Index Incarceration Research Articles

Prediction of Recidivism and Detection of Risk Factors Under Different Time Windows Using Machine Learning Techniques

Following a comprehensive analysis of the initial three generations of prisoner risk assessment tools, the field has observed a notable prominence in the integration of fourth-generation tools and machine learning techniques. However, limited efforts have been made to address the explainability of data-driven prediction models and their connection with treatment recommendations. Our primary objective was to develop predictive models for assessing the likelihood of recidivism among prisoners released from their index incarceration within 1-year, 2-year, and 5-year timeframes. We aimed to enhance interpretability using SHapley Additive exPlanations (SHAP). We collected data from 20,457 in-prison records from February 10, 2005, to August 25, 2021, sourced from a Southwestern China prison’s data management system. Recidivism records were officially determined through data mining from an official website and combined identification data from neighboring prisons. We employed five machine learning algorithms, considering sociodemographic, physical health, psychological assessments, criminological characteristics, crime history, social support, and in-prison behaviors as factors. For interpretability, SHAP was applied to reveal feature contributions. Findings indicated that young prisoners accused of larceny, previous convictions, lower fines, and limited family support faced higher reoffending risk. Conversely, middle-aged and senior prisoners with no prior convictions, lower monthly supermarket expenses, and positive psychological test results had lower reoffending risk. We also explored interactions between significant predictive features, such as prisoner age at incarceration initiation and primary accusation, and the duration of current incarceration and cumulative prior incarcerations. Notably, our models consistently exhibited high performance, as shown by AUC on the test dataset across time windows. Interpretability results provided insights into evolving risk factors over time, valuable for intervention with high-risk individuals. These insights, with additional validation, could offer dynamic prisoner information for stakeholders. Moreover, interpretability results can be seamlessly integrated into prison and court management systems as a valuable risk assessment tool.

Prevalence of Mental Health and Addiction Service use Prior to and During Incarceration in Provincial Jails in Ontario, Canada: A Retrospective Cohort Study.

Individuals with mental illness and addiction are overrepresented in prisons. Few studies have assessed mental health and addiction (MHA)-related service use among individuals experiencing incarceration using health administrative data and most focus on service use after prison release. The objective of this study was to determine the prevalence of MHA-related service use in the 5 years prior to and during incarceration. We used linked correctional and administrative health data for people released from Ontario provincial jails in 2010. MHA-related service use in the 5 years prior to the index incarceration was categorized hierarchically into four mutually exclusive categories based on the type of service use: psychiatric hospitalization, MHA-related emergency department (ED) visit, MHA-related outpatient visit (from psychiatrist or primary care physician), and no MHA-related service use. Demographic, diagnostic, and incarceration characteristics were compared across the four service use categories. MHA-related service use during the index incarceration was assessed by category and length of incarceration. A total of 48,917 individuals were included. Prior to incarceration, 6,116 (12.5%) had a psychiatric hospitalization, 8,837 (18.1%) had an MHA-related ED visit, and 15,866 (32.4%) had an MHA-related outpatient visit. Of the individuals with any MHA-related service prior to incarceration, 60.4% did not receive outpatient care from a psychiatrist prior to incarceration and 65.6% did not receive MHA-related care during incarceration. Despite a high prevalence of mental illness and addiction among people experiencing incarceration, access to and use of MHA-related care prior to and during incarceration is poor. Increasing the accessibility and use of MHA-related services throughout the criminal justice pathway is warranted.

Impact of methadone treatment initiated in jail on subsequent arrest

BackgroundThere are limited data from randomized trials about the impact of starting methadone treatment in jail on subsequent arrest after release for adults with opioid use disorder (OUD). MethodsOfficial arrest records were obtained for 212 participants with OUD who were enrolled in a three-group randomized controlled trial of initiating methadone treatment in jail either with or without patient navigation vs. enhanced treatment-as-usual in Baltimore, Maryland. Participants treated for opioid withdrawal in jail were assigned to: 1) interim methadone (IM) with patient navigation (PN; IM + PN); 2) IM without PN (IM); or 3) enhanced treatment-as-usual (ETAU). Participants in both IM groups were able to continue treatment at a community-based methadone treatment program with counseling upon release, while ETAU participants received overdose information and a city-wide treatment assessment/referral number. Likelihood of arrest, time to first subsequent arrest, and severity of arrest charges in the 12 months following release were examined for: 1) combined IM + PN and IM groups compared to ETAU; and 2) IM + PN compared to IM. ResultsWithin 12 months of release from the index incarceration, 50.5% of the sample had been arrested. The majority of arrest charges (71%) were for low-level, nonviolent crimes. On an intention-to-treat basis, there were no significant differences between the combined IM + PN and IM groups vs. ETAU or IM + PN vs. IM in the likelihood of arrest, time to first subsequent arrest, or severity of arrest charges. ConclusionInitiating IM with or without PN during pretrial detention did not have a significant effect on subsequent arrest during a 12-month post-release follow-up compared to not starting methadone maintenance during detention, despite the high rate of methadone treatment entry in the community following release. This finding may be attributable to the considerable attrition from treatment in the community or other systematic factors. Additional interventions may be needed to reduce the likelihood of subsequent arrest.

Observational study of continuity of HIV care following release from correctional facilities in South Africa

BackgroundWe sought to describe linkage to care, ART continuity, and factors associated with linkage to care among people with HIV following release from incarceration in South Africa.MethodsWe conducted a study of South African correctional service community reentrants who were receiving ART at the time of release. The study was implemented in three of 46 correctional service management areas. Participants were enrolled prior to corrections release and followed up to 90 days post-release to obtain self-reported linkage to care status and number of days of ART provided at corrections release. Clinic electronic and paper charts were sought and abstracted to verify self-reported linkage to care. Log-binomial regression, adjusted for facility, was used to identify associations with post-release linkage to care (self-reported and verified). We sought to specifically assess for associations with HIV diagnosis during index incarceration, ART initiation during index incarceration, and duration of incarceration.ResultsFrom May 2014 to December 2016, 554 inmates met eligibility and 516 (93%) consented to participate; 391 were released on ART, 40 of whom were excluded from analysis post-release. Of the remaining 351, 301 (86%) were men and the median age was 35 years (interquartile range 30, 40). Linkage to care was self-reported by 227 (64%) and linkage to care could be verified for 121 (34%). At most, 47% of participants had no lapse in ART supply. Initiating ART during the index incarceration showed a trend toward increased self-reported post-release linkage to care. Age > 35 years was associated with increased verified linkage to care while HIV diagnosis outside of a correctional setting and ART initiation during the index incarceration showed trends toward association with increased verified linkage to care.DiscussionThe results of our study are the first description of retention in care following correctional facility release from an African setting and indicate high levels of attrition during the transition from correctional facility to community care. Initiating ART within a correctional facility did not impair post-release linkage to care.

Non-fatal drug overdose after release from prison: A prospective data linkage study

BackgroundAdults released from prison are at increased risk of poor health outcomes and preventable mortality, including from overdose. Non-fatal overdose (NFOD) is a strong predictor of future overdose and associated with considerable morbidity. This study aims to the determine the incidence, predictors and clinical characteristics of NFOD following release from prison. MethodsWe used pre-release interview data collected for a randomised controlled trial in 2008–2010, and linked person-level, state-wide ambulance, emergency department, and hospital records, from a representative sample of 1307 adults incarcerated in Queensland, Australia. The incidence of NFOD following release from prison was calculated. A multivariate Andersen-Gill model was used to identify demographic, health, social, and criminal justice predictors of NFOD. ResultsThe crude incidence rate (IR) of NFOD was 47.6 (95%CI 41.1–55.0) per 1000 person-years and was highest in the first 14 days after release from prison (IR = 296 per 1000 person-years, 95%CI 206–426). In multivariate analyses, NFOD after release from prison was positively associated with a recent history of substance use disorder (SUD), dual diagnosis of mental illness and SUD, lifetime history of injecting drug use, lifetime history of NFOD, being dispensed benzodiazepines after release, a shorter index incarceration, and low perceived social support. The risk of NFOD was lower for people with high-risk alcohol use and while incarcerated. ConclusionsAdults released from prison are at high risk of non-fatal overdose, particularly in the first 14 days after release. Providing coordinated transitional care between prison and the community is likely critical to reduce the risk of overdose.

Linkage to hepatitis C care after incarceration in jail: a prospective, single arm clinical trial

BackgroundHepatitis C virus (HCV) is a major public health problem in correctional settings. HCV treatment is often not possible in U.S. jails due to short lengths of stay. Linkage to care is crucial in these settings, but competing priorities complicate community healthcare engagement and retention after incarceration.MethodsWe conducted a single arm clinical trial of a combined transitional care coordination (TCC) and patient navigation intervention and assessed the linkage rate and factors associated with linkage to HCV care after incarceration.ResultsDuring the intervention, 84 participants returned to the community after their index incarceration. Most participants were male and Hispanic, with a history of mental illness and a mean age of 45 years. Of those who returned to the community, 26 (31%) linked to HCV care within a median of 20.5 days; 17 (20%) initiated HCV treatment, 15 (18%) completed treatment, 9 (11%) had a follow-up lab drawn to confirm sustained virologic response (SVR), and 7 (8%) had a documented SVR. Among those with follow-up labs the known SVR rate was (7/9) 78%. Expressing a preference to be linked to the participant’s existing health system, being on methadone prior to incarceration, and feeling that family or a loved one were concerned about the participant’s wellbeing were associated with linkage to HCV care. Reporting drinking alcohol to intoxication prior to incarceration was negatively associated with linkage to HCV care.ConclusionWe demonstrate that an integrated strategy with combined TCC and patient navigation may be effective in achieving timely linkage to HCV care. Additional multicomponent interventions aimed at treatment of substance use disorders and increasing social support could lead to further improvement.Trial registrationClinicaltrials.gov NCT04036760 July 30th, 2019 (retrospectively registered).

Resumption of injecting drug use following release from prison in Australia

IntroductionEx-prisoners with a history of injecting drug use (IDU) experience disproportionate drug-related harm. Rapid resumption of substance use following prison release is common and evidenced in high rates of overdose mortality. However, few studies have documented the rate of IDU resumption following prison release or identified risk factors for relapse. MethodsStructured interviews were conducted with 533 adults with a history of IDU in Queensland, Australia prior to release from prison and approximately 1, 3 and 6 months post-release. Incidence of self-reported IDU resumption was calculated overall and for each follow-up interval. Risk factors associated with time to resumption of IDU were estimated using discrete-time survival analysis. ResultsIDU resumption was reported by 41% of participants during a median of 98days of follow-up (IQR=94–121), an overall crude incidence of 1.06 per person-year. The highest rate was observed in the first month (23%; crude incidence 2.24 per person-year). In adjusted discrete-time survival analyses, being unemployed at the previous interview (AHR=1.59; 95%CI:1.10–2.30), shorter incarceration (≤90days vs. >365days; AHR=2.20; 95%CI:1.33–3.65), and IDU during the index incarceration (AHR=2.80; 95%CI:1.92–4.09) were significantly associated with time to IDU resumption; parole was protective (AHR=0.66; 95%CI:0.47–0.92). ConclusionsEvidence-based efforts to prevent IDU in prison and IDU resumption after release are important for both prisoner and public health. Enhancing opportunities for employment and capitalising on the short-term benefits of parole for ex-prisoners may delay resumption of IDU after release from prison. These strategies should complement rather than replace harm reduction efforts for this high-risk population.

Clinical risk factors for death after release from prison in Washington State: a nested case-control study.

While mortality rates after prison release are high, little is known about clinical risk factors for death. We sought to identify risk and protective factors for all-cause and accidental poisoning (overdose) death. Nested case-control study of people released from prison. Washington State Department of Corrections, Washington, USA. Cases (699 all-cause deaths, of which 88 were among women, and 196 additional overdose deaths, of which 76 were among women) between 1999 and 2009 matched 1 : 1 to controls on sex, age and year of release using risk set sampling. Prison medical charts were abstracted for clinical information. Independent associations between clinical characteristics and all-cause and overdose mortality were assessed using conditional logistic regression. Key independent risk factors for all-cause mortality included homelessness [odds ratio (OR) = 1.53, 95% confidence interval (CI) = 1.06, 2.23], injection drug use (OR = 1.54, 95% CI = 1.16, 2.06), tobacco use (OR = 1.51, 95% CI = 1.07, 2.13), cirrhosis (OR = 4.42, 95% CI = 1.63, 11.98) and psychiatric medications before release (OR = 2.38, 95% CI = 1.71, 3.30). Independent risk factors for overdose mortality included substance use disorder (OR = 2.33, 95% CI = 1.32, 4.11), injection drug use (OR = 2.43, 95% CI = 1.53, 3.86), panic disorder (OR = 3.87, 95% CI = 1.62, 9.21), psychiatric prescriptions before release (OR = 2.44, 95% CI = 1.55, 3.85) and problems with opiates/sedatives (OR = 2.81, 95% CI = 1.40, 5.63). Substance use disorder treatment during the index incarceration was protective for all-cause (OR = 0.67, 95% CI = 0.49, 0.91) and overdose (OR = 0.57, 95% CI = 0.36, 0.90) mortality. Injection drug use and substance use disorders are risk factors for death after release from prison. In-prison substance use disorder treatment services may reduce the risk.

Sex Trading Among Hazardously Drinking Jailed Women

For women involved in sex trading, both alcohol problems and passage through the criminal justice system are highly prevalent. This study is the 1st to conduct a focused examination of factors associated with sex trading among hazardously drinking, pretrial, jailed women. Cocaine use, social support for alcohol abstinence, and more days incarcerated in the 90 days leading up to the index incarceration were significantly associated with sex trading involvement among alcoholic women. Helping incarcerated alcoholic women reduce cocaine use and improve sober support networks during and following an incarceration may minimize sex trading after release.

A Phase 4, Pilot, Open-Label Study of VIVITROL® (Extended-Release Naltrexone XR-NTX) for Prisoners

This was a Phase 4, pilot, open-label feasibility study of extended-release injectable naltrexone (XR-NTX) administered to pre-release prisoners having a history of pre-incarceration opioid disorder. We evaluated the relationship between XR-NTX adherence and criminal recidivism (re-arrest and re-incarceration) and opioid and cocaine use. Twenty-seven pre-release male and female prisoners who had opioid disorders during the year prior to index incarceration were recruited and received one XR-NTX injection once each month for 7months (1 injection pre-release from prison and 6 injections in the community) and of those 27, 10 (37%) were retained in treatment at 7-months post release. Results indicate those completing 6 compared to those completing <6 injections were less likely to test positive for opioids in the community (0% vs. 62.5%, respectively; p=0.003). Although not statistically significant, individuals who did not complete all 6 injections were more likely to be re-arrested compared to those completing all 6 community injections (31.3% vs. 0%, respectively; p=0.123). Contingent upon further study of a randomized controlled trial, XR-NTX may be a feasible option in the prison setting in view of the lack of potential for diversion. Furthermore, these data suggest that completing the entire course of treatment (6 injections) may reduce opioid use and, to a lesser degree, re-arrest and re-incarceration.

Individual Patient and Program Factors Related to Prison and Community Treatment Completion in Prison-Initiated Methadone Maintenance Treatment

While prison-initiated methadone maintenance treatment is effective, it is largely unknown as to what patient and program factors are related to outcomes. These issues were studied in a secondary analysis of data from 67 male prerelease prison inmates with preincarceration heroin addiction. Three outcomes are examined: completed prison treatment; completed 1 year of community treatment; and number of days in community treatment. Being employed (p=.045) during the three years prior to index incarceration was significantly and positively related to community treatment completion. Increased frequency of urine tests taken was significantly associated with a greater number of days in community treatment (p<.001). Limitations, policy implications, and directions for future research are discussed.

Factors associated with smoking among adolescent males prior to incarceration and after release from jail: a longitudinal study

BackgroundThe prevalence of cigarette smoking among incarcerated adult men and women is three-four times higher than in the general population, ranging from 70-80%. However, little is known about factors associated with smoking among incarcerated adolescents, especially upon their re-entry into communities after release from jail. The current study explores factors associated with smoking among adolescent males prior to incarceration and one year after their release from jail.MethodsWe conducted a secondary data analysis of the Returning Educated African-American and Latino Men to Enriched Neighborhoods (REAL MEN) study, which was designed to reduce HIV risk, substance use, and recidivism among 16–18 year old males leaving jail. We examined differences between smokers and non-smokers at the time of their incarceration (N = 552) and one year after their release from jail (N = 397) using t-tests and chi-square tests. Using logistic and linear regression we examined factors associated with current smoking status, frequency of smoking, and quantity of cigarettes smoked per day both prior to the young men’s incarceration and one year after their release from jail.ResultsPrior to incarceration, 62% of the young men reported smoking, and one-year after jail release, 69% reported smoking. Prior to incarceration, foster care history, not living with parents, not attending school, drug sales, number of sex partners, gang involvement, current drug charges, and number of prior arrests were positively associated with smoking indicators prior to incarceration. Having violent charges was inversely associated with smoking indicators prior to incarceration. One-year after release from jail, foster care history and number of prior arrests before the index incarceration were associated with smoking indicators.ConclusionsSeveral problem behaviors may be associated with adolescent males’ smoking behaviors prior to incarceration. However, the young men’s histories of difficult life circumstances and engagement in illegal activity may have long-term consequences on smoking for these young men during their transition between jail and community. Findings suggest a need for comprehensive risk reduction interventions in settings in which disadvantaged young men are institutionalized, starting in childhood.

Understanding the Revolving Door: Individual and Structural-Level Predictors of Recidivism Among Individuals with HIV Leaving Jail

Incarceration, particularly when recurrent, can significantly compromise the health of individuals living with HIV. Despite this, the occurrence of recidivism among individuals with HIV has been little examined, particularly among those leaving jail, who may be at especially high risk for return to the criminal justice system. We evaluated individual- and structural-level predictors of recidivism and time to re-incarceration in a cohort of 798 individuals with HIV leaving jail. Nearly a third of the sample experienced at least one re-incarceration event in the 6 months following jail release. Having ever been diagnosed with a major psychiatric disorder, prior homelessness, having longer lifetime incarceration history, having been charged with a violent offense for the index incarceration and not having health insurance in the 30 days following jail release were predictive of recidivism and associated with shorter time to re-incarceration. Health interventions for individuals with HIV who are involved in the criminal justice system should also target recidivism as a predisposing factor for poor health outcomes. The factors found to be associated with recidivism in this study may be potential targets for intervention and need to be further explored. Reducing criminal justice involvement should be a key component of efforts to promote more sustainable improvements in health and well-being among individuals living with HIV.

Incarceration Among Adults Who Are in the Public Mental Health System: Rates, Risk Factors, and Short-Term Outcomes

Incarceration of people with mental illness has become a major social, clinical, and economic concern, with an estimated 2.1 million incarcerations in 2007. Prior studies have primarily focused on mental illness rates among incarcerated persons. This study examined rates of and risk factors for incarceration and reincarceration, as well as short-term outcomes after incarceration, among patients in a large public mental health system. The data set included 39,463 patient records combined with 4,544 matching incarceration records from the county jail system during fiscal year 2005-2006. Risk factors for incarceration and reincarceration were analyzed with logistic regression. Time after release from the index incarceration until receiving services was examined with survival analysis. During the year, 11.5% of patients (N=4,544) were incarcerated. Risk factors for incarceration included prior incarcerations; co-occurring substance-related diagnoses; homelessness; schizophrenia, bipolar, or other psychotic disorder diagnoses; male gender; no Medicaid insurance; and being African American. Patients older than 45, Medicaid beneficiaries, and those from Latino, Asian, and other non-Euro-American racial-ethnic groups were less likely to be incarcerated. Risk factors for reincarceration included co-occurring substance-related diagnoses; prior incarceration; diagnosed schizophrenia or bipolar disorder; homelessness; and incarceration for three or fewer days. Patients whose first service after release from incarceration was outpatient or case management were less likely to receive subsequent emergency services or to be reincarcerated within 90 days. Modifiable factors affecting incarceration risk include homelessness, substance abuse, lack of medical insurance, and timely receipt of outpatient or case management services after release from incarceration.

Parole Revocation Among Prison Inmates With Psychiatric and Substance Use Disorders

Parole Revocation Among Prison Inmates With Psychiatric and Substance Use Disorders