All-cause mortality in criminal justice clients with substance use problems--a prospective follow-up study

Background: Mortality in previously incarcerated individuals is known to be elevated, with high proportions of drug-related deaths. However, there is less documentation of whether specific substance use patterns and other clinical characteristics predict increased mortality in the group.

Methods: This is a follow-up study of mortality and causes of death in ex-prisoners with substance use problems prior to incarceration (N=4081), who were followed during an average of 3.6 years from release from prison until death or until data were censored. Baseline predictors of mortality, derived from interviews with Addiction Severity Index (ASI) in prison, were studied in a Cox regression analysis.

Results: During follow-up, 166 subjects (4.1%) died. Standardized mortality ratios were 7.0 (3.6-12.2) for females and 7.7 (5.6-9.0) for males. In 84% of cases, deaths were unnatural or due to substance-related disease. Most common causes of death were accidental poisoning (27%), transport accidents (13%), poisoning/injury with undetermined intent (12%), and suicide (10%). Death was positively predicted by heroin use, overdose, and age, and negatively predicted by a history of depression.

Conclusions: A vast majority of deaths after release from prison in individuals with substance use are due to violent or substance-related causes. Significant predictors identified were mainly related to patterns of drug use, and need to be addressed upon incarceration as risk factors of death. The findings have implications for referral and treatment upon release from prison.