Association of Hospital Procedure Volume and Outcomes in Patients with Colon Cancer at High Risk for Recurrence

BACKGROUND: Studies that use registry data have demonstrated superior long-term overall survival after curative surgical resection of colon cancer at hospitals where the volume of such surgeries is high. However, because such administrative data lack information on cancer recurrence, the true nature of this relation remains uncertain. OBJECTIVE: To determine whether hospital procedure volume predicts long-term outcomes of colon cancer surgery. DESIGN: Nested cohort study within a randomized clinical trial. SETTING: Intergroup 0089 national adjuvant colon cancer study conducted between 1988 and 1992. PATIENTS: 3161 patients with high-risk stage II and stage III colon cancer. MEASUREMENTS: Overall survival and recurrence-free survival, by hospital procedure volume as defined by Medicare claims data. RESULTS: With a median follow-up of 9.4 years, 5-year overall survival significantly differed across tertiles of hospital procedure volume (63.8% for patients who had resection at low-volume hospitals compared with 67.3% at high-volume hospitals; P = 0.04). After adjustment for other predictors of colon cancer outcome, the hazard ratio for overall mortality in patients treated at low-volume centers was 1.16 (95% CI, 1.03 to 1.32). However, the risk for cancer recurrence was not associated with hospital procedure volume. Five-year recurrence-free survival was 63.9% for patients who had resection at low-volume hospitals compared with 63.0% at high-volume hospitals (adjusted hazard ratio, 1.03 CI, 0.89 to 1.18). These findings did not materially change after stratification by other potential demographic and clinical predictors of outcome. CONCLUSIONS: According to prospectively recorded data from a large clinical trial, patients whose colon cancer was resected at low-volume hospitals experienced a higher risk for long-term mortality; however, this increased mortality was not attributable to differences in colon cancer recurrences.