Summary: In 22,220 older US adults without diabetes, daily vitamin D 2000 IU did not significantly reduce incident type 2 diabetes over a median 5.3 years (hazard ratio 0.91, 95% CI 0.76 to 1.09), and it had no effect on glycaemic traits at 2 years. A pre-specified meta-analysis of four trials suggested a modest pooled reduction (hazard ratio 0.89, 95% CI 0.80 to 0.99).
PICO Summary
| Element | Detail |
|---|---|
| Population | 22,220 US adults without type 2 diabetes at enrolment (mean age 67.2 years, 51% female, mean BMI 27.5 kg/m2); ancillary analysis (VITAL-T2D) of the double-blind, placebo-controlled VITAL 2×2 factorial trial. |
| Intervention | Cholecalciferol 2000 IU/day (vitamin D arm of the factorial design). |
| Comparison | Matching placebo. |
| Outcome | Incident type 2 diabetes at median 5.3 years: 3.98 vs 4.37 cases/1000 person-years; HR 0.91 (95% CI 0.76 to 1.09), non-significant. No effect on glycaemic traits at 2 years (subcohort n=911). Pre-specified meta-analysis of 4 trials (5205 participants, 936 cases): pooled HR 0.89 (95% CI 0.80 to 0.99). |
Vitamin D to prevent type 2 diabetes
RCT · older adults · median 5.3 years
Daily vitamin D 2000 IU did not significantly reduce incident type 2 diabetes in unselected older adults. A pooled meta-analysis hinted at a modest benefit, but the effect is hypothesis-generating only.
Expert Commentary
The verdict for the primary question is clear and negative. In a general population of older adults who did not have prediabetes selected at entry, daily vitamin D 2000 IU produced no statistically significant reduction in incident type 2 diabetes, and the confidence interval (0.76 to 1.09) comfortably crosses unity. The absence of any effect on glycaemic traits at 2 years reinforces that this was a true null, not merely an underpowered signal. The accompanying meta-analysis pooling four trials does reach significance (HR 0.89, 0.80 to 0.99), but the upper bound hugs the null, the benefit is modest, and most of the contributing weight comes from earlier prediabetes-enriched cohorts, so this should be read as hypothesis-generating rather than as proof of prevention. A key limitation is that VITAL-T2D was an ancillary analysis of a trial designed and powered for cancer and cardiovascular endpoints, not for diabetes incidence, so the primary comparison may itself be underpowered for a small true effect. Can I use this with my patients? For an unselected older adult with normal glucose tolerance and replete vitamin D, no: there is no justification for prescribing vitamin D to prevent diabetes. Reserve supplementation for genuine deficiency. Future trials should target high-risk, vitamin D deficient populations before any preventive claim is entertained.
References
Tobias DK, Pradhan AD, Duran EK, Li C, Song Y, Buring JE, et al. Vitamin D supplementation vs. placebo and incident type 2 diabetes in an ancillary study of the randomized Vitamin D and Omega-3 Trial. Nat Commun. 2025;16(1):3332. doi:10.1038/s41467-025-58721-6