The primary study objective was to ascertain whether a prior finding that the baseline cocaine urine toxicology predicted treatment dropout for cocaine dependent outpatients could be extended to three additional cocaine dependent outpatient treatment samples and whether the urine toxicology also predicted attainment of initial abstinence for the four samples. A secondary objective was to ascertain the extent to which other baseline variables accounted for additional outcome variance over and above that afforded by urine toxicology. To evaluate the first objective, the relationships between the baseline cocaine urine and each of two measures of within treatment response--the completion of treatment or the attainment of initial abstinence--were determined for each of the treatment samples. The second objective was evaluated by a stepwise, hierarchical logistic regression analysis, with the urine toxicology entered in the first step, baseline Addiction Severity Index (ASI) variables in the second step, and achievement of initial abstinence as the outcome. In all four samples, patients with a urine indicative of recent cocaine use were less than half as likely to complete treatment or achieve initial abstinence. Individual ASI baseline variables did not contribute statistically significant variance over and above that predicted by the cocaine urine toxicology. The findings confirm the utility of the initial cocaine urine as a predictor of unfavorable outpatient treatment response.