Specifying the Relationship between Arrestee Drug Test Results and Recidivism

Abstract

Over the last fifteen years, research has confirmed what many criminal justice practitioners have long suspected: among criminal populations, drug use is a predictor of future criminal activity. Improved drug testing technology has significantly enhanced the ability to obtain information about an arrestee's recent drug use. However, the use of this information is a source of considerable debate. Some propose evidence of an arrestee's recent drug use should play a part in decisions about pretrial detention, including what requirements should be imposed as a condition of release. Opponents contend the results future criminality, but only because the drug test information duplicates other risk factors, such as the accused's prior criminal history. The opponents argue that widespread drug testing of arrestees is unnecessary because information on these other risk factors is readily available, cheaper and less intrusive. In this paper, we examine two empirical issues in this debate. First, are the results of drug tests at the time of arrest related to recidivism after the effects of other risk factors have been taken into account? Second, does the ability of drug test results to predict recidivism vary depending on an individual's other risk factors?