Background: Surgery for prostate cancer can lead to adverse outcomes, such as incontinence and erectile dysfunction. Studies have identified hospital variation in survival after surgery for other types of cancer; few studies have assessed hospital variation in outcomes after prostate cancer surgery. We sought to characterize hospital variation in incontinence/erectile dysfunction (IED) and overall survival (OS) after prostate surgery. Methods: Using SEER-Medicare claims, we identified men > 65 years old who had surgery for prostate cancer from 2009-13, excluding those with metastatic disease, chemotherapy, or radiation therapy. We computed the difference in the number of days with claims for incontinence or erectile dysfunction in the year after versus the year before prostate surgery (IED score). Then, we assigned each patient to the hospital where prostate surgery was performed, and derived the mean IED score for hospitals that treated ≥10 patients (n = 239). Hierarchical linear and cox regression models were used to assess the association of patient, tumor, and hospital factors with the IED score and OS, respectively. The Kaplan-Meier method was used to estimate 3-year OS. Results: Among 10,850 Medicare-insured men with prostate cancer, the mean difference in claim-days for incontinence or erectile dysfunction during the year after versus before prostate surgery was +2.34 (standard deviation [SD] 3.72, minimum -18, maximum 54). After rescaling the IED score using a 0 (worst) to 100 (best) scale, the mean unadjusted hospital performance was 52.7 (SD 8.0). Controlling for other factors, hospital, NCI-cancer center designation (comprehensive vs. none), region in the US (south or west vs. northeast), and type of prostate surgery (open vs. not) were associated with a higher IED score. The mean 3-year OS across all hospitals was 98% (SD 3%). Conclusions: This claims-based analysis identified significant variation in incontinence/erectile dysfunction after prostate cancer surgery by hospital, but not by patient socio-demographic factors. The IED score may help identify opportunities to improve outcomes. Efforts to validate claims as indicators of patient-centric outcomes and to explore the reasons for hospital-level variation are needed.