Vieira · Annals of oncology : official journal of the European Society for Medical Oncology 2017 · systematic review and dose-response meta-analysis of prospective cohort studies · n=111 cohort studies

Foods and beverages and colorectal cancer risk: a systematic review and meta-analysis of cohort studies, an update of the evidence of the WCRF-AICR Continuous Update Project.

Cited 451 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis of prospective cohort studies

PubMed 28407090 · doi:10.1093/annonc/mdx171 · record verified 2026-08-27

What was done

As part of the WCRF-AICR Continuous Update Project, researchers conducted a systematic review and dose-response random-effects meta-analysis of prospective cohort studies from PubMed and other databases up to May 31, 2015. They analyzed adjusted relative risk estimates for specific foods and beverages and colorectal, colon, and rectal cancer risk across roughly 400 estimates from 111 unique cohort studies.

What was found

Colorectal cancer risk increased by: - 12% per 100 g/day increase in red and processed meat intake (95% CI: 4% to 21%, I² = 70%, p_heterogeneity < 0.01) - 7% per 10 g/day increase in ethanol intake from alcoholic drinks (95% CI: 5% to 9%, I² = 25%, p_heterogeneity = 0.21) Colorectal cancer risk decreased by: - 17% per 90 g/day increase in whole grains (95% CI: 11% to 21%, I² = 0%, p_heterogeneity = 0.30, 6 studies) - 13% per 400 g/day increase in dairy products (95% CI: 10% to 17%, I² = 18%, p_heterogeneity = 0.27, 10 studies) - Vegetables: RR 0.98 per 100 g/day (95% CI: 0.96 to 0.99, I² = 0%, p_heterogeneity = 0.48, 11 studies; weak effect) - Fish: RR 0.89 per 100 g/day (95% CI: 0.80 to 0.99, I² = 0%, p_heterogeneity = 0.52, 11 studies; driven by one study) No association was found for fruits, coffee, tea, cheese, poultry, or legumes.

Why it matters

This large meta-analysis solidifies evidence that red/processed meat and alcohol increase colorectal cancer risk in a dose-dependent manner, whereas whole grains and dairy products show consistent inverse associations.

Limits

All included data are observational and subject to residual confounding. High heterogeneity was observed for red and processed meat (I² = 70%). The fish association was driven by a single study, the vegetable association was very weak, and total participant counts across all cohorts were not reported in the abstract.

Cited by