To develop the risk-assessment model, researchers used data from two large population-based case-control studies. Several factors that have been previously associated with colorectal cancer risk were shown to be predictive of a colorectal cancer diagnosis in those two studies, including age; family history of colorectal cancer; consumption of vegetables; body mass index; cigarette smoking; use of aspirin or other non-steroidal anti-inflammatory drugs; physical activity; use of hormone replacement therapy; previous history of sigmoidoscopy and/or colonoscopy; and history of polyps. Estimates of relative risk (comparisons of risk in one group to another) from the case-control studies were combined with population-based data on colorectal cancer incidence from NCI's SEER (Surveillance, Epidemiology and End Results) cancer registries to make the model broadly applicable in the United States.

"This colorectal cancer risk model should provide physicians and their patients a new tool to help make informed decisions about cancer screening and other cancer prevention strategies. It may also assist policy makers in evaluating the usefulness of current and future population colorectal cancer screening approaches," said Andrew Freedman, Ph.D., lead author of the paper that describes the development of the risk-assessment model.


To test the accuracy of the risk-assessment model, the researchers compared expected numbers of colorectal cancer cases predicted by the model to the observed numbers of cases identified in the NIH-AARP Diet and Health Study, a large study that follows AARP members and collects information about nutrition and health. From information about individual risk factors that was collected when participants entered the study, the researchers used the new model to estimate the number of men and women who would be expected to develop colorectal cancer. According to Ruth Pfeiffer, Ph.D., who was the senior author of the validation study, "The colorectal cancer risk-assessment tool predicted the numbers of colorectal cancer diagnoses well overall, and in most risk categories."

Because the majority of participants in the two case-control studies used to develop the model were non-Hispanic whites age 50 or older, the researchers were unable to estimate relative risks for other age and racial/ethnic groups. However, there are plans to expand the tool to include these populations in the future. In addition, the tool is not applicable to individuals with certain gastrointestinal disorders (such as ulcerative colitis or Crohn's disease), certain inherited genetic conditions (such as familial adenomatous polyposis or hereditary nonpolyposis colorectal cancer) or a personal history of colorectal cancer. These conditions are known to carry a very high risk of developing colorectal cancer.

More Industry Headlines