Context: Colorectal cancer is an important cause of cancer-related morbidity and mortality in the United States. Screening has the potential to reduce the morbidity and mortality from colorectal cancer through early detection and removal of early-stage cancers or precancerous adenomatous polyps.
Objective: We conducted a systematic review for the US Preventive Services Task Force to assess the effectiveness and cost-effectiveness of different colorectal cancer screening tests.
Data sources: We used recently conducted systematic reviews, the second edition of the Guide to Clinical Preventive Services, the British National Health Service Economic Evaluation database, and focused searches of MEDLINE from 1966 through September 2000 to identify relevant studies for inclusion. We also conducted hand-searches, review of bibliographies, and consultations with context experts to assure completeness.
Study selection: When available, we included the most recent high-quality systematic review and then supplemented that review with a search for more recent articles. Full MEDLINE searches were performed to examine the accuracy of double-contrast barium enema, the rates of complications for each of the available screening tests, and for studies of the cost-effectiveness of screening. Two reviewers examined the results of each of the full searches and determined by consensus which articles should be abstracted into evidence tables.
Data extraction: One reviewer abstracted the information from the final set of studies into evidence tables, and a second reviewer checked them for accuracy.
Data synthesis: Direct evidence from multiple well-conducted randomized trials supports the effectiveness of fecal occult blood testing (FOBT) in decreasing colon cancer incidence and reducing mortality from colorectal cancer compared with no screening for average-risk adults over age 50. Data from well-conducted case-control studies support the effectiveness of sigmoidoscopy and possibly colonoscopy in reducing colon cancer mortality as well. A nonrandomized trial and indirect evidence support the use of combination FOBT and sigmoidoscopy. Indirect evidence from diagnostic accuracy studies suggests that double-contrast barium enema or virtual colonoscopy may also be effective compared with no screening. Data are insufficient to determine with confidence and precision the most effective or cost-effective strategies or the age at which screening should be stopped.
Conclusions: Colorectal cancer screening is effective in reducing mortality from colorectal cancer. Current data are insufficient to determine the most effective or cost-effective strategy for screening, although all major strategies have favorable cost-effectiveness ratios compared with no screening.