs. 82%, respectively), and this effect was retained in a multivariate model controlling for participant socio-demographics and criminal history. Similarly, in the second cohort evaluation (Miller & Miller, 2015), a significantly lower proportion of the intervention group recidivated relative to the comparison group (29% vs. 73.3%, respectively) or received a parole violation (9.7% vs. 46.7%); however, the effect of program participation was significant only for parole violations, but not re-arrest, in multivariate models controlling for participant characteristics.

In a subsequent study, Miller et al. (2016) evaluated the effects of participation in a re-entry program for men receiving SUD treatment in jail that included 90 days of aftercare. Program participants had lower rates of recidivism than the matched comparison group (27.7% vs. 75%, p < 0.001), significantly lower odds of new charge recidivism (OR = 0.14) and any recidivism (OR = 0.07), and increased time to recidivism (HR = 0.25), but no significant difference occurred for probation revocations.

Four RCTs examined criminal justice outcomes among individuals who received re-entry interventions to link them to MOUD. None of these studies (Farabee et al., 2020; Kelly et al., 2020; Lee et al., 2015; McKenzie et al., 2012) found significant intervention effects across various measures of criminal behavior outcomes, including self-reported criminal behavior, re-arrest, re-incarceration, time to re-arrest, or severity of arrests over follow-up periods ranging from 4 weeks to 24 months.

Two other RCTs examined recidivism outcomes for individuals who received interventions focused on motivation, engagement, and linkage to SUD treatment following discharge. In Prendergast et al. (2017) no differences existed in arrests at 12 months for individuals who received either a motivational intervention and brief treatment while in jail, along with postrelease brief treatment or linkage to community SUD treatment for those who screened at a high level of substance use severity, relative to a control group that received standard referral to community SUD treatment.

The RMC study found no direct effect of RMC on criminal justice outcomes; however, women who were abstinent were also less likely to be involved in any illegal activity, re-arrested, or re-incarcerated over the initial 90-day follow-up period (Scott & Dennis, 2012). Moreover, probation status had a stronger effect on criminal justice outcomes than the experimental intervention over 36 months (Scott et al., 2017). A secondary analysis of data from this study found that having custody of one’s children exerted a strong protective effect on reducing recidivism for women over the 36-month follow-up period, regardless of treatment group (Scott et al., 2016).

Last, a process of pre-release intensive discharge planning was provided in the Needels et al. (2005) study to adolescent males and adult females combined with postrelease case management, crisis intervention, and counseling for those in the experimental condition. Among a variety of criminal justice-related outcomes examined, only serious arrest charges (i.e., murder, assault, robbery, burglary) were significantly reduced among women in the experimental group, although the overall rate of serious arrest was low (less than 5%); no differences occurred between groups on any criminal justice outcomes for the adolescent males. In secondary analyses conducted with data from this study, overall illegal activity declined significantly in the year after release regardless of condition for both adult women and adolescent males; postrelease employment and health insurance were associated with lower re-arrest rates for both, whereas having a child at home was protective against drug-related crime and arrest for women only (Freudenberg et al., 2005).

3.2.3. Substance use outcomes

Ten studies (10/14) examined substance use outcomes; 3 found positive intervention effects at some point postrelease; 7 found no significant intervention effects, although in some cases effects varied by type of substance.

The Lee et al. (2015) study found significant intervention effects for individuals who received XR-NTX prior to jail release combined with brief motivational counseling and referral to community-based treatment compared to those who received postrelease referral only. The experimental group had lower rates of opioid relapse and higher rates of opioid abstinence and opioid negative urine tests over 8 weeks following release. However, no differences occurred in rates of injection drug use, cocaine use, or overdose. Similarly, in McKenzie et al. (2012) individuals who received pre-release initiation of MMT had lower rates of relapse to heroin use at 6 months compared with groups that received postrelease referral to MMT (with or without financial assistance for treatment). Cocaine use was also significantly lower in the pre-release MMT group relative to the other groups.

In the Stein et al. (2010) study, women assessed with hazardous alcohol use who were in a combined prison/jail received brief motivational interventions both prior to release and 1–3 months following release. At 3-month follow-up, women in the experimental group had more abstinent days relative to a control group (68% vs. 57%) and fewer alcohol-related problems, but no significant differences existed between groups in number of drinks per drinking day at 3 months nor in any alcohol-related outcomes at 6 months.

The remaining studies (Cunningham et al., 2018; Farabee et al., 2020; Myers et al., 2018; Needels et al., 2005; Prendergast et al., 2017; Schwartz et al., 2020; Schwartz et al., 2021; Scott et al., 2017; Scott & Dennis, 2012) found no significant direct intervention effects on substance use outcomes, although several studies noted that overall significant reductions occurred in substance use at follow-up across groups (Cunningham et al., 2018; Schwartz et al., 2020) or that the intervention had indirect effects by increasing treatment participation, which in turn was associated with better substance use outcomes (Scott & Dennis, 2012).

4. Discussion

This review examined a range of re-entry interventions that have the goal of closing the gap between services received inside jails and in the community for individuals with SUD and other health and social challenges. The review differentiated studies based on intervention targets, including MOUD-specific treatment and nonspecific SUD services, as well as interventions targeting individuals with HIV/AIDS who had high rates of SUD and need for treatment. The synthesis of outcomes reported areas of convergence across studies and gaps that future research should address. We address the implications of this study’s findings for research and intervention development, following the suggestions of Levac, Colquhoun, and O’Brien (2010) on strategies to improve the methodology of scoping reviews and their relevance to health care delivery.

4.1. Treatment linkage, retention, adherence

Overall, this review found evidence to support the effectiveness of a diverse range of interventions in initial engagement in or linkage to services, and yet limited support existed for their effects on longer-term treatment retention or medication adherence. Thus, the interventions largely accomplish the goal of facilitating the connection of individuals to services following their release from jail through assertive referral and linkage, navigation, and case management interventions. Yet following the initial engagement in SUD treatment or treatment with MOUD, high rates of treatment attrition occurred, along with limited retention or medication adherence over time.

None of the included studies evaluated comparative treatment outcomes for different forms of MOUD, despite prior research showing more favorable postrelease treatment engagement for individuals treated with buprenorphine versus methadone prior to release from jail in the absence of linkage interventions (Magura et al., 2009). Similarly, a recent comparative effectiveness trial of different buprenorphine formulations (without a re-entry linkage component) found better treatment retention at 8 weeks postrelease for individuals who received pre-release extended-release buprenorphine versus daily sublingual buprenorphine-naloxone (Lee et al., 2021). Comparative effectiveness trials in this area are especially needed to evaluate differential effects of pre-release MOUD initiation combined with re-entry interventions that aim to sustain treatment engagement and medication adherence over time.

Intervention effects were strongest in studies that used more assertive mechanisms to link to nonspecific SUD treatment (e.g., intensive case management services, Recovery Management Check-ups), whereas studies using brief motivational interventions prior to and after release did not yield significant effects on treatment entry. Several studies succeeded in transitioning individuals to community-based MOUD providers, following in-jail initiation of treatment, using case management, and/or PN. However, retention in MOUD treatment fell sharply after 90 days, and differences in treatment participation between groups were limited beyond that time. Few studies examined treatment re-engagement over time, although the RMC study that employed periodic check-ups to re-link participants who had indicators of treatment need yielded higher rates of treatment participation over time among participants that were not currently on probation (Scott et al., 2017).

Additional research should examine linkage interventions that are coupled with expanded components or longer duration to promote retention, medication adherence, and treatment re-linkage, as needed. Moreover, the linkage interventions examined typically included a bundle of services and components, making it hard to discern which specific components are “essential” to achieve the desired outcomes, which are extraneous, and the optimal packaging of components. Intervention components should be theory-driven and target factors that research has identified as associated with treatment retention, medication adherence, recidivism, and relapse, including structural determinants of health, such as access to employment, health care, social and recovery support, and housing (Binswanger et al., 2012). Considerable overlap occurred across intervention models (e.g., case management, peer or patient navigation), and variability in intervention staff, duration, targeted outcomes, and follow-up time frames. Future studies need to clearly delineate intervention models and targeted outcomes, and systematically examine core components and service arrays to understand the relationship of intervention components to specific outcomes. Relatedly, assertive interventions, such as those employing intensive case management or patient navigation, were generally effective at linking and engaging individuals in services. However, since these more intensive interventions are likely to have higher costs (Zarkin et al., 2020), as well as provide a longer duration of engagement, cost-effectiveness studies are essential to evaluate these dimensions of passive versus assertive, limited versus comprehensive, and shorter versus longer duration.

4.2. Criminal justice outcomes

Study findings were mixed in terms of the types of criminal justice outcomes examined and evidence of intervention effects was weak. Quasi-experimental studies that matched experimental participants with comparable controls found superior outcomes on recidivism for the intervention groups using administrative records on arrests and probation violations. However, in RCTs that evaluated criminal justice outcomes, no demonstrated intervention effects occurred across a range of measures apart from one that was limited to decreases in serious arrests among women who received transitional planning and intensive case management (Needels et al., 2005). Moreover, studies that focused on HIV+ individuals released from jail did not measure criminal justice outcomes, which is a missed opportunity.

4.3. Substance use outcomes

Findings regarding pre-release initiation of treatment with MOUD on postrelease substance use were mixed. Pre-release initiation of XR-NTX was associated with better opioid outcomes in Lee et al. (2015), but follow-up was limited to 8 weeks. Similarly, individuals who received pre-release initiation of MMT and referral to community providers had lower rates of heroin relapse at 6 months (McKenzie et al., 2012). In contrast, interim methadone prior to release in Schwartz et al. (2020, 2021) or pre-release initiation of XR-NTX in Farabee et al. (2020), coupled with patient navigation, showed no beneficial impact on opioid or other substance use outcomes over 12–24 months.

Evidence of intervention effects for studies that targeted nonspecific SUD treatment was weak, although, generally, overall reductions occurred in substance use postrelease, but studies showed no discernable group effects. Similarly, in studies that primarily targeted linkage to HIV services along with SUD treatment and other services, little evidence existed of intervention impact on substance use outcomes.

4.4. Need to examine interventions for specific target populations

As noted, most of the included studies had samples that were majority male; only 2 studies targeted women in jail and one had a women’s specific component (Richie, Freudenberg, & Page, 2001). The number of women confined in jail increased by 11.4% from 2005 to 2019 (from 94,000 to 110,500), whereas the number of men decreased by 8.5% over this same period (Zeng & Minton, 2021). Given the challenges that women in jail face related to parenting responsibilities, risk of interpersonal violence, and stigma, more studies should address gender-based issues in the re-entry process (Beeble & Hampton, 2021).

Similarly, only one study targeted a jail in a rural community (Miller & Miller, 2010, 2015). Individuals who are released to rural communities face greater challenges to engaging in SUD treatment and other services in communities with limited treatment options, heightened stigma, and fewer housing and employment resources (Singer & Kopak, 2021); and lack of access to continuity of care from jail is a risk factor for recidivism in rural communities that lack adequate services (Snell-Rood, Willging, Showalter, Peters, & Pollini, 2020; Victor, Comartin et al., 2021). Although increases have occurred in recent years, rural areas have less access to physicians who can dispense MOUD, particularly buprenorphine (Andrilla & Patterson, 2021), including longer-acting formulations, which are a promising approach to avert relapse among individuals with OUD at re-entry (Lee et al., 2021; Winkelman & Silva, 2021).

4.5. Need for implementation research on jail re-entry interventions

The research infrastructure in these studies often supported the development and implementation of re-entry interventions, such as covering staffing costs and resources for participants’ follow-up, which in turn may have enhanced participants’ ongoing connection to the intervention itself. Implementation research is needed to understand how to ensure that successful interventions can be integrated and sustained in the absence of a research infrastructure. Moreover, implementing a re-entry program in a jail setting may change the dynamics and culture of the jail itself (Miller, 2014). Several implementation frameworks have been successfully applied to developing cross-system interventions that depend upon successful integration across corrections and community-based treatment (Ferguson et al., 2019; Hanna et al., 2020) and need to be further expanded in future research.

Where available, this review incorporated findings from qualitative components of the included studies that provided more in-depth information on how participants experienced the re-entry interventions or the challenges they faced in the re-entry process as well as the perspectives of intervention staff. Such information is a valuable companion to interpreting intervention outcomes and developing strategies to improve implementation. Perspectives of participants receiving MOUD at jail re-entry have highlighted the importance of the social and community contexts to which they return, and the risks they encounter from lack of access to housing, employment, social services, adequate health services, and recovery supports (Hu et al., 2020; Velasquez et al., 2019).

4.6. Expand interventions to address multiple risks

Although more recent studies of interventions focused on individuals with OUD have also incorporated overdose prevention components, this was by no means universal. Jail re-entry interventions provide an excellent opportunity to leverage overdose prevention education and training, including naloxone distribution (Wenger et al., 2019). Moreover, overdose prevention education should not be limited to individuals with OUD, given the risk of overdose from stimulants and polysubstance use (Curtis et al., 2018) as well as the need to equip individuals to intervene in situations where others are overdosing. Future interventions should incorporate overdose prevention within re-entry interventions, including harm reduction, such as use of fentanyl test strips (Taylor et al., 2021).

Similarly, given the demonstrated link between drug use and HIV, the opportunity to incorporate a dual focus on substance use and HIV services and risk reduction was evident in some studies, but could be further expanded (Nunn et al., 2010; Taweh et al., 2021). A recent systematic review found strong support that treatment with MOUD increases HAART adherence and viral suppression for people with HIV/AIDS (McNamara et al., 2021). Conversely, methamphetamine use was associated with reduced viral suppression among individuals receiving re-entry peer navigation in the Cunningham et al. (2018) study (Goodman-Meza et al., 2019), underscoring the importance of integrating SUD treatment within interventions aiming to link individuals to HIV services following release from jail.

4.7. Study limitations

Study limitations stem from the nature of scoping reviews, which aim to characterize the size and scope of research on a topic, but do not include a systematic methodological assessment of studies nor quantitative synthesis of common outcomes (Grant & Booth, 2009). Thus, this review included studies that ranged in study design and intervention model, as appropriate to the research questions. The review included only peer-reviewed publications, which excluded unpublished findings and gray literature, such as evaluation reports. Studies varied in the level of detail provided on the specific intervention components, which may have led to more limited descriptions of the full range of services included.

5. Conclusion

The lack of prolonged engagement in community-based SUD treatment and MOUD adherence following jail discharge and lack of sustained improvements in substance use and criminal behavior observed across these studies suggest that more intensive and longer-duration interventions are needed. Researchers and providers can usefully apply the service cascade framework to extend the focus of jail re-entry interventions beyond the initial phase of treatment linkage and engagement to longer-term treatment retention, medication adherence, substance use recovery, and behavioral change. Studies participating in the NIDA-sponsored Justice Community Opioid Innovation Network (JCOIN)—which include (1) comparative effectiveness trials of MOUD; (2) natural experiments capitalizing on ongoing state policy rollouts; and (3) randomized trials examining different linkage strategies to connect individuals with community-based treatment services at re-entry (Ducharme, Wiley, Mulford, Su, & Zur, 2021)—address many of the methodological issues and intervention gaps that this review identifies. Moreover, many are hybrid effectiveness-implementation trials and promise to yield findings that will enhance implementation of jail re-entry interventions for individuals with SUD and other behavioral health and service needs, with the aim of improving their long-term recovery and health outcomes.