Studies have been inconsistent in their findings as to whether high-fiber diets protect against colon cancer. Most prospective studies and randomized trials have found no protective effect, contrary to the findings of observational studies. Park and colleagues performed a pooled analysis to evaluate more fully the correlation between dietary fiber intake and colorectal cancer risk.

The authors analyzed data from 13 prospective cohort studies on dietary fiber intake and incident colon cancer. Distinctions were made between whole-grain and refined-grain foods. The 13 studies included 7,328,414 person-years and 8,081 incidences of colon cancer. Follow-up time varied from six to 20 years.

In a model adjusted for age, there was a 16 percent lower risk of colon cancer in the highest fiber consumption quintile compared with the lowest fiber consumption quintile; this was attenuated slightly when adjusted for other factors. Additional adjustments further attenuated the risk reduction until it was no longer statistically significant. Results were similar among European and North American studies. Factors considered included age, sex, body mass index, smoking status, alcohol consumption, and duration of follow-up.

In an analysis of absolute intake cut points, the risk of colorectal cancer was significantly higher in persons consuming at least 10 g of fiber per day compared with those consuming less than 10 g per day. Tumor site was modestly associated with fiber intake, with rectal tumors showing the greatest inverse association. There was no correlation between the development of cancer and intake of refined- versus whole-grain foods.

The authors conclude that, although age-adjusted analysis shows a statistically significant association between dietary fiber intake and colorectal cancer, this association is not maintained after adjustment for other factors.