Risk of hospitalisation after primary treatment for prostate cancer.
Academic Article
Overview
abstract
OBJECTIVE: To compare the risk of hospitalisation and associated costs in patients after treatment for prostate cancer. PATIENTS AND METHODS: We identified 29 571 patients aged 66-75 years without significant comorbidity from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database who were diagnosed with localised prostate cancer between 2004 and 2009. We compared the rates of all-cause and treatment-related hospitalisation that occurred within 365 days of the initiation of definitive therapy. We used multivariable logistic regression analysis to identify determinants associated with hospitalisation. RESULTS: Men who underwent radical prostatectomy (RP) rather than radiotherapy (RT) had lower odds of being hospitalised for any cause after therapy [odds ratio (OR) 0.80, 95% confidence interval (CI): 0.74-0.87]. Patients who underwent RP rather than RT had higher odds of being hospitalised for treatment-related complications (OR 1.15, 95% CI: 1.03-1.29). However, men who underwent external beam RT (EBRT)/intensity modulated RT (IMRT) (OR 0.84, 95% CI: 0.72-0.99) had a 16% lower odds of hospitalisation from treatment-related complications than patients undergoing RP. Using propensity score-weighted analyses there was no significant difference in the odds of hospitalisation from treatment-related complications for men who underwent RP vs RT (OR 1.06, 95% CI: 0.92-1.21). Patients hospitalised for treatment-related complications after RT were costlier than patients who underwent RP (Mean $18 381 vs $13 203, P < 0.001). CONCLUSIONS: With the exception of men who underwent EBRT/IMRT, there was no statistically significant difference in the odds of hospitalisation from treatment-related complications. Costs from hospitalisation after treatment were significantly higher for men undergoing RT than RP. Our findings are relevant in the context of penalties linked to hospital readmissions and bundled payment models.