Impact of vitamin D supplementation on all-cause mortality: Randomized trials revisited

Vitamin D insufficiency and deficiency are common worldwide and linked to adverse health outcomes, including higher all-cause mortality. Two large randomized controlled trials (VITAL and D-Health), conducted in mostly vitamin D-sufficient populations, found no mortality benefits of vitamin D supplementation. This study aims to estimate the expected effects of vitamin D supplementation in target populations with vitamin D insufficiency or deficiency..

We emulated the VITAL and D-Health trials using data from the UK Biobank cohort to estimate expected effects of the observed increases in serum 25-hydroxyvitamin-D (25(OH)D) concentrations by 30 nmol/L and 38 nmol/L. In alternative analyses, study populations meeting the trial inclusion criteria (n = 237,502 and 185,809) were either weighted to yield distributions of 25(OH)D as observed in the trials, or restricted to people with vitamin D insufficiency or deficiency. Expected effects on all-cause mortality over the mean trial follow up times (5.3 and 5.7 years) were estimated using Cox models.

Emulated trials with study populations weighted to the 25(OH)D distributions of the original trials yielded null results similar to those reported (hazard ratios [HR] 0.97 [95 % CI: 0.92-1.02] and 1.02 [95%CI: 0.97-1.07]). In contrast, major mortality reduction was expected in emulated trials that were restricted to people with vitamin insufficiency (HR 0.85 [95%CI: 0.79-0.91] and 0.81 [95%CI: 0.76-0.86]) or deficiency (HR 0.79 [95%CI: 0.72-0.87] and 0.75 [95%CI: 0.69-0.81])..

Null effects of vitamin D supplementation were to be expected in trials conducted in vitamin D sufficient populations. Emulated trials suggest a potential for major mortality reduction in vitamin D insufficient and deficient populations..