Predictors of re-offending in German substance-abusers in legal detention : a prospective population-based cohort study

Abstract

This study aimed to evaluate whether the elapsed time after release to first re-offense is longer for those who complete coerced forensic addiction treatment than for those who fail to complete the treatment. It is also aimed to identify predictors of re-offending for both those who complete such treatments as compared with those who do not. Two hundred and sixty-one patients discharged from five German forensic addiction clinics were investigated on 65 anamnestic, socio-demographic, and therapeutic process variables. One hundred and fifty-one patients were prematurely discharged and returned to prison (group A; time at risk [TAR] = 58.7 months), 110 finished treatment successfully (group B; TAR = 44.2 months). Federal Criminal Register data were used for follow-up. Univariate survival analyses and multivariate stepwise Cox-regression models were computed. Twelve predictors in group A revealed a four-factor model: age at admission to treatment, duration of concurrent prison sentence, number of entries in the criminal register, and attendance at prior substitution programs. The model for group B covered five out of eight predictors: regular working activities before first diagnosis of mental illness, type of main offense, (secondary) traffic offenses, outpatient rehabilitation treatments, and escapes during treatment. Successful completers of forensic addiction treatment are slower to re-offend than non-completers.