INDEPENDENT THINKING. WHOLE-FOOD LIVING.NO ANIMALS / NO OIL
THENANODIET
← Oncology / Health Desk

Fiber sources and colorectal cancer: the estimates are not interchangeable

A source-specific dose-response review separates cereal fiber from vegetable, fruit, and legume fiber.

Whole-food ingredients prepared without added oil

Illustrative food image, not a photograph of the study intervention.

A different question from whole-grain servings

This meta-analysis examined fiber sources rather than treating every gram of fiber as interchangeable. It included ten prospective studies and restricted eligible publications to those reporting cereal, vegetable, fruit, and legume fiber, an effort to make comparisons more comparable. Searches ran through August 2018. The question differs from asking how many servings of whole grains someone should eat: cereal fiber is the measured exposure, not a standardized whole-grain meal. Nor was this a randomized food trial. The results describe relationships between reported intake and later colorectal outcomes, leaving causal interpretation more limited than the dose-response language might suggest.

Oh H, Kim H, Lee DH, Lee A et al.. The British journal of nutrition, 2019. PMID 31495339 · DOI

Cereal fiber and the boundary of significance

For each 10 g/day higher cereal fiber intake, the summary relative risk of colorectal cancer was 0.91, with a 95% confidence interval of 0.82 to 1.00. That upper boundary deserves to appear alongside the point estimate. In analyses adjusted for folate intake, the cereal-fiber association remained statistically significant: relative risk 0.89, confidence interval 0.80 to 0.99. These are relative estimates, not absolute percentage-point reductions in cancer occurrence. The abstract judged the evidence strongest for fiber from cereals or grains, but it did not demonstrate that adding a particular bowl of oats prevents colorectal cancer.

Oh H, Kim H, Lee DH, Lee A et al.. The British journal of nutrition, 2019. PMID 31495339 · DOI

Why a larger point estimate is not a winner

Other sources had wider or null-crossing intervals. Per 10 g/day, vegetable fiber had a relative risk of 0.95 (95% CI 0.87–1.03), fruit fiber 0.91 (0.78–1.06), and legume fiber 0.84 (0.63–1.13). The legume estimate looks most favorable numerically, but its uncertainty does not justify naming it the best source for prevention. Vegetable and fruit fiber also showed evidence of nonlinear dose-response patterns. A single incremental estimate cannot capture every shape of association. These comparisons favor careful labels over a league table that ranks foods by their most attractive point estimates alone.

Oh H, Kim H, Lee DH, Lee A et al.. The British journal of nutrition, 2019. PMID 31495339 · DOI

Cancer and adenomas are different endpoints

The review separately considered adenomas, including whether they were newly detected or recurrent. Vegetable and fruit fiber were inversely associated with incident adenoma, with relative risks of 0.84 and 0.78 per 10 g/day, respectively; their confidence intervals excluded 1. The cereal estimate for incident adenoma was 0.81, but its interval was 0.54 to 1.21. The abstract did not find associations with recurrent adenoma. These distinctions prevent a cancer-incidence association from becoming a claim about recurrence. An oat-groat meal illustrates a grain-centered dinner, not a tested prescription for preventing either cancer or a returning adenoma.

Oh H, Kim H, Lee DH, Lee A et al.. The British journal of nutrition, 2019. PMID 31495339 · DOI