Vitamin D supplementation to prevent tuberculosis infection and disease: systematic review and individual participant data meta-analysis of randomised controlled trials

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background

Several randomised controlled trials of vitamin D supplementation for prevention of tuberculosis infection and disease have been conducted, but no individual participant data (IPD) meta-analysis has been conducted.

Methods

We conducted a systematic review, one-stage and two-stage IPD meta-analysis (PROSPERO CRD420251173446) of double-blind, placebo-controlled randomised trials of vitamin D supplementation for prevention of tuberculosis infection or disease. We searched MEDLINE, Embase, Scopus, Cochrane and ClinicalTrials.gov from inception to October 29, 2026. The co-primary outcomes were incident tuberculosis infection (evidenced by tuberculin skin test or Interferon Gamma Release Assay conversion) and incident tuberculosis disease. Mixed-effects models estimated adjusted odds ratios (aORs), accounting for trial and prespecified covariates. Prespecified subgroup analyses investigated effect modification by sex and age, body-mass index, vitamin D status, and BCG vaccination status.

Findings

Four trials contributed IPD for 14,619 participants. Vitamin D did not reduce risk of incident tuberculosis infection (9,557 participants; aOR 0.97, 95% CI 0.81-1.18; p=0.79) or incident tuberculosis disease (14,533 participants; aOR 0.78, 95% CI 0.56-1.10; p=0.15). There was no evidence of effect modification by participant characteristics investigated in sub-group analyses, including in participants with 25-hydroxyvitamin D <25 nmol/L (infection: aOR 0.95, 95% CI 0.67-1.34; disease: 1.01, 0.45-2.30). Serious adverse events and mortality were unaffected.

Interpretation

Vitamin D supplementation did not reduce the risk of incident tuberculosis infection or disease in the populations studied, either overall or within sub-groups. The 95% confidence interval of the effect estimate for prevention of tuberculosis disease remains compatible with a potentially important benefit, and further trials for this indication may be warranted.

Article activity feed