Vitamin D supplementation to prevent acute respiratory infections: a systematic review and meta-analysis of aggregate data from randomised controlled trials
David A Jolliffe, Carlos A Camargo, John D Sluyter, Mary Aglipay, John F Aloia, Davaasambuu Ganmaa, Peter Bergman, Heike A Bischoff-Ferrari, Arturo Borzutzky, Camilla T Damsgaard, Gal Dubnov-Raz, Susanna Esposito, Clare Gilham, Adit A Ginde, Inbal Golan-Tripto, Emma C Goodall, Cameron C Grant, Christopher J Griffiths, Anna Maria Hibbs, Wim Janssens, Anuradha Vaman Khadilkar, Ilkka Laaksi, Margaret T Lee, Mark Loeb, Jonathon L Maguire, Paweł Majak, David T Mauger, Semira Manaseki-Holland, David R Murdoch, Akio Nakashima, Rachel E Neale, Hai Pham, Christine Rake, Judy R Rees, Jenni Rosendahl, Robert Scragg, Dheeraj Shah, Yoshiki Shimizu, Steve Simpson-Yap, Geeta Trilok-Kumar, Mitsuyoshi Urashima, Adrian R Martineau
The lancet. Diabetes & endocrinology·2021·9(5):276-292
The authors report no dedicated external funding for this publication.
Competing interests
The source discloses financial, employment, advisory or material-support relationships involving Pharma Nord, DSM, Pfizer, Roche. See the publication for the authors and nature of each relationship.
Study snapshot
Design
Systematic review and meta-analysis
Model
Human evidence synthesis
Sample
43 trials with primary-outcome data for 48,488 participants
Intervention
Vitamin D versus placebo or lower-dose control; schedules varied
This publication examines acute respiratory infection risk, serious adverse events. Study population or evidence base: 43 trials with primary-outcome data for 48,488 participants. Intervention or exposure: Vitamin D versus placebo or lower-dose control; schedules varied.
Key findings
The 2021 pooled analysis reported a small reduction in respiratory infection risk. Subgroup patterns did not establish reliable differences between dosing or demographic groups.
What this study can and cannot tell us
The larger 2025 update no longer found a statistically significant overall effect. These analyses share many source trials and should not be counted as independent confirmations.
