OBJECTIVES: Tolterodine-ER is a newly-released medication for Overactive Bladder/Urinary Incontinence (OAB/UI). This retrospective study investigated the difference in direct health-care cost, hospital utilization, and medication persistence between OAB/UI patients initiated with tolterodine-ER or oxybutynin-ER. METHODS: Newly started adult patients (age>=18) on either tolterodine-ER or oxybutynin-ER from October 2001 to May 2002 with 1-year washout period, were included and followed up until the end of study period or disenrollment. The log-transformed direct healthcare cost (excluding OAB-related pharmacy cost per member per month) was analyzed by ANCOVA. Control variables include demographics, previous hospital utilizations, medication pattern, prior OAB diagnosis and comorbidities. A Cox Proportional Hazard model was applied to examine the effect of different initial treatments on persistence described by time to switch and time to discontinuation. Logistic regression was used to assess the risk of hospitalization associated with the first prescription. RESULTS: Of 1811 patients, 1021 patients started with tolterodine-ER, and 790 with oxybutynin-ER. The average follow-up period for both groups was five months. No significant difference was found in the converted adjusted costs PMPM between tolterodine-ER group (US$602) and oxybutynin-ER group (US$648) with P=0.324. Two groups had similar physician encounter frequency (one visit PMPM) and emergency room visit rate (2%), but tolterodine-ER initiated patients were less frequently hospitalized (9.11% vs. 13.16%, P=0.006). A higher proportion of oxybutynin-ER initiated patients discontinued (56.6% vs. 52.5 %, P=0.08) and switched (7.85% vs. 7.54%, P=0.808). The type of initial therapy did not significantly affect time to discontinuation or time to switch, but initiating tolterodine-ER could reduce the probability of hospitalization by 32% (OR: 0.677, P=0.037). CONCLUSIONS: Initial drug selection did not significantly change direct healthcare cost, medical utilization, and medication persistence in OAB/UI patients. However, patients initiating with oxybutynin-ER were exposed to a higher risk of hospitalization than those initiating with tolterodine-ER.