Using data on arrestees in New York City, this study examined whether recent drug use is significantly associated with pretrial misconduct and whether consideration of recent drug use enhances risk classification among a sample of persons on pretrial release.
To measure recent drug use, urine samples were collected from persons shortly after their arrest and tested for four drugs: heroin, cocaine, PCP, and methadone. Of the 2,606 cases considered in this analysis, 1,967 (75.5 percent) had at least 1 day free prior to disposition. Two measures of pretrial misconduct were considered: whether a defendant was rearrested prior to case disposition and whether a defendant failed to appear for a scheduled court data. Censored probit models were used to estimate the statistical association between drug test results and pretrial misconduct. Results show that drug test results were significantly associated with pretrial misconduct over and above the information typically available to judges at the time release decisions were made. Some implications of these findings for pretrial decision-making are discussed. 8 tables and 19 references