COLORECTAL cancer (CRC) is a major cause of morbidity and mortality, accounting for 140000 new cases and 60 000 deaths per year in the United States.1 At present, the most promising strategy to reduce this burden is some form of periodic screening. The goal of screening is early detection of surgically curable cancers, as well as their likely precursors, adenomatous polyps; both may be present for years before disseminated, incurable cancer develops.
Sigmoidoscopic screening at intervals of 3 to 5 years has been widely recommended2-5 because sigmoidoscopy has high sensitivity for detecting early cancers and adenomas. However, routine sigmoidoscopic screening is controversial6-8 because there is no direct evidence from a ramdomized controlled trial to show that screening reduces CRC mortality. Also, the effort involved in applying sigmoidoscopic screening is very large. Thus, despite long-standing recommendations, sigmoidoscopic screening is performed in less than 5% to 10% of eligible