Background: Understanding the relationship of multivitamin use with diabetes risk is important given the wide use of multivitamin supplements among U.S. adults.

Methods: We prospectively examined supplemental use of multivitamins and individual vitamins and minerals assessed in 1995-1996 in relation to self-reported diabetes diagnosed after 2000 among 232,007 participants in the NIH-AARP Diet and Health Study. Multivitamin use was assessed by food frequency questionnaire at baseline. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated by logistic regression models, adjusted for potential confounders. In total, 14,130 diabetes cases diagnosed after year 2000 were included in the analysis.

Results: Frequent use of any multivitamins was not associated with risk of diabetes after adjustment for potential confounders and uses of individual supplements. Compared with nonusers of any multivitamins, the multivariate ORs among users were 1.07 (95% CI: 0.94-1.21) for taking <1 time/week, 0.97 (0.88-1.06) for 1-3 times/week, 0.92 (0.84-1.00) for 4-6 times/week, and 1.02 (0.98-1.06) for ≥7 times/week (P for trend=0.64). Significantly lower risk of diabetes was associated with the use of vitamin C or calcium supplements. The multivariate ORs comparing daily users with nonusers were 0.91 (0.86-0.97) for vitamin C supplements and 0.85 (0.80-0.90) for calcium supplements. Use of vitamin E or other individual vitamin and mineral supplements were not associated with diabetes risk.

Conclusions: In this large cohort of U.S. older adults, multivitamin use was not associated with diabetes risk. The findings of lower diabetes risk among frequent users of vitamin C or calcium supplements warrant further evaluations.