In this case-control study, we examined the food sources of fiber and fiber solubility to determine whether particular components of dietary fiber were differentially associated with risk of colon and rectal cancer. In Western New York, cases with pathologically confirmed, single, primary cancers of the colon and rectum as well as age-, sex- and neighborhood-matched controls were interviewed from 1975–1986. The sample included 428 colon case-control pairs (223 females, 205 males) and 422 rectal case-control pairs (145 females, 277 males). Subjects were interviewed regarding usual quantity and frequency of consumption of foods. For the colon, risk decreased with intake of grain fiber for both females and males and with intake of fruit/vegetable fiber for males only. Insoluble grain fiber was more strongly associated with risk than soluble grain fiber. For the rectum, fruit/vegetable fiber was associated with decreased risk, whereas grain fiber was not. There was no difference in risk for soluble and insoluble fiber components for the rectum. Analysis of risk associated with fiber by food source and by components of the fiber may provide insight into possible mechanisms of a fiber effect on cancer of the colon and rectum.


This investigation was supported in part by United States Public Health Service Grants CA 11535 and CA 09051-11 from the National Cancer Institute, Department of Health and Human Services. Part of these data in preliminary form was presented at the annual meeting of the Society for Epidemiological Research, 1989.

This content is only available via PDF.