Assessing the Impact of Time Spent in Custody and Mortality on the Estimation of Recidivism

This article reports on findings from a methodological study which investigated the impact of two potentially biasing factors - time spent in custody and mortality - on the estimation of recidivism. Using survival analysis techniques to derive estimates of reoffending, the study compared adjusted and unadjusted recidivism rates and assessed how rates vary for different offender populations and over different follow-up periods. In contrast to many previous studies, the research found that adjusting for time spent in custody and mortality makes little difference to the two-year recidivism rates of large offender populations. However, for certain offender groups and over shorter follow-up periods, the under-estimation of recidivism is more marked. The study concludes that current methods of estimating population-level recidivism rates are adequate and do not require wholesale re-calibration to account for either factor.