Relation of hospital volume to colostomy rates and survival for patients with rectal cancer

Abstract

Background: Postoperative mortality after some types of cancer surgery is inversely related to the number of operations performed at a hospital (i.e., hospital volume). This study assessed the association of hospital volume with colostomy rates and survival for patients with rectal cancer in a large representative cohort identified from the California Cancer Registry. Methods: We identified 7257 patients diagnosed from January 1, 1994, through December 31, 1997, with stage I-III rectal cancer who underwent surgical resection. Registry data were linked to hospital discharge abstracts and ZIP-code-level data from the 1990 U.S. Census. Associations of hospital volume with permanent colostomy and 30-day mortality were assessed with the Mantel-Haenszel trend test and logistic regression. Overall survival was examined with the Kaplan-Meier method and a multivariable Cox proportional hazards model. Multivariable analyses adjusted for demographic and clinical variables and patient clustering within hospitals. All tests of statistical significance were two-sided. Results: In unadjusted analyses across decreasing quartiles of hospital volume, we observed statistically significant increases in colostomy rates (29.5%, 31.8%, 35.2%, and 36.6%; P <001). The adjusted risks of permanent colostomy (odds ratio [OR] = 1.37, 95% confidence interval [CI] = 1.10 to 1.70), 30-day mortality (OR = 2.64, 95% CI = 1.41 to 4.93), and 2-year mortality (hazard ratio = 1.28, 95% CI = 1.15 to 1.44) were greater for patients at hospitals in the lowest volume quartile than for patients at hospitals in the highest volume quartile. Stratification by tumor stage and comorbidity index did not appreciably affect the results. Adjusted colostomy rates varied statistically significantly (P