Vitamin D

We grade the evidence separately for each outcome; the evidence grading method explains what each grade requires.

Does vitamin D affect overall cognition?

Evidence is mixed: pooled trials show a small effect, while the largest single trials show none. A preregistered 2024 meta-analysis of 24 randomized trials with 7,557 participants from mixed healthy and vulnerable populations (mean age 65.21 years, 78.54% women) found that vitamin D improved global cognition compared with control, with a small effect size (Hedges' g = 0.128), and no effect on specific cognitive domains [1]. In subgroups, the effect was larger in vulnerable populations (g = 0.414) and in people who were vitamin D-deficient at the start (g = 0.480) [1].

In two VITAL substudies of adults aged 60 or older (3,424 tested by phone over 2.8 years and 794 in person over 2.0 years), 2,000 IU of vitamin D3 a day did not change the annual rate of change in a global composite score compared with placebo (mean difference 0.01; 95% CI -0.01 to 0.02), the prespecified primary outcome [2]. VITAL used study product donated by Pharmavite LLC [8]. The result did not depend on baseline 25(OH)D (p-interaction = 0.84) [2]. Black participants (19%) on vitamin D kept cognitive scores better than those on placebo, a subgroup finding the authors say needs confirmation [2]. In DO-HEALTH, a factorial trial in 2,157 adults aged 70 or older across Europe, 2,000 IU a day for 3 years had no effect on the Montreal Cognitive Assessment (MoCA), one of six primary outcomes [3]. DO-HEALTH was co-funded by DSM Nutritional Products, Roche, NESTEC, Pfizer and Streuli, alongside public funders [3].

The VitaMIND trial randomized 620 adults aged 50 or older with mild to moderate vitamin D deficiency and early cognitive impairment, a clinical population, to vitamin D or placebo for 24 months; it found no benefit on executive function, its primary outcome, or on other cognitive measures [5]. In a secondary analysis of a 2-year trial in 273 adults aged 60 or older after knee replacement, researchers 2,000 IU with 800 IU a day and found no difference in any cognitive endpoint, including in the subgroup that started out deficient [4]. A 5-year Finnish trial in 2,492 largely vitamin D-sufficient older adults found that medium or high daily doses did not change how many people were diagnosed with dementia [6].

For healthy adults outside these groups the evidence is thin: a 2022 systematic review of trials in healthy non-athlete adults did not pool results and called for more trials [7].

Does vitamin D affect mood?

In VITAL-DEP, 18,353 US adults aged 50 or older without clinically relevant depressive symptoms at baseline took 2,000 IU of vitamin D3 a day or placebo for a median of 5.3 years, with study product donated by Pharmavite LLC [8]. Mood scores on the PHQ-8 questionnaire did not change between groups (mean difference 0.01 points; 95% CI -0.04 to 0.05), and the risk of depression or clinically relevant symptoms was similar (hazard ratio 0.97; 95% CI 0.87 to 1.09) [8]. In the D-Health trial in Australians aged 60-84, monthly vitamin D3 for 5 years did not change PHQ-9 scores compared with placebo (mean difference 0.02; 95% CI -0.06 to 0.11) in 20,487 participants [9].

In D-Health subgroups, participants with a predicted 25(OH)D below 50 nmol/L had a lower risk of starting antidepressants with vitamin D, though the confidence interval included no difference, while in those at 50 nmol/L or above it was higher; the difference between subgroups was statistically significant (p-interaction = 0.01) [9].

Meta-analyses that pool clinical populations reach a different answer. A 2023 meta-analysis of 41 placebo-controlled trials (n = 53,235), including clinical populations and people with systemic diseases, found lower depressive-symptom scores with vitamin D (Hedges' g = -0.317), but heterogeneity was high (I2 = 88.16%), most trials had risk-of-bias concerns, and the certainty of evidence was rated very low [10]. A 2024 dose-response meta-analysis of 31 trials with 24,189 participants with and without depression found short-term reductions in depressive-symptom scores, but no significant effect in trials lasting 24 to 52 weeks (SMD -0.13; 95% CI -0.28 to 0.02) or longer than 52 weeks (SMD 0.14; 95% CI -0.16 to 0.44) [11].

Does vitamin D affect muscle strength?

Evidence is mixed, and these meta-analyses included trials that also gave calcium [12][13][14]. A 2014 meta-analysis of 30 randomized trials with 5,615 people (mean age 61.1 years) found a small increase in global muscle strength with vitamin D compared with placebo or control (SMD 0.17) and no effect on muscle mass or muscle power [12]. The effect was larger in people with 25(OH)D below 30 nmol/L and in people aged 65 or older (SMD 0.25; 95% CI 0.01 to 0.48) than in younger people (SMD 0.03; 95% CI -0.08 to 0.14) [12].

A larger 2021 meta-analysis of 54 placebo-controlled trials with 8,747 people, excluding athletes, found no benefit [13]. People on vitamin D took slightly longer on the timed up-and-go test (0.15 seconds) and had lower maximum knee flexion strength (-3.3 Newton) than people on placebo, and subgroups with low vitamin D levels did not differ from placebo [13]. The authors wrote that vitamin D may have adverse effects on muscle health [13].

A 2011 meta-analysis of 17 trials with 5,072 participants found no effect on grip or proximal lower-limb strength in adults with 25(OH)D above 25 nmol/L, while two studies in deficient participants (below 25 nmol/L) showed a large effect on hip muscle strength (SMD 3.52) [14]. In athletes, a meta-analysis of 8 trials with 284 participants found no overall effect on muscle strength (SMD 0.05), with an effect on lower-limb strength in a subgroup analysis (SMD 0.55) [15].

What the evidence does not show

The large trials above did not find that 2,000 IU a day improved cognitive or mood scores in older adults [2][3][8], and the pooled analysis found no effect on specific cognitive domains [1]. Sleep quality is not graded: a 2022 review pooled only 3 randomized trials in mixed populations (Pittsburgh Sleep Quality Index mean difference -2.33 versus placebo) [16].

The large trials on this page used vitamin D3 [2][3][8][9]. None of the studies cited here compared morning and evening intake.

Regulators and reference bodies give the following upper levels for adults, and for Canada a product maximum. Above an upper intake level, the potential risk of adverse effects may increase [33].

Table: Upper levels and product maximums by market (adults unless stated)

The US committee noted that 10,000 IU (250 micrograms) a day was not associated with classic toxicity, but set the limit lower in view of emerging data on adverse outcomes, such as all-cause mortality, at lower intakes [33]. EFSA based its limit on persistent high urinary calcium (hypercalciuria), with a lowest adverse effect level of 250 micrograms a day, and judged that European populations are unlikely to exceed it, except regular users of high-dose supplements [20].

Toxicity is mostly a supplement problem [23]. ODS reports that vitamin D toxicity can cause hypercalcemia (high blood calcium), hypercalciuria and high 25(OH)D, and in extreme cases kidney failure, calcification of soft tissues, heart rhythm problems and death, and that it almost always results from supplements rather than sun exposure [23]. A clinical review lists confusion, apathy, recurrent vomiting, abdominal pain, frequent urination, thirst and dehydration as the most common symptoms, and notes that 25(OH)D above 375 nmol/L is the hallmark of overdosing [19].

High doses have produced adverse findings in trials [17][18]. In a double-blind, placebo-controlled trial in 2,256 community-dwelling women aged 70 or older at high risk of fracture, a 500,000 IU dose given once a year increased the risk of both falls and fractures compared with placebo, the outcomes it was meant to reduce [17]. In a 3-year trial in 311 healthy adults aged 55 to 70, 4,000 or 10,000 IU a day resulted in lower radial bone density than 400 IU a day, with no difference in bone strength; the authors said more research is needed to know whether this is harmful [18].

Health Canada's high-dose vitamin D monograph requires products to tell people taking antacids, anticonvulsants, digoxin (a medicine with a narrow safe range), cholestyramine, colestipol, mineral oil, steroids, statins or thiazide diuretics to ask a health care practitioner before use [27].

• Thiazide diuretics. Combined with vitamin D supplements, which increase calcium absorption from the gut, these diuretics might lead to hypercalcemia, especially in older adults and people with reduced kidney function or hyperparathyroidism [23]. A systematic review reported the same risk with calcium and vitamin D supplements [21].
• Statins, such as atorvastatin. In 16 patients taking atorvastatin, 800 IU of vitamin D a day for 6 weeks lowered blood concentrations of atorvastatin and its active metabolites, while LDL and total cholesterol were also lower [22]. ODS says vitamin D might reduce the potency of atorvastatin, lovastatin and simvastatin [23].
• Orlistat and steroids. Orlistat can reduce vitamin D absorption, and steroids can reduce calcium absorption and impair vitamin D metabolism [23].

These sources describe what has been reported, not what is right for any individual. If you take any of these medicines, talk to your doctor or pharmacist before combining them with vitamin D.