OBJECTIVE: To examine the incidence of overactive bladder (OAB)-related clinical events for elderly patients receiving pharmacotherapy compared with those receiving non-pharmacologic management. METHODS: Data were obtained from the PharMetrics Patient-Centric Database on continuously benefit-eligible patients aged ³65 years diagnosed with OAB between January 2002 and December 2003. Patients were categorized into 2 cohorts: those receiving pharmacotherapy, including long-acting tolterodine, or immediate- or extended-release oxybutynin and those receiving non-pharmacologic management. Patients were matched 1:1 by the estimated propensity score for OAB pharmacotherapy using a logistic regression model that included selected demographic and clinical characteristics. The incidence of clinical events, including assessment of depression, urinary tract infections (UTIs), and falls or fractures was evaluated. Crude event rates were reported, as well as the risk of an event using Cox proportional hazards models adjusting for important demographic and clinical characteristics. RESULTS: A total of 1681 matched pairs were identified. Mean ± SD age was 78±8 years; 60% were women. After matching, differences in baseline patient characteristics between cohorts were not significant. The incidence of depression (10.2% for pharmacotherapy vs 11.0% for non-pharmacologic management; p=NS) and fractures (8.6% vs 10.1%; p=NS) was numerically lower in the drug-treated cohort, and the incidence of UTI was significantly lower (28.1% vs 36.5%; p<0.0001). Adjusted rates using Cox proportional hazards models were similar, with no difference in the risk of depression (hazard ratio [HR]=0.980; 95% confidence interval [CI], 0.795–1.208; p=NS) or fractures (HR=0.863; 95% CI, 0.691–1.079; p=NS) between the cohorts, and the risk of UTI was 28% lower among pharmacotherapy patients (HR=0.718; 95% CI, 0.636–0.810; p<0.0001). CONCLUSIONS: Pharmacotherapy may impart selected clinical benefits for some elderly patients with OAB compared with non-pharmacologic management. Careful consideration should be given to the selection of treatment approaches in this high-risk population.