The purpose of this review is to evaluate the published literature on adherence to colorectal cancer (CRC) screening with fecal occult blood testing (FOBT) and sigmoidoscopy. Specifically, the review addresses the following: 1) prevalence of FOBT and sigmoidoscopy; 2) interventions to increase adherence to FOBT and sigmoidoscopy; 3) correlates or predictors of adherence to FOBT and sigmoidoscopy; and 4) reasons for nonadherence. Other objectives are to put the literature on CRC screening adherence in the context of recently reported findings from experimental interventions to change prevention and early detection behaviors and to suggest directions for future research on CRC screening adherence. CRC screening offers the potential both for primary and for secondary prevention. Data from the 1992 National Health Interview Survey show that 26% of the population more than 49 years of age report FOBT within the past 3 years and 33% report ever having had sigmoidoscopy. The Year 2000 goals set forth in Healthy People 2000 are for 50% of the population more than 49 years of age to report FOBT within the past 2 years and for 40% to report that they ever had sigmoidoscopy. Thus, systematic efforts to increase CRC screening are warranted. To date, attempts to promote CRC screening have used both a public health model that targets entire communities, e.g., mass media campaigns, and a medical model that targets individuals, e.g., general practice patients. Most of these efforts, however, did not include systematic evaluation of strategies to increase adherence. The data on FOBT show that the median adherence rate to programmatic offers of FOBT is between 40% and 50%, depending on the type of population offered the test, e.g., patients or employees. Approximately, 50% of those initially offered testing in unselected populations will respond to minimal prompts or interventions. A salient issue for FOBT, however, is whether or not the behavior can be sustained over time. Fewer studies examined adherence to sigmoidoscopy. Adherence was highest in relatives of CRC cases and in employer-sponsored programs offered to workers at increased risk of CRC. At present, we know very little about the determinants of CRC screening behaviors, particularly as they relate to rescreening.