Abstract
To bridge a gap in access to community services for releasing state inmates, the Boston Offender Needs Delivery (BOND) project was developed as a longitudinal study (2014-2016) of adult inmates with a history of substance use and co-occurring mental health disorders returning to the community following detention in one of Massachusetts’ State correctional facilities. Pre-release inmates who were wrapping up their sentence (i.e., no community supervision) and presented with mental illness and/or substance use disorder were recruited. Participants were provided substance abuse treatment immediately post-release, as well as a variety of recovery supports for a follow-up period of 6 months. This first manuscript is intended to describe the rationale behind the BOND project, as well as the methods and procedure used to collect the data.
Keywords
Introduction
The increasing number of mentally ill individuals coming into contact with the criminal justice system has been associated with several changes occurring across the multiple waves of deinstitutionalization beginning in the 1960s (Lamb & Bachrach, 2001; Wallace, Mullen, & Burgess, 2004). The objective of psychiatric deinstitutionalization was to provide mentally ill individuals with mental health services in the community. Unfortunately, adequate and sufficient outpatient community-based services for persons with a mental illness did not follow these changes. In parallel, important modifications were brought to legislation pertaining to involuntary commitment of individuals with a mental illness (Laberge & Morin, 1995; Lamb, Weinberger, & Gross, 1999). The reduced access to assessment and treatment for mentally ill individuals, and the increasingly complex psychosocial profiles in this population (e.g., substance use, homelessness), has led to an increasing number of individuals with mental illness interacting with the criminal justice system (Wallace et al., 2004).
Community reentry of vulnerable and institutionalized populations has been the subject of increased awareness considering the criminalization cycle that can result from poorly planned community transitions. In Massachusetts, the prison population has more than tripled since the 1980s and grown from 2,754 to 10,014 in 2016, which has corresponded to the number of prisoners released from prison doubling as well to 3,329 in 2015 (Brooks et al., 2008; Cannata et al., 2016; Kohl, Hoover, McDonald, & Solomon, 2008). Massachusetts Department of Corrections (MADOC) revealed that 24% of their inmates had an open mental health case (i.e., Axis I disorders, mental retardation, dementia, or other cognitive disorders), 8% of which were for a serious mental illness (Kohl & Lockmer, 2012). It is estimated that 50% of those who are released from prison meet Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) criteria for drug dependence or abuse (Mumola & Karberg, 2007). However, over the years, rates of community supervision have drastically decreased: It was reported that 40% of released inmates have no community supervision (Jones & Forman, 2016). Without continued supervision and support, mentally ill inmates returning to the community have difficulty obtaining the treatment needed and face restrictions on access to community-based services (Hartwell, 2004), which results in high rates of recidivism (Jacoby & Kozie-Peak, 1997; Wilson, Tien, & Eaves, 1995).
Recidivism refers to reentry into the criminal justice system following release (Wormith et al., 2007). A recent report focused on recidivism in Massachusetts reveal that two thirds of defendants sent to state or county jail and prison facilities each year had previously been incarcerated (Jones & Forman, 2016). In 2012, in the state of Massachusetts, 3 years following release, 32% of released inmates returned to prison (Cannata et al., 2016). Although recidivism rates in Massachusetts have declined from 40% in 2002 (1,008 recidivists out of 2,524 released) to 32% in 2012 (753 recidivists out of 2,346 released), the number of prisoners being released into the community remains high enough to warrant public concern (Brooks et al., 2008). Recidivism can be precipitated by a variety of factors. When inmates are released to the community, they must adapt to a less structured environment, where they face several issues including difficulties finding housing, reconnecting with family, limited employment opportunity, drug and alcohol use, continued crime involvement, shortage of supervised release programs, and stigma (Seiter & Kadela, 2003). Two types of criminogenic risk factors are found in the scientific literature (i.e., factors that predict recidivism): static factors (i.e., factors that are stable over time) and dynamic factors (i.e., changeable factors that can be targeted in treatment) (Andrews, Bonta, & Hodge, 1990; Gendreau, Little, & Goggin, 1996). Static or historical risk factors such as younger age, being male, having a history of criminal justice involvement, juvenile criminal justice involvement, and family criminality have been shown to significantly increase the risk of recidivism (Bonta, Blais, & Wilson, 2014; Bonta, Law, & Hanson, 1998; Brekke, Prindle, Bae, & Long, 2001; Fisher et al., 2014; Gendreau et al., 1996; Skeem et al., 2006). Although relevant for risk assessment, static risk factors are less informative when it comes to treatment and supervision planning. Empirically validated dynamic risk factors, which can be targeted in treatment, include but are not limited to substance use disorder, symptoms of major mental illness, personality disorder, treatment/supervision response, living situation, employment/leisure/recreation, personal support, antisocial peer association and impulsivity (Andrews et al., 1990; Bonta et al., 2014; Skeem, Winter, Kennealy, Louden, & Tatar, 2014). To be effective, reentry initiatives must parallel offenders’ criminogenic needs (Andrews et al., 1990).
Reentry initiatives not only serve inmates’ individual needs but also promote public safety. Most reentry initiatives focus on monitoring (i.e., parole and probation) rather than treatment and rehabilitation. According to Hartwell (2003b), few nonmandated reentry services are available in the states of Massachusetts. Solomon and Draine (1995, 1999) have shown that monitoring rather than rehabilitating mentally ill individuals released from the correctional system does little in terms of improving their lives and increases their likelihood of returning for technical violations. Furthermore, following principles of the Good Lives Model, which is a strength-based approach to rehabilitation, an efficient way to reduce recidivism is to provide individuals with the tools needed to live a fulfilling life (Ward, 2004). Research reveals that ex-inmates report the need for rehabilitative resources that provide programming, mentoring, and housing (Hartwell, 2008).
Rehabilitation approaches can occur within prison or after release from prison and include correctional, therapeutic, and cognitive-behavioral approaches, among others (Wormith et al., 2007). These services can even extend to include general health treatment, mental health treatment, faith-related programs, and programs specific to culture (Wormith et al., 2007). Treatment initiatives targeted toward drug abuse are most common due to the high prevalence of drug use among offenders. Programs offered during incarceration have been shown to reduce recidivism rates among participants (Aos Miller, & Drake, 2006) as well as likelihood of relapse when compared with nonparticipants (Aos et al., 2006; Pelissier et al., 2000). In one randomly assigned quasi-experimental study of 2,315 federal inmates, drug treatment was found to have statistically significant reductions in recidivism and drug use for inmates (Rhodes et al., 2001; Seiter & Kadela, 2003). Aftercare treatment has been noted as critical in the transition between prison and the community for inmates (Wormith et al., 2007). Research further suggests that the longevity of positive outcomes increases when aftercare treatment programs are combined to in-prison program (Knight, Simpson, & Hiller, 1999; Martin, Butzin, Saum, & Inciardi, 1999; Pelissier et al., 2000; Wexler, Melnick, Lowe, & Peters, 1999; Wormith et al., 2007). A study revealed that when participants receive in-prison treatment only, without aftercare treatment, reductions in recidivism and relapse rates significantly diminished after 3 years (Wormith et al., 2007). These findings support the notion that the positive impact of interventions is not permanent and can fade over time.
Though risk factors are thought to be the same for individuals with and without mental illness (Bonta et al., 2014; Bonta et al., 1998), risk management strategies must be adapted for individuals who present with mental illness, as psychiatric illnesses have been shown to complicate reentry processes for inmates (Morrissey, Meyer, & Cuddeback, 2007; Pogorzelski, Wolff, Pan, & Blitz, 2005; Wolff & Draine, 2004). In 1998, Massachusetts Department of Mental Health (DMH) established the Forensic Transition Team (FTT) to assist mentally ill inmates returning to the community. The objectives of the FTT was to coordinate services pre-release such as housing, mental health, and substance use treatment for up to 3 months post-release. The team serves individuals eligible to receive services by DMH (i.e., presence of an Axis I diagnosis). Studies looking at FTT reveal that the majority of mentally ill offenders served by the forensic teams were incarcerated following a misdemeanour charge. The majority of service users presented with substance use problems (71%), while 38% anticipated homelessness at release (Hartwell, 2003a). Three month follow-up studies concluded that nearly 72% of FTT service users remain in the community, while 23% are hospitalized and less than 5% are reincarcerated. Consistent with the literature, studies on FTT reveal that those who committed misdemeanours, those who presented with substance use disorder, and those with criminal history were more likely to be reincarcerated after 3 months (Hartwell, 2003a).
The review of literature presented above reveals a general consensus that rehabilitation programs are effective in assisting transition between prison and the community, as well as with relapse and recidivism (Seiter & Kadela, 2003). The review also reveals the importance of combining both in-prison treatment and aftercare services for inmates wrapping up their sentences. It has been suggested that the longer the person stays in the community, the less likely they are to be reincarcerated (Blumstein & Nakamura, 2009). The period post incarceration is critical in ensuring community tenure, as individuals are at high risk of relapse during that transition period. Many releasing offenders with mental illness and/or substance abuse are without parole supervision, and may not be eligible to receive DMH services. These ex-inmates find themselves with limited support to cope with the pressures that accompany release from prison and resettlement in the community. To bridge this gap in services, we conducted a project designed to assist moderate- to high-risk inmates who are noneligible for DMH services, wrapping up their sentences (i.e., not on parole) and returning to the greater Boston area.
Current Study
The Boston Offender Needs Delivery (BOND) project was a longitudinal study (January 1, 2014, to September 31, 2016) of adult inmates with a history of substance use and co-occurring mental health disorders returning to the Boston area following detention in one of Massachusetts’ State correctional facilities. This project was service oriented and funded by the Substance Abuse and Mental Health Service Administration (SAMHSA/TI 24794). Pre-release inmates who were wrapping up their sentences (i.e., no community supervision) and presented with mental illness and/or substance use disorder were recruited. Participants were provided with substance abuse treatment immediately post-release, as well as a variety of recovery support services for a follow-up period of 6 months. For the purpose of this project, our service partner was Span, Inc., a nonprofit organization with a focus on ex-inmates post prison release incorporated in 1978 in Boston.
Methodology and Approach
Participants
Detainees who were within 3 months of release without parole from one of Massachusetts’s State Correctional facilities and were returning to the greater Boston area were identified by correctional professionals and submitted an application to receive Span, Inc.’s services. To be eligible, inmates had to meet DSM-IV criteria for substance abuse and had to score medium or high risk on the COMPAS (Correctional Offender Management Profiling for Alternative Sanctions; Brennan, Dieterich, & Oliver, 2007). Similar to other states, MADOC uses the COMPAS to evaluate risk of recidivism. Referrals went to the project director who then assigned a reentry services specialist to screen the individual at the facility. Individuals who were 18 years and older, and who agreed and appeared capable of providing written informed consent to the research component of the project were included in the project and evaluation. Individuals with a known severe mental illness were excluded and referred to the DMH due to their higher service need. Individuals with a physical condition that would make participation (in the opinion of the study clinical director) medically hazardous or that required immediate treatment (e.g., imminent suicide risk) were also excluded. Last, individuals not able to receive and access services in the greater Boston area were not part of our sample. A reentry services specialist enrolled all eligible individuals through Span, Inc.’s intake and conducted a transitional needs plan that included mental health counselling through a social worker with specific training in mental health.
Sample Description
The majority of our sample (n = 125) was male with a mean age was 40.63 years old. Our sample was predominately White, followed by African American and Hispanic. The large majority of our sample had never served in the Armed Forces, the Reserves, or in the National Guard. Over a quarter of participants had not received a high school diploma or equivalent.
The majority of participants had committed a nonviolent or drug-related offense; followed by a violent offense, and a small proportion incarcerated following a technical violation (i.e., parole violation). According to aggregated statistics provided by the MADOC regarding 213 of our participants, the average maximum sentence length was less than 5 years. The large majority of our sample had at least one prior incarceration. Based on the inmate’s general and violent risk to recidivate (most recent COMPAS assessment prior to release), 157 (74%) scored high, 35 (16%) scored moderate, eight (4%) scored low, and 13 (6%) were not screened. Inmates may not be screened due to their shorter length of sentence. Finally, more than half of our sample was on probation while participating in the BOND project.
Study Design
Recruited participants first had a pretreatment visit at the correctional facility prior to release. Discharge plans were discussed and a meeting at Span, Inc., post release was scheduled. Participants were screened on various tools at their first meeting at Span, Inc. (see “Measures” section below) for the baseline interview. For the 6 months that followed, each participant received various services such as case management, and substance use and recovery counselling services, and participated in the 6-month discharge evaluation assessments (see “Measures” section below). Clients who completed their 6-month follow-up interview at the time of discharge from the BOND project received a US$20 gift card.
A total of 222 participants were assessed at baseline, and 162 completed the 6 months follow-up assessment (73%); of those, 35 (22%) withdrew from services prior to the end of the 6 months period and 125 (78%) completed services.
Measures
At baseline, participants were evaluated on the M.I.N.I International Neuropsychology Interview, the GPRA (Government Performance and Results Act), the Addiction Severity Index (ASI) as well as the Trauma-Symptom Checklist (TSC) (following release). At 6 months follow-up, participants were reinterviewed with the GPRA and ASI measures to track outcome data.
The M.I.N.I. is a self-reported brief structured interview for the major Axis I psychiatric disorders in DSM-IV (Sheehan et al., 1998). The M.I.N.I. was used to establish whether the participant presented with an Axis I disorder.
The Government Performance and Results Act Measure (GPRA Measure) is a performance measurement used by the Substance Abuse and Mental Health Services Agency (SAMHSA) that tracks substance use and mental health–related outcomes over the past 30 days. This measure is based on self-report, as well as case managers’ reports. Included in the GPRA measure is a list of services that may be offered to participants. GPRA’s description of services can be found in the appendix.
Addiction Severity Index Lite (McLellan et al., 1992; McLellan, Luborsky, Woody, & O’Brien, 1980; baseline and follow-up) is a standardized, clinical research interview based on self-report to assess problem severity in various domains. The alcohol and other drug use domain were analysed. Participants were asked, “in the 30 days that preceded your incarceration, how many days have you spent” . . . using alcohol, drugs, feeling anxious, feeling depressed, and so on.
Services
Services were offered at Span, Inc., on an individual basis, group basis, and over the phone, as well as in the community through referrals. Different services were provided and tracked through the GPRA measure over the course of the project (see the appendix for GPRA official definition of services). Services were provided by Span, Inc., professionals, by a licensed social worker specialising in recovery support and by a peer coach.
Results
The following results pertain to all 222 participants.
Baseline Measures: M.I.N.I, ASI, and GPRA Measures
Based on the baseline M.I.N.I interview, half of our total sample presented with an Axis I mental disorder, most of which presented with co-occurring substance use disorder. The most frequent diagnosis was one of depression, followed by posttraumatic stress disorder (PTSD), anxiety, bipolar disorder, and schizophrenia (see Table 1).
Descriptive Statistics.
Note. PTSD = posttraumatic stress disorder; MH = mental health; SUD= Substance use disorder.
Based on the ASI measure, at baseline, the mean number of days spent consuming alcohol in the 30 days that preceded incarceration was reported to be 10.97 (SD = 12.6). Mean number of days using drugs was 20.26 (SD = 13.11), and mean number of days using both drugs and alcohol was 9 (SD = 12.19). Based on the GPRA measure, at baseline, the mean number of days spent homeless in the month that preceded incarceration was 6.84 (SD = 12.16). The mean number of days spent feeling anxious or tense was 16.69 (SD = 13.50) and the mean number of days feeling depressed was 11.95 (SD = 13.25). The mean number of days having trouble controlling violent behavior was 3.5 (SD = 8.45). At baseline, 46 (23.6%) participants were employed, 171 (77%) were housed, 28 (12.6%) were in a shelter or on the streets.
Services Received During the Project
Based on the GPRA measure (reported by Span, Inc., professionals), during the 6 months study period, most participants received services on an outpatient basis, only three participants had to receive inpatient treatments. The large majority of participants (185, 81.9%) were provided with a treatment/recovery plan; 34 (15%) participants were referred for treatment outside of Span, Inc., services; 215 (95.1%) participants received case management services; 175 (77.4%) participants received transportation services in the form of public transport passes or being transported by Span, Inc., staff members; 194 participants received recovery support services (85.8%); 131 (58%) participants received individual counselling services provided by a licensed social worker; 102 (45.1%) participants received group counselling provided by Span, Inc., professionals or a licensed social worker; and 76 (33.6%) participants received peer-to-peer coaching in individual and/or group modality. The mean number of services provided was 3.95 (SD = 1.4) services, with a maximum of six different types of services.
Here, we report the number of sessions attended by participants who completed one or more of the six most prevalent categories of services received. The mean number of case management sessions was 164 (SD = 49) with a minimum of two and maximum of 180 during the 6 months follow-up period. The mean number of recovery support sessions received was 168.62 (SD = 41.5) with a minimum of one and a maximum of 180. The mean number of transportation services received was seven (SD = 3) with a minimum of one and a maximum of 18.
The mean number of group counselling received was 4.7 (SD = 5.3), with a minimum of one and a maximum of 42. The mean number of peer-to-peer (PTP) coaching received was 2.2 (SD = 2.5), with a minimum of one and maximum of 11. The mean number of individual counselling sessions was 5.17 (SD = 5.42), with a minimum of one and maximum of 42.
Discussion
The BOND program provided case management, substance use, and recovery counselling services for a period of 6 months following the release to community of state inmates presenting with substance use and mental health needs. Services were provided at Span, Inc., and participants were assessed on various tools at release from incarceration and following intervention (6 months after release). The present manuscript is the first to document the rationale behind the Boston Offenders Needs Delivery project as well as the methodology and procedure used to collect the data. A description of our sample demographics, baseline measures, and services received was also provided. Our results reveal that an important number of participants had co-occurring mental health difficulties. Furthermore, in the days that preceded their incarceration, inmates were presenting several difficulties such as substance use, anxiety, depression, unemployment, and homelessness. The services that have been mostly used by our participants were case management meetings, recovery support, transportation, individual and group counselling as well as peer-to-peer coaching.
Strengths, Limitations, and Future Direction
One of the strength of this project is that it bridges a gap in services for inmates who face important adaptation difficulties following long periods of incarceration. For state inmates released from long-term incarceration, the stress caused by a return to the community without support of services often results in relapse which increases the risk of criminal involvement (Lamb et al., 1999; Lindqvist & Allebeck, 1990). This project therefore targeted an especially vulnerable population who would otherwise face limited or no access to comprehensive mental health services upon their return in the community (Hartwell, 2004). Furthermore, in accordance with best practices in correctional treatment, services were provided while participants were still incarcerated. Another strength of this project was to partner with a nonprofit organization which has been a pioneer in offering services to ex-inmates in Boston since 1978. This procedure ensured that our participants received services tailored to their needs. Furthermore, for the purpose of the BOND project, Span, Inc., recruited a social worker trained to provide mental health services to our participants; which is not a service Span, Inc., usually offers.
The main limitation of this study is the absence of a control group, limiting the observation of a treatment specific effect for our future analyses. Furthermore, the study relies on self-reports, which are subject to various types of biases; especially, regarding questions at baseline, which ask participants to report symptoms in the month that preceded their incarceration. Furthermore, a proportion of our initial sample failed to complete services and/or follow-up interview, limiting our sample size.
Conclusion
For multi-problemed populations coming out of prisons, nonmandated service options, particularly from an agency where prisoner reentry is central to its mission, are value added in improving all facets of the transition experience. Our next steps are to analyse outcome measures of participants who completed the 6-month follow-up interview. We intend to further investigate the impact of services on mental health and community outcomes. Furthermore, by obtaining individual recidivism data, we will measure the extended impact of reentry services on community outcomes and to describe which community reentry services seem to be better suited for different profiles of inmates.
Footnotes
Appendix
Acknowledgements
This study could not have been possible without the full collaboration of Span, Inc.’s Team: (Kevin Davis, April Robbins, Sean Harding, Amanda Ofria, Jack Harper, Abby Smith, and Lyn Levy). The authors would like to acknowledge Massachusetts Department of Corrections for their collaboration.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. Salem received a FQR-SC postdoctoral fellowship to complete this study. The Boston Offender Needs Delivery project is funded through the Substance Abuse and Mental Health Service Administration (SAMHSA/TI 24794).