Effect of Vitamin D Supplementation on the Incidence of Diabetes Mellitus.

Systematic review and meta-analysis of randomized trials assessing vitamin D supplementation for prevention of type 2 diabetes mellitus. Adults included people with prediabetes as well as participants with normal glucose tolerance or average risk at baseline, across diverse settings and with varying age and BMI profiles.

Vitamin D supplementation was evaluated across randomized trials using multiple formulations, including cholecalciferol (vitamin D3), ergocalciferol (vitamin D2), and an active analogue (eldecalcitol). Doses ranged from lower-dose regimens to moderate/high-dose supplementation (at least 1000 IU/day), with some trials using bolus dosing and some combining vitamin D with calcium.

Overall, vitamin D supplementation did not reduce incident type 2 diabetes mellitus when all populations and doses were pooled: 3,424 cases; RR 0.96 (95% CI, 0.90-1.03); P = 0.30; I2 = 3%. In post hoc subgroup analyses, moderate/high-dose vitamin D (≥1000 IU/day) was associated with lower diabetes incidence: 1,019 cases; RR 0.88 (95% CI, 0.79-0.99); P = 0.03; I2 = 0%, whereas lower-dose supplementation was not effective: RR 1.02 (95% CI, 0.94-1.10); P = 0.68; I2 = 0%. Benefit also appeared restricted to non-obese participants, with BMI <30 showing RR 0.68 (95% CI, 0.53-0.89) versus BMI ≥30 showing RR 0.98 (95% CI, 0.83-1.16).

The pooled analysis showed no overall effect, and the apparent benefit came from post hoc subgroup analyses by dose, BMI, and prediabetes status. Trial populations, vitamin D formulations, and cointerventions such as calcium were heterogeneous, which limits direct comparability and generalizability. The authors also called for participant-level analyses, indicating uncertainty about effect modifiers and the robustness of subgroup findings.