Objective: To test the hypothesis that individual surgical volume (SV) is an independent predictor of radical prostatectomy (RP) total charges.
Methods: We used the Florida State Inpatient Data File. ICD-9 codes 60.5 (RP) and 185 (prostate cancer) identified all men treated with RP for prostate cancer between January 1 and December 31, 1998. Among 1,923,085 records, 3167 RPs were selected. SV represented the predictor. Total RP charges represented the outcome. Age, race, and comorbidity represented covariates. Univariate and multivariate linear regression models were used.
Results: All 3167 RPs were performed by 81 surgeons. SV ranged from 2 to 162 (mean, 68). Charges were 4755 dollars to 140,201 dollars (mean, 18,200 dollars). In the multivariate model, each SV increment corresponding to one RP reduced hospital charges by 25 dollars (p < or = 0.001).
Conclusions: Redistribution of RPs from low to high SV users could result in significant savings. For example, 4 million dollars could be saved if 1000 RPs were redistributed from surgeons with an SV of 18 to surgeons with an SV of 200.