We report a sample pretreatment approach for the analysis of total cocaine residues in wastewater that eliminates the need for two key assumptions often made in estimating cocaine utilization from measurement of its benzoylecgonine metabolite: that benzoylecgonine is neither degraded nor generated during transport in a sewer system, and that it is excreted as a constant fraction of cocaine ingested. By adding NaOH and incubating samples at 55 °C, cocaine and its principal metabolites are efficiently hydrolyzed into ecgonine, anhydroecgonine, and norecgonine. Ecgonine, estimated to represent between 37% and 90% (on a molar basis) of cocaine residues, can be directly determined (without preconcentration via solid-phase extraction (SPE)) by reversed-phase (RP) or hydrophilic interaction liquid chromatography-tandem mass spectrometry (LC/MS/MS). If samples are subjected to SPE, anhydroecgonine can also be determined; this metabolite (and its precursors) represents ≈7% of urinary cocaine residues (based on spot collections from living individuals). Although a reference standard for norecgonine is not commercially available, such nortropanes are also a minor fraction (up to 2%) of urinary cocaine residues. The stability of two human markers (cotinine and creatinine) to the hydrolysis procedure was also investigated. Results obtained by applying the hydrolysis approach for the analysis of total cocaine in an untreated municipal wastewater sample (obtained from Baltimore, MD) were generally in excellent agreement with those obtained from split samples analyzed using a more comprehensive solid-phase extraction RPLC/MS/MS method as described in our previous work. In particular, total tropane-based cocaine residues were found to be hydrolyzed to ecgonine with 98-99% efficiency.