Vitamin D pills failed to fend off respiratory infections

Low blood levels of vitamin D have repeatedly been linked with a higher risk of respiratory tract infections, which has helped make vitamin D a popular winter purchase. An association, though, is not proof that topping up the vitamin prevents anything, since people with low levels may differ in many other ways from people with high levels. The only way to settle the question is to randomise healthy people to vitamin D or a control condition and count the infections that follow. This pooled analysis gathered every such trial available at the time to ask whether supplementation genuinely protects.

The authors searched PubMed, Embase and Cochrane databases through January 2013 for randomised controlled trials of vitamin D supplementation with respiratory tract infection as an outcome in healthy populations, also reviewing the reference lists of retrieved papers. Seven trials met the criteria, covering 4827 participants in total. Results were pooled into a single relative risk, with the statistical approach chosen according to how much the trials disagreed with one another. The team also checked whether age, dosing regimen or follow-up length shifted the result, and tested for signs of publication bias.

The pooled relative risk for people taking vitamin D was 0.98, which is statistically indistinguishable from no effect at all. The result barely moved when the analysis accounted for participants' age, the dosing regimen used, or the length of follow-up. Removing any single trial made little difference either, meaning no one study was propping up or masking an effect. There was also no evidence of publication bias, the distortion that arises when trials with negative results quietly go unpublished.

This null sits awkwardly against the consistent observation that people with low vitamin D levels suffer more respiratory infections, and it suggests the link may not be straightforwardly causal, or may matter only in groups these trials did not capture. The authors themselves flag the open questions: larger studies, and clarity on whether baseline vitamin D status, dose and study length change the answer. People who start out deficient are the obvious candidates for a different result. As of this analysis, routine supplementation of healthy people for infection prevention lacked support.

If you are healthy and buy vitamin D mainly as insurance against winter infections, this pooled trial evidence says that purpose was not being served. That is not a verdict on vitamin D's other roles in the body, and it says nothing about people who are genuinely deficient, a group these trials were not designed to isolate. It is a clean reminder that a blood marker tracking with a disease is not the same thing as a pill preventing it.