To evaluate clinician- and geographic-level variation in utilization of sacral neuromodulation (SNM) and onabotulinumtoxinA injections as index treatment for overactive bladder (OAB) among United States Medicare beneficiaries.

This is a cross-sectional study of a 100% sample of fee-for-service Medicare beneficiaries undergoing first-time SNM test procedures or onabotulinumtoxinA injections from 2014-2016. The primary outcomes were clinician- and geographic-level variation in utilization of SNM or onabotulinumtoxinA injections. Secondary outcomes included variation in use of SNM test procedures (percutaneous nerve evaluation [PNE] and stage 1 permanent tined lead placement [stage 1]). Geographic region was defined using hospital referral regions (HRRs). Mixed-effect logistic regression models were used to calculate the median odds ratio (MOR) for clinician and HRR (higher MOR = greater variability between groups) and to identify individual-level predictors of utilization.

Overall, 48,580 Medicare beneficiaries underwent SNM (47.1%) or onabotulinumtoxinA injections (52.9%) during the study period. There was a considerable amount of variation according to clinician and HRR, which were more influential than patient-level characteristics in the type of procedure received. The adjusted MOR for SNM versus onabotulinumtoxinA injections was 33.1 for clinician and 4.24 for HRR. For stage 1 versus PNE, the MOR was 13.12 for clinician and 2.44 for HRR.

Among Medicare beneficiaries undergoing first-time SNM or onabotulinumtoxinA injections, there was considerable variation according to performing clinician and geographic region. These findings suggest that non-clinical variables—who a patient sees and where they live—are significant drivers of minimally invasive OAB therapy utilization in the United States.