“Vitamin D supplementation reduces mortality”

VITAL randomly assigned 25,871 US adults - men aged 50 and over, women aged 55 and over - to 2,000 IU a day of vitamin D3, the form used in most supplements, or a dummy pill. After a median of 5.3 years the two groups had the same rate of death from any cause, the same rate of major heart and blood-vessel events, and the same number of cancer diagnoses [1].

Cancer deaths were the one signal worth noting. VITAL counted 17% fewer of them in the vitamin D group, but that gap was small enough to be chance [1]. Pooling 52 randomised trials covering 75,454 adults found the same direction and a firmer result: about 16% fewer cancer deaths, alongside no effect on deaths from any cause and none on deaths from heart disease [2].

Whether vitamin D helps people who are actually short of it is still open, because most trial volunteers started with healthy levels and so the trials cannot answer the question. One 2026 analysis rebuilt the VITAL and D-Health trials inside UK health records covering more than 185,000 adults: when the modelled group was matched to the vitamin D levels seen in the real trials the result was null, but when it was restricted to people who were deficient the model predicted roughly a fifth to a quarter fewer deaths [3].

That is a modelling study built on observed health records, not a randomised trial, so it points to a hypothesis worth testing rather than a proven benefit. No trial has yet recruited only deficient people and counted how many died.

The practical takeaway is to know your number rather than guess. Most laboratories flag a vitamin D blood level below 20 nanograms per millilitre as deficient, and a result in that range is a reason to talk to a doctor; if your level is already normal, the trial evidence says extra vitamin D is unlikely to add years.