You have accessJournal of UrologyGeneral & Epidemiological Trends & Socioeconomics: Practice Patterns, Quality of Life and Shared Decision Making I1 Apr 2016


    View All Author Information


    Overactive bladder (OAB) affects nearly half of older men and women. Anti-cholinergic medications are a mainstay of treatment, yet often have only moderate efficacy and potentially substantial side effects. These side effects are particularly concerning for older adults and may lead to switching among drug agents in pursuit of greater efficacy or fewer side effects. Despite this important clinical problem, no prior study has examined drug switching patterns in the Medicare population. Our objective was to describe variation in medication switching within 6 months of initial pharmacotherapy for OAB symptoms among Medicare Part D beneficiaries.


    We performed a retrospective cohort study of beneficiaries using the Medicare 5% sample. Between 2007-2013, we identified subjects with an outpatient encounter consistent with OAB symptoms and a Medicare Part D drug claim for a first generation (pre-2000 FDA approval: oxybutynin, tolterodine) or a second generation (post-2000 FDA approval: darifenacin, fesoterodine, mirabegron, solifenacin, trospium) agent. No subject had a prior OAB drug claim. We followed subjects for 6 months after the initial claim, and defined switching as any new prescription for a different OAB-directed agent. We used a multivariable log-binomial regression model to estimate the relative probability of medication switch, controlling for beneficiary demographics, comorbidity, and geographic variation.


    During the study period, 15,923 beneficiaries had Part D enrollment for at least 6 months after the index drug claim. The average subject was aged 78 ± 7.5 years, 80% were female, and 89% were white. Tolterodine or oxybutynin constituted 56% of initial prescriptions. Within 6 months, 2,457 (15%) switched agents. In multivariable models, females were 25% more likely to switch, as were nursing home residents (RR 1.29, p <0.001). Switching was also more likely when the prescribing physician was a urologist (RR 1.18, p<0.001) and when the initially prescribed agent was a second generation medication. In sensitivity analyses, out-of-pocket patient cost was not associated with medication switching. The only initial agent not associated with switching within 6 months was mirabegron.


    Switching is common among Medicare beneficiaries treated with medication for OAB symptoms. Changing prescriptions is associated with patient characteristics, provider and initial agent, but not patient cost. Further investigation is required to understand drivers and outcomes of these switching patterns to improve care in this vulnerable population.