PERIODIC HEALTH EXAMINATION, 1994 UPDATE .2. SCREENING STRATEGIES FORCOLORECTAL-CANCER

Objective: To make recommendations on the effectiveness of screening f or colorectal cancer in asymptomatic patients over 40 years of age. Op tions: Multiphase screening that begins with test for fecal occult blo od, uniphase screening with sigmoidoscopy and uniphase screening with colonoscopy. Options included screening repeated at different interval s and different procedures for patients with selected risk factors. Ou tcomes: Rates of death, death from cancer and cancer detection; compli ance, feasibility and accuracy of each manoeuvre. Evidence: A MEDLINE search for articles pubished between January 1966 and June 1993 with t he use of MeSH terms ''screening'' and ''colorectal neoplasia,'' a che ck with the reference sections of review articles published before Jun e 1993 and a survey of content experts. Articles were weighted accordi ng to the Canadian Task Force on the Periodic Health Examination level s of evidence. Values: The highest value was assigned to manoeuvres th at lowered the rate of death from cancer and had a low rate of false-p ositive results and acceptable cost and compliance. Recommendations we re determined by consensus of the authors, members of the task force a nd colorectal cancer experts. Benefits, harms and costs: There is evid ence that annual fecal occult blood testing with the use of the rehydr ated Hemoccult test has a small but significant benefit in lowering th e rate of death from cancer after more than 10 years of screening; how ever, the high rate of false-positive results (9.8%) and the poor sens itivity of annual (49%) and biennial (38%) screening make this a poor method for detecting colorectal cancer. There is fair evidence that sc reening with sigmoidoscopy may improve survival rates; however, this m ay be due to volunteer bias. The high cost of and poor compliance with colonoscopic screening make this an unfeasible strategy. Recommendati ons: There is insufficient evidence to support the inclusion or exclus ion of fecal occult blood testing or sigmoidoscopic or colonoscopic sc reening of asymptomatic patients over 40 years of age. There is fair e vidence to support screening with colonoscopy of patients in kindreds with the cancer family syndrome and patients with ulcerative colitis. Randomized controlled trials are needed to determine the benefit of sc reening with sigmoidoscopy. Development of better risk stratification for screening is a high research priority. Validation: These recommend ations are unchanged from the task force recommendations made in 1989 and are similar to those of the US Preventive Services Task Force. The American Cancer Society, however, recommends annual screening with th e Hemoccult test and screening with flexible sigmoidoscopy every 3 to 5 years in patients over 40 years of age. Sponsor: These guidelines we re developed and endorsed by the Canadian Task Force on the Periodic H ealth Examination, which was funded by Health Canada.