"Consistent with the results reported above, we found that drug court participation led to significantly less drug use on both outcome measures in Table 4-3.6. Not surprisingly, we also found that a greater frequency of drug use at baseline significantly predicted a greater frequency at 18 months for both drug court and comparison offenders. Among other background characteristics, a younger age, male sex, black race, having been classified with depression (based on a multi-item screening tool), and having been classified with anti-social personality disorder (also based on a multi-item tool), all predicted greater drug use on at least one if not both of the outcome measures in Table 4-3.6. On the other hand, our results did not provide any evidence of a relationship between offender social ties (e.g., based on employment/school status, marital status, involvement of blood relatives with crime or drugs) and less drug use; nor did we find that prior criminal history predicted future drug use. Overall, the strongest and most consistent predictors of drug use outcomes were participation in the drug court, less frequent drug use at baseline, and the absence of mental health problems at baseline."