C L Ow1, H J Lemar, M J Weaver. 1. Department of Medicine, Fitzsimons Army Medical Center, Aurora, CO 80045-5001.
Abstract
OBJECTIVE: To determine the strength of the evidence in the literature that screening proctosigmoidoscopy reduces colorectal cancer mortality. DESIGN: All English-language studies reporting mortality or survival from screening proctosigmoidoscopy published since 1960 were critically reviewed. DATA EXTRACTION: Fifteen references reported on five studies of screening proctosigmoidoscopy. Two authors independently reviewed each reference using explicit methodologic criteria, particularly for potential sources of bias. MEASUREMENTS AND MAIN RESULTS: Of the five studies, four used historical controls and were susceptible to bias, especially self-selection and lead-time bias. Only the Kaiser-Permanente Multiphasic Health Check-up study collected a representative patient sample from a defined population, had randomly allocated controls, and avoided multiple sources of bias. There was a reduction in mortality from a group of seven potentially postponable causes of mortality, including colorectal cancer, although no difference in overall mortality between screened and control groups was found. This study was not designed to determine specifically the impact of screening proctosigmoidoscopy on mortality from colorectal cancer, and suggested that most of the reduction in colorectal cancer deaths was due to a lower incidence in the screened group, which could not be attributed to polypectomy. The benefit of screening proctosigmoidoscopy in this study, if any, was small. CONCLUSIONS: The evidence in the literature is inadequate to determine whether or not screening proctosigmoidoscopy has an impact on colorectal cancer mortality, but the best available data suggest that the benefit is small, at best.
OBJECTIVE: To determine the strength of the evidence in the literature that screening proctosigmoidoscopy reduces colorectal cancer mortality. DESIGN: All English-language studies reporting mortality or survival from screening proctosigmoidoscopy published since 1960 were critically reviewed. DATA EXTRACTION: Fifteen references reported on five studies of screening proctosigmoidoscopy. Two authors independently reviewed each reference using explicit methodologic criteria, particularly for potential sources of bias. MEASUREMENTS AND MAIN RESULTS: Of the five studies, four used historical controls and were susceptible to bias, especially self-selection and lead-time bias. Only the Kaiser-Permanente Multiphasic Health Check-up study collected a representative patient sample from a defined population, had randomly allocated controls, and avoided multiple sources of bias. There was a reduction in mortality from a group of seven potentially postponable causes of mortality, including colorectal cancer, although no difference in overall mortality between screened and control groups was found. This study was not designed to determine specifically the impact of screening proctosigmoidoscopy on mortality from colorectal cancer, and suggested that most of the reduction in colorectal cancer deaths was due to a lower incidence in the screened group, which could not be attributed to polypectomy. The benefit of screening proctosigmoidoscopy in this study, if any, was small. CONCLUSIONS: The evidence in the literature is inadequate to determine whether or not screening proctosigmoidoscopy has an impact on colorectal cancer mortality, but the best available data suggest that the benefit is small, at best.