Interventions to reduce harms related to drug use among people who experience incarceration: systematic review and meta-analysis.

Background: Mortality, suicide, self-harm, and substance use are elevated among people who are incarcerated. There is a wide range of heterogeneous interventions aimed at reducing these harms in this population. Previous reviews have focused on specific interventions or limited their findings to drug use and recidivism and have not explored interventions delivered after release from prison. Our aim is to examine the effect of interventions delivered to people who use drugs during incarceration or after release from incarceration, on a wide range of outcomes.

Methods: In this systematic review and meta-analysis, we searched Embase, MEDLINE, and PsycINFO databases up until Sept 12, 2023 for studies published from Jan 1, 1980 onwards. All studies evaluating the effectiveness of any intervention on drug use, recidivism outcomes, sexual or injecting risk behaviours, or mortality among people who use psychoactive drugs and who were currently or recently incarcerated were included. Studies without a comparator or measuring only alcohol use were excluded. Data extracted from each study included demographic characteristics, interventions, and comparisons. Pooled odds ratios and risk ratios were calculated using random-effects meta-analyses.

Findings: We identified 126 eligible studies (47 randomised controlled trials and 79 observational studies) encompassing 18 interventions; receiving opioid-agonist treatment (OAT) in prison reduced the risk of death in prison (one study; hazard ratio 0·25; 95% CI 0·13-0·48), whereas receiving OAT in the first 4 weeks following release reduced risk of death in the community (two studies; relative risk 0·24; 95% CI 0·15-0·37). Therapeutic community interventions reduced re-arrest at 6-12 months (six studies; odds ratio [OR] 0·72; 95% CI 0·55-0·95) and reincarceration at 24 months (two studies; OR 0·66; 95% CI 0·48-0·96). There was scarce evidence that OAT and syringe service provision are effective in reducing injecting risk behaviours and needle and syringe sharing.

Interpretation: There are effective interventions to reduce mortality and recidivism for people who use drugs who have been incarcerated. Nonetheless, there are also substantial gaps in the research examining the effect of interventions on risk behaviours and mortality during incarceration and a need for randomised designs examining outcomes for people who use drugs after release.

Funding: Australian National Health and Medical Research Council.