Qualitative urinalysis methods of monitoring cocaine use may over-detect frequency of use, possibly decreasing the ability of clinical trials to detect effective treatments. Quantitative urinalysis and newly developed criteria for identifying new cocaine use were evaluated as alternative measures of cocaine use. Urine specimens collected in a cocaine dosing study in non-treatment-seeking subjects (n = 5) and a cocaine treatment trial (n = 37) were analyzed for the cocaine metabolite, benzoylecgonine, with qualitative and quantitative methods. Pharmacokinetic criteria (‘New Use’ rules) were applied to quantitative data to identify occasions of new cocaine use. Results were compared to known cocaine administrations in the laboratory study and to self-reported drug use and qualitative urinalysis for subjects in the clinical trial. New Use criteria correctly identified cocaine administrations in the cocaine dosing study in all but a small number of specimens. In the clinical trial, quantitative urinalysis and estimated New Uses provided more information about patterns and frequency of use than qualitative urinalysis in the different treatment conditions in the clinical trial. Interpretation of quantitative urinalysis with New Use rules appears to be a useful method for monitoring treatment outcome and may be more accurate than traditional qualitative urinalysis in estimating frequency of cocaine use.