UCSF

Whether incarceration confers excessive risk for accelerated aging or premature mortality remains an open question. Earlier studies suggested excess risk for aging and premature mortality but the data were either anecdotal or using population comparisons. One of the major risk factors for persons incarcerated is a history of substance abuse which alone is associated with adverse outcomes. Thus, we chose to examine the question of incarceration and adverse outcomes among a sample restricted to drug users, to determine if there was an independent risk related to incarceration. In terms of drug use, it is well documented that the risk of mortality is highest within the first couple of weeks after release. We framed the question, does a history of incarceration have a latency for a long term effect on accelerated aging and premature mortality after accounting for drug use. To examine this question, we used data from the AIDS Linked to the IntraVenous Experience (ALIVE) cohort, which consisted of injection drug users (IDU) presenting to a community clinic in Baltimore and followed semiannually for up to 30 years. Data were collected from 2005-2013 and history of incarceration was based on self-report for when the participant was entered into the cohort study anytime between 1988 – 2005. Outcome data were frailty using Fried’s criteria (For frailty, operationalized according to Fried, we used a three-level scale of non-frail, prefrail, and frail) that was collected from 2005 onward, and all-cause mortality collected from NDI-Plus. For statistical analysis, ALIVE participants were divided into two cohorts, those recruited before the introduction of highly active antiretroviral therapy (HAART) when HIV infection was a significant cause of premature mortality, and those recruited after the advent of HAART when the risk of death from HIV dropped significantly. We started the investigation with a literature review that included scholarly studies published from 1990 until 2017 on the association between incarceration and prevalence of chronic disease. In the cross-sectional study, the independent variable was a history of incarceration, operationalized as the total number of reported incarcerations in the six months prior to baseline. We examined two dependent variables: Frailty and mortality. No statistically significant effect of incarceration on frailty was found even after controlling for age, gender, race, educational attainment, HIV status, or current injection drug use. For mortality by a history of incarceration a survival analysis showed no significant difference even after controlling for age, gender, and HIV status. The adjusted relative hazard (95% CI) of mortality for those with a history of incarceration was 1.14 (0.81, 1.60) among those enrolled in the Pre-HAART era cohort and 1.19 (0.68, 2.10) for those enrolled in the HAART era cohort. Although earlier studies observed excess mortality soon after release, our data suggests that the role of past incarceration may have modest if any impact on the long term occurrence of frailty or mortality.