OBJECTIVE: To examine levels of persistence and compliance with long-acting tolterodine (TOL) versus immediate- or extended-release oxybutynin (OXY-I or OXY-X) among commercially-insured patients with overactive bladder (OAB). METHODS: Data were obtained from the PharMetrics Patient-Centric Database on patients diagnosed with OAB who started therapy with TOL, OXY-I, or OXY-X between January 2001 and December 2002. Medication adherence was tracked for 12 months. Persistence with medication was calculated based on the time from the index prescription to first discontinuation. Discontinuation was defined as a gap in therapy exceeding 2 times the therapy-days supplied on the previous prescription. Compliance was assessed as the ratio of the total number of days supplied to the number of days persistent with medication. Results were compared between groups descriptively. Chi-square tests were performed comparing the proportions persistent at 1, 2, 3, and 12 months. RESULTS: A total of 15,394 TOL, 7,934 OXY-X, and 7,963 OXY-I patients were available for analysis. Compliance did not differ between TOL (77.4%) and OXY-X (74.3%), but was substantially lower for OXY-I patients (60.9%). The mean (±SD) number of days on therapy was higher among TOL (139±132) than OXY-X (115±122) or OXY-I (60±85) patients. Persistence dropped more sharply over time among OXY-I and OXY-X than among TOL patients. At 1 month, 94.1% of TOL patients were persistent, declining to 45.9% at 3 months, and 19.9% at 12 months. Among OXY-X patients, 89.0%, 38.2%, and 13.9% were persistent after 1, 3, and 12 months, respectively. Corresponding figures for OXY-I patients were 65.6%, 16.6%, and 4.3%. Between-group differences (ie, OXY-X and OXY-I vs TOL respectively) were statistically significant (p values <0.0001) at each time point. CONCLUSIONS: Although compliance was comparable among TOL and OXY-I patients, TOL was associated with longer duration of use relative to OXY-X and OXY-I.