U.S. flag

An official website of the United States government, Department of Justice.

Recidivism Rates for Drug Court Graduates: Nationally Based Estimates, Final Report

NCJ Number
Date Published
66 pages
Publication Series
This federally supported study provides nationally representative estimates of recidivism rates for the average drug court graduate and are intended to be used as benchmarks in later research.
This study supported by the U.S. Department of Justice, National Institute of Justice presents a general estimate of recidivism among a nationally representative sample of drug court graduates. Its intent is to assist policymakers by developing a single estimate of recidivism rates from a consistent data source and to create a benchmark for drug court practitioners and researchers. In this study, recidivism was measured as any arrest for a serious offense resulting in the filing of a charge. A major contribution of this study was the use of a single, consistent measure of recidivism drawn from a consistent data source, the internal FBI criminal history database. It is suggested that at least 95 percent of police agencies consistently report data to this database. In order to measure the average recidivism rate for drug court graduates, this study consisted of 2,020 graduates in 1999 and 2000 from 95 drug courts, which is designed to be representative of approximately 17,000 annual drug court graduates. The criminal history of these graduates was analyzed to estimate the likelihood of re-arrest during the first year after graduation and weighted to represent all graduates of the 95 courts during this period. Study results indicate: (1) that within 1 year 16.4 percent of drug court graduates had been arrested and charged with a serious offense, rising to 27.5 percent within 2 years; (2) that in the first year after graduation, drug court graduates averaged 2.23 serious crimes per person and 0.50 serious crimes per person in the first 2 years after graduation; (3) that drug courts with high recidivism rates are serving the most difficult to reach populations; and (4) that recidivism among drug court graduates appears to be related to the size of the drug court. It is believed that these benchmark estimates will assist in efforts to measure drug court impact. However, these estimates should be considered in isolation due, in part, to the fact that drug courts operate in multifaceted environments. Study limitations included underestimating the recidivism rates by not counting all arrests and by not taking into account that not all participants can be matched to their FBI records. In addition, due to definitional issues, the study may overestimate the recidivism rate. In summary, there is no single estimate that can, or should be used to measure whether an individual drug court is successful. Figures and appendices A-C

Date Published: January 1, 2002