While criminal legal involvement is a structural determinant of health, both administrative and national longitudinal cohort data are collected and made available in a way that prevents a full understanding of this relationship. Administrative data are both collected and overseen by the same entity and are incomplete, delayed, and/or uninterpretable. Cohort data often only ask these questions to the most vulnerable, and do not include all types of criminal legal involvement, when this involvement occurs in someone's life, or family and community involvement. To achieve a more optimized data landscape and to facilitate population-level research on criminal legal involvement and health, (1) individual administrative level data must be made available and able to be linked across carceral systems, (2) a national data archive must be made to maintain and make criminal legal data available to researchers, and (3) a nationally representative, longitudinal study focused on those with criminal legal involvement is necessary. By beginning to critically think about how future data could be collated and collected, we can begin to provide more robust evidence around how the criminal legal system impacts the health of our society and, in turn, create policy reform.

Individuals involved with community supervision experience multi-level obstacles impacting health outcomes. This is a high-risk period for HIV acquisition due to potential reengagement in unprotected sex and/or unsafe injection drug practices. This study aimed to assess the congruence between actual and perceived HIV risk and the degree to which individual, social, and behavioral factors impact risk perception among individuals on community supervision. While all participants were clinically indicated for PrEP, most participants (81.5%) did not consider themselves at risk for HIV (69.5%) or were not sure of their risk (12.0%). Among those with no or unsure perceived risk, 94% engaged in sexual behaviors that put them at-risk of HIV. Perceived HIV risk was associated with sharing injection equipment (aPR = 1.8, 95% CI [1.02, 3.3]), identifying as a sexual minority (aPR = 2.3, 95% CI [1.3, 3.9]), and having sex with a partner living with HIV (aPR = 2.4, 95% CI [1.3, 4.3]). Having sex with a partner living with HIV was the only sexual risk behavior associated with a perceived risk of HIV. These findings indicate a substantial discrepancy between actual and perceived HIV risk, highlighting the need for targeted interventions to improve risk perception accuracy and enhance risk prevention among individuals on community supervision.

After release from incarceration, individuals are rarely connected to primary care or to social services despite bearing a disproportionate burden of poor health (e.g., chronic illness) and structural determinants of health (e.g., housing instability). The Rhode Island Transitions Clinic (RITC) works to fill this gap, particularly for patients with complex needs, by linking individuals to primary care and social services. However, prior work has not formally assessed how successful the TCN is at connecting patients to social services. Our objective was to assess the relationship between RITC and connections to healthcare use and services addressing structural determinants of health.

Purpose: To understand how allostatic load - cumulative physiologic burden of stress - varies by amount and timing of arrests stratified by race/ethnicity and by sex. Methods: Using The National Longitudinal Study of Adolescent to Adult Health, we calculated descriptive statistics and mean differences in bio-marker measured allostatic load by arrest history stratified by race/ethnicity and sex. Results: One-third of participants experienced at least one arrest, and most experienced arrests only as adults. Allostatic load scores were higher for those that had ever experienced an arrest compared to never (mean difference: 0.58 (0.33, 0.84)). Similar results held for men and women and across race/ethnicity, but Black non-Hispanic individuals had higher allostatic load at all levels compared to other individuals. Conclusions: Experiencing both any arrest and multiple arrests were associated with higher allostatic load. The stress of arrests may contribute to physiological maladaptations and poor health. The public health and law enforcement fields must recognize the detrimental consequences of arrests on physiological stress and search for non-carceral solutions.

Purpose: Identifying predictors of opioid overdose following release from prison is critical for opioid overdose prevention. Methods: We leveraged an individually linked, state-wide database from 2015-2020 to predict the risk of opioid overdose within 90 days of release from Massachusetts state prisons. We developed two decision tree modeling schemes: a model fit on all individuals with a single weight for those that experienced an opioid overdose and models stratified by race/ethnicity. We compared the performance of each model using several performance measures and identified factors that were most predictive of opioid overdose within racial/ethnic groups and across models. Results: We found that out of 44,246 prison releases in Massachusetts between 2015-2020, 2237 (5.1%) resulted in opioid overdose in the 90 days following release. The performance of the two predictive models varied. The single weight model had high sensitivity (79%) and low specificity (56%) for predicting opioid overdose and was more sensitive for White non-Hispanic individuals (sensitivity = 84%) than for racial/ethnic minority individuals. Conclusions: Stratified models had better balanced performance metrics for both White non-Hispanic and racial/ethnic minority groups and identified different predictors of overdose between racial/ethnic groups. Across racial/ethnic groups and models, involuntary commitment (involuntary treatment for alcohol/substance use disorder) was an important predictor of opioid overdose.