Effects of First-Time Imprisonment on Postprison Mortality

Abstract

Objectives: To examine the effects of first-time imprisonment on postprison mortality. Method: Data are used from a longitudinal study examining criminal behavior and mortality over a 25-year period in a representative group of 2,297 Dutch offenders who had their criminal case adjudicated in 1977. Of these offenders, 597 were imprisoned for the first-time in their lives in 1977. The remaining 1,700 offenders got a noncustodial sentence. Ex-prisoners' mortality rates and causes of death are compared with those in the general population and those in a matched control group of non-imprisoned offenders. Propensity score matching is used to minimize selection bias. Odds ratios with 95% confidence intervals are used to examine whether mortality among the ex-prisoners differ significantly from the general population or from the non-imprisoned controls. Results: About 18 percent of the imprisoned offenders died over the 25-year follow-up period. Compared with the general population (age and gender adjusted), ex-prisoners are three times as likely to die during the 25-year follow-up (odds ratio [OR] = 3.21). Compared with a more appropriate control group of non-imprisoned offenders (matched on age, gender, and propensity score), ex-prisoners are no longer significantly more likely to die (OR = 1.40). Conclusions: The results of the present study emphasize the importance of constructing appropriate comparison groups when examining the effects of imprisonment on postprison mortality.