A summary of published research, not medical advice. See the Disclaimer.
Studies: every indexed trial and review on this topic, scored
Vitamin D is a fat-soluble vitamin that the skin makes in sunlight and a few foods supply. The body converts it into a hormone that regulates calcium absorption.
Best-supported effect: Fewer acute respiratory infections in children and adults. Probably helps a little (moderate evidence).
Most-cited weak claim: Fewer fractures in adults not selected for deficiency. Makes little or no difference (strong evidence).
How we grade evidence · Last reviewed: October 2026
What you need to know
Vitamin D slightly reduced acute respiratory infections in children and adults, with moderate evidence, when given daily or weekly and not as large intermittent bolus doses. In adults, vitamin D alone made little or no difference to fractures, falls, cancer incidence, or heart attacks and strokes, with strong evidence. Trials disagree on fractures with added calcium and on depression. High blood calcium was more common on vitamin D (moderate evidence), with about 4 extra cases per 1,000 people at 3,200 to 4,000 IU a day, and a single yearly dose of 500,000 IU was followed by more falls and fractures. With 1,000 mg of calcium a day, kidney stones were more common over 7 years. The large long-term trials used vitamin D3 at 2,000 IU a day for a median of 5.3 years.
| Effect | Verdict | What was measured | Studied in |
|---|---|---|---|
| Fewer acute respiratory infections | Probably helps a little (moderate evidence) Pooled analyses of 25 and 46 trials, small effect, inconsistent across trials |
Measured | Children and adults, most dosed daily or weekly |
An individual-participant meta-analysis of 25 RCTs (data from 10,933 participants) found 12% lower odds of at least one acute respiratory infection with vitamin D (OR 0.88) [1]. The effect appeared with daily or weekly dosing but not with large bolus doses, and among people dosed daily or weekly it was larger in those whose starting blood level was below 25 nmol/L [1]. A later aggregate meta-analysis of 46 RCTs (75,541 participants) found a smaller reduction (OR 0.92), seen mainly with daily doses [2]. Probably helps a little (moderate evidence): downgraded for inconsistency between trials and a small absolute effect.
| Effect | Verdict | What was measured | Studied in |
|---|---|---|---|
| Fewer fractures (vitamin D alone) | Makes little or no difference (strong evidence) Pooled analysis of 81 trials and the VITAL trial found no fracture change |
Hard outcome | Adults, including midlife and older adults not selected for deficiency or low bone mass |
| Fewer falls | Makes little or no difference (strong evidence) Review that pooled 81 trials found no change in falls |
Hard outcome | Adults |
| Lower cancer incidence | Makes little or no difference (strong evidence) VITAL trial and a pooled analysis of 10 trials found no change in new cases |
Hard outcome | Adults, including men 50 and over and women 55 and over |
| Fewer heart attacks and strokes | Makes little or no difference (strong evidence) VITAL trial in 25,871 adults found no reduction in major heart and blood vessel events |
Hard outcome | Men 50 and over, women 55 and over |
More on these results under Claimed but unsupported or contested.
| Claim | Verdict | What was measured | Studied in |
|---|---|---|---|
| Lower cancer mortality | May help a little (weak evidence) Few trials reported cancer deaths, VITAL trial result could have been chance |
Hard outcome | Adults in trials with 3–10 years of follow-up |
| Fewer fractures (vitamin D with calcium) | Unclear: trials disagree (weak evidence) One pooled analysis found fewer fractures, another in adults living at home found none |
Hard outcome | Adults 50 and over |
| Less depression | Trials disagree (moderate evidence) Pooled analysis of 31 trials, lower scores mainly in shorter trials |
Self-reported | Adults, mostly in trials of 24 weeks or less |
A meta-analysis of 10 RCTs found no change in cancer incidence, but deaths from cancer were about 13% lower (RR 0.87) in the five trials that reported them, mostly with daily rather than bolus dosing [3]. In VITAL, 25,871 adults received 2,000 IU/day or placebo for a median of 5.3 years, and cancer deaths were lower without reaching significance (HR 0.83, 95% CI 0.67–1.02) [4]. May help a little (weak evidence): few trials reported cancer deaths, and VITAL's confidence interval included no effect.
A meta-analysis of 81 RCTs (53,537 participants) found no effect of vitamin D on total fractures (RR 1.00), hip fractures (RR 1.11) or falls (RR 0.97) [5]. In VITAL's fracture analysis (25,871 adults, 5.3 years), total fractures did not change (HR 0.98) at any starting blood level [6]. Very high yearly or daily doses raised falls in some trials; that harm is listed under Side effects [19][22]. The claim comes mainly from observational studies that link low blood levels with fractures.
Results for vitamin D with calcium differ by population. A meta-analysis of 29 RCTs (63,897 participants) of calcium alone or with vitamin D in people 50 and over found 12% fewer fractures in the 17 trials that reported them, and 24% fewer in trials with high compliance [7]. A meta-analysis of 33 RCTs (51,145 participants) in community-dwelling adults over 50 found no reduction in hip or total fractures with the combination [8]. Unclear: trials disagree (weak evidence): the positive analysis mixed calcium alone with the combination, and the analysis of the combination in people living at home found no clear effect.
VITAL found no reduction in invasive cancer (HR 0.96) or major cardiovascular events (HR 0.97) at 2,000 IU/day [4]. The 10-trial meta-analysis also found no change in cancer incidence (RR 0.98) [3].
The depression claim comes largely from observational studies that link low blood levels with depressive symptoms. A dose-response meta-analysis of 31 RCTs (24,189 participants) reported lower depressive symptom scores per 1,000 IU/day, but mainly in trials of 24 weeks or less and not in longer trials [9]. Trials disagree (moderate evidence): the effect faded in longer trials.
Few large trials have enrolled only people with very low starting levels, so the effect of correcting a true deficiency on fractures, cancer and heart disease is largely untested.
Shown in humans
Shown only in animals or cells
Theory only
| Side effect | Verdict | How often vs placebo | Source | Studied in |
|---|---|---|---|---|
| Hypercalcemia (high blood calcium) | Probably happens (moderate evidence) Pooled analyses show raised risk, one 2,000 IU/day trial showed no increase |
RR 1.54 in 37 trials, not related to dose [18]; RR 2.21 at 3,200–4,000 IU/day, about 4 extra cases per 1,000 people [19]; RR 1.93 at 2,800 IU/day or more [20]; no increase at 2,000 IU/day over 5.3 years in one 25,871-person trial [4] | [4][18][19][20] | 25,871 adults |
| Other adverse events at 2,000 IU/day for 5.3 years | No harm found (strong evidence) One 25,871-person trial showed no increase over a median of 5.3 years, meaning half were followed longer |
No increase vs placebo | [4] | 25,871 adults |
| Hypercalciuria (high urine calcium) | May happen (weak evidence) More cases across 14 trials, two other sources saw no rise |
RR 1.64 in 14 trials [18]; no difference at 3,200–4,000 IU/day [19]; 30.6% of women on vitamin D with calcium, equally common on placebo [23] | [18][19][23] | Women |
| Kidney stones (vitamin D alone) | May not increase it (weak evidence) Estimates in 9 and 5 trials point in opposite directions |
RR 0.66 in 9 trials [18]; RR 1.26, not significant, in 5 trials [20]; no difference at 3,200–4,000 IU/day [19] | [18][19][20] | |
| Kidney stones (400 IU/day with 1,000 mg calcium) | Probably happens (moderate evidence) 36,282 women on added calcium, more stones over 7 years |
HR 1.17 over 7 years in 36,282 women | [21] | 36,282 women |
| More falls with very high intermittent or daily doses | Probably happens (moderate evidence) Excess falls at very high doses, but 37 trials saw no change |
83.4 vs 72.7 falls per 100 person-years after 500,000 IU once a year (RR 1.15) [22]; RR 1.25 at 3,200–4,000 IU/day (4 trials) [19]; no change in falls across 37 trials, with similar results in high-dose trials and above 800 IU/day [5] | [5][19][22] | Older women |
| Fractures after a single yearly dose of 500,000 IU | Probably happens (moderate evidence) One trial in 2,256 older women who did not know what they got, more fractures on vitamin D |
4.9 vs 3.9 per 100 person-years | [22] | 2,256 older women |
| Hospital admission at 3,200–4,000 IU/day | May happen (weak evidence) 1.16 times the risk across 7 trials, review flagged incomplete safety reporting |
RR 1.16 (7 trials) | [19] | |
| Total adverse events at 2,800 IU/day or more for a year or longer | May not increase it (weak evidence) 1.05 times the risk across 10 trials of a year or longer |
RR 1.05 in 10 trials | [20] | |
| Vitamin D toxicity with severe hypercalcemia | Unknown (very weak evidence) Reports of single patients only, toxic doses not tested in trials |
Case reports only | [10][24] | People with dysregulated vitamin D metabolism; 58-year-old woman with normal starting vitamin D |
The hypercalcemia meta-analyses agree that risk rises, by about 1.5 to 2.2 times; one found the rise did not depend on dose [18], and VITAL found no increase at 2,000 IU/day [4]. The absolute rise was small, about 4 extra cases per 1,000 people at 3,200–4,000 IU/day, and that review rated safety reporting at high risk of being incomplete [19]. In the falls trial, the extra falls were concentrated in the first 3 months after each yearly dose (RR 1.31, then 1.13) [22]. Toxicity mostly follows excessive long-term intake or accidental use of very high-dose preparations, or a disorder of vitamin D metabolism [10]. Trials do not test toxic doses, so the risk at those doses is unknown.
This lists what studies and labels report. A pharmacist or prescriber can say whether it applies to a specific person.
This lists what studies and labels report. A pharmacist or prescriber can say whether it applies to a specific person.
| Combined with | What happened | Verdict | Source | Studied in |
|---|---|---|---|---|
| Calcium (1,000 mg/day, with 400 IU/day vitamin D) | More kidney stones than placebo over 7 years (HR 1.17) | Probably happens (moderate evidence) One trial of 36,282 women, more stones than placebo over 7 years |
[21] | 36,282 women |
| Thiazide diuretics (with calcium and vitamin D) | Hypercalcemia reported in older people and those with poor kidney function or overactive parathyroid glands; evidence mostly non-randomised or case reports | Unknown (very weak evidence) Mostly studies without random assignment or reports of single patients, in higher-risk people |
[25] | Older people |
| Atorvastatin | Vitamin D lowered atorvastatin levels and atorvastatin raised vitamin D levels, in low-quality studies | Unknown (very weak evidence) Each changed the other's blood levels, seen only in low-quality studies |
[25] | |
| Orlistat (a fat-absorption blocker) | Vitamin D levels fell after 1 month in 17 adolescents despite a daily multivitamin with 400 IU; a systematic review judged the evidence insufficient | Unknown (very weak evidence) Pilot study in 17 adolescents who knew what they got, review judged evidence insufficient |
[25][26] | 17 adolescents |
| Antiepileptic drugs, antimicrobials, HIV antiretrovirals, H2 blockers | Evidence insufficient to say whether they change vitamin D levels | Unknown (very weak evidence) Formal review could not tell whether levels change |
[25] |
These are the amounts used in the studies. They are not recommendations.
The large long-term trials used vitamin D3 at 2,000 IU/day for a median of 5.3 years [4][6]. In the respiratory trials, the benefit appeared with daily or weekly dosing and not with large intermittent bolus doses [1]. In the 46-trial analysis it was clearest at daily dose equivalents of 400–1,000 IU for up to 12 months [2]. In the calcium trials, effects were larger with at least 1,200 mg/day of calcium and, in combined regimens, at least 800 IU/day of vitamin D [7].