Getting under the hood of the volume-outcome relationship for radical cystectomy

Purpose: Hospital and surgeon volumes independently explain variations in outcomes for a host of surgical procedures. However, the mediators of the volume effect remain unclear. We assessed whether differences among hospitals could explain some or all of the volume effect on short-term outcomes after cystectomy for bladder cancer. Materials and Methods: Using the Nationwide Inpatient Sample a 20% sampling of hospital discharges in the United States and the American Hospital Association file we applied International Classification of Diseased, 9th revision, clinical modification procedure codes to identify 1,847 patients who underwent cystectomy for bladder cancer in 2003. Multivariable mixed models were fit to quantify the differences in measures of hospital structure (capacity, staffing and health services) by hospital volume. Separate models were fit to determine the impact of accounting for these differences on the volume-outcome relationship. Results: There were substantial differences in hospital structure according to radical cystectomy volume, including those characterizing capacity, staffing levels and the breadth of available health services. For example, 40.7% of low and 87.8% of high volume hospitals for radical cystectomy offered open heart surgery (OR 10.4, 95% CI 1.3-85.3). After adjusting for case mix patients treated at low volume centers were 3.2 times (95% Cl 0.8-13.4) more likely to die postoperatively. Accounting for differences in hospital structure attenuated the volume effect by 59% (OR 1.9, 95% Cl 0.4-8.6). Conclusions: Measurable differences in the availability and breadth of consultative, diagnostic and ancillary services may at least partially explain the association between procedure volume and short-term cystectomy outcomes.