Vitamin and multiple-vitamin supplement intake and incidence of colorectal cancer: a meta-analysis of cohort studies

Abstract

This paper systematically evaluated the association of intake of different vitamins and multiple-vitamin supplements and the incidence of colorectal cancer. Relevant studies were identified in MEDLINE via PubMed (published up to April 2014). We extracted data from articles on vitamins A, C, D, E, B9 (folate), B2, B3, B6, and B12 and multiple-vitamin supplements. We used multivariable-adjusted relative risks (RRs) and a random-effects model for analysis and random effects. With heterogeneity, we looked for the source of heterogeneity or performed sensitivity and stratified analyses. We found 47 articles meeting the inclusion criteria. The multivariable-adjusted RR for pooled studies for the association between the highest versus lowest vitamin B9 (folate) intake and colorectal cancer was 0.88 [95 % confidence interval (95 % CI) 0.81-0.95]. Vitamin D was 0.87 (95 % CI 0.77-0.99); vitamin B6, 0.88 (95 % CI 0.79-0.99); vitamin B2, 0.86 (95 % CI, 0.76-0.97); vitamin A, 0.87 (95 % CI, 0.75-1.03); vitamin C, 0.92 (95 % CI, 0.80-1.06); vitamin E, 0.94 (95 % CI, 0.82-1.07); vitamin B12, 1.10 (95 % CI, 0.92-1.32); vitamin B3, 1.18 (95 % CI, 0.76-1.84). Vitamin B9 (folate), D, B6, and B2 intake was inversely associated with risk of colorectal cancer, but further study is needed. Our study featured unacceptable heterogeneity for studies of multiple-vitamin supplements, so findings were inconclusive.