Vitamin D

What it is

Vitamin D supplements come mainly as vitamin D3 (cholecalciferol) or vitamin D2, and trials have tested both [1]. Trials have given it daily, weekly or as large occasional doses, and the dosing pattern appears to matter [2].

What the research says

Bones, fractures and falls. A 2026 BMJ meta-analysis of 69 trials with 153,902 people found that vitamin D alone had no effect on fractures (36 trials, 92,045 people; high-certainty evidence), and that calcium, vitamin D or both had little to no effect on falls [3].

Type 2 diabetes. In three trials designed for people with prediabetes, vitamin D lowered the risk of developing diabetes by 15%, or 3.3 fewer cases per 100 people over three years, and made a return to normal blood sugar 30% more likely [4]. An earlier analysis of eight trials found the benefit mainly in people without obesity [5].

Living longer and cancer. Across 52 trials with 75,454 people, vitamin D did not lower deaths from all causes or from heart disease, but reduced cancer deaths by 16% [1]. A later analysis of 14 trials (104,727 people) found a smaller, non-significant 6% reduction overall, with a 12% reduction in trials that gave vitamin D daily rather than in large occasional doses [2].

Heart disease. Pooling five randomised trials, vitamin D did not reduce major cardiovascular events, strokes or cardiovascular deaths [6].

Infections. A 2025 meta-analysis of 40 trials with 61,589 people found no statistically significant protection against acute respiratory infections [7].

Autoimmune disease. In the VITAL trial of 25,871 older adults, 2,000 IU a day for about five years reduced confirmed autoimmune diseases such as rheumatoid arthritis and psoriasis by 22% [8].

Mood. Across 31 trials with 24,189 people, vitamin D3 slightly reduced depressive symptoms, with larger effects in people with depression and in shorter trials; trials longer than a year showed no effect [9].

Testosterone. Two meta-analyses of randomized trials found vitamin D supplements did not raise total testosterone in men: no significant effect across eight trials (mean difference 0.20, not significant) [10], and no clear effect across 11 comparisons, with low certainty [11]. A 12-week trial in 98 healthy men with vitamin D levels below 75 nmol/L found no change [12], and three small trials in men with heart failure, nursing-home residents and deficient immigrants also found no increase [13].

Doses used in studies

The VITAL trial used 2,000 IU a day [8]. The prediabetes trials used 4,000 IU a day or 20,000 IU a week of cholecalciferol [4]. A single yearly dose of 500,000 IU increased falls and fractures [14].

Safety and interactions

• Very large single doses: older women given 500,000 IU once a year fell 15% more often and had 26% more fractures than those on placebo [14].
• High blood calcium: in studies of very high 'mega-doses', side effects — measured as high calcium in blood or urine — occurred in 1.9% to 18.5% of patients [15].
• Kidney stones and calcium at usual trial doses: in the prediabetes trials, kidney stones, high blood calcium and high urine calcium were not significantly more common with vitamin D, but the trials were not designed to detect safety problems [4].
• Serious side effects: across 38 infection trials, serious adverse events were no more common with vitamin D than with placebo [7].
• Upper limit and long-term use: the upper tolerable intake for the general population is 4,000 IU a day [16]. In 22 trials lasting six months or longer, 3,200 to 4,000 IU a day raised the risk of high blood calcium (risk ratio 2.21, about 4 cases per 1,000 people), and also raised falls and hospital stays; the authors judged safety reporting in these trials to be incomplete [16].
• Pregnancy: a 2024 Cochrane review of 10 trials found the evidence on pre-eclampsia, gestational diabetes and preterm birth very uncertain [17].
• Kidney disease: in adults with chronic kidney disease, vitamin D therapy raised blood calcium and did not reduce deaths from any cause [18].
• Medicines: this page does not yet cover drug interactions; ask a pharmacist or doctor before combining vitamin D with other medicines.

Caveats and open questions

Most people in the fracture trials lived independently and were not at high risk of fractures or falls, and evidence for high-risk people or care-home residents is more limited [3]. The cancer-death benefit depends on which trials are pooled and on daily dosing [2] [1].