Evidence File 027 · Vitamin D · Method v0.14 of 4 claims researched

Vitamin DD3 / cholecalciferol · 4 claims

A textbook case of an effect that looked real, got bigger trials, and evaporated. Worth reading if you want to understand how evidence actually moves.

4Claims scored

7Sources

≤1000 IUStudied dose

Jul 26Reviewed

Claim 01

Prevents acute respiratory infections

3.1/10

Weak·64,086 participants · the effect disappeared

In 2021 a meta-analysis of 43 randomised trials found a statistically significant protective effect: odds ratio 0.92 (95% CI 0.86–0.99). That result drove a great deal of supplementation advice.

Six further trials then reported, including one with 15,804 participants. The updated analysis pooled 64,086 participants across 46 RCTs stratified by baseline vitamin D status and age. The confidence interval now includes 1.00 — no statistically significant protection.

Independent analyses agree. In older adults, 12 trials and 41,552 participants gave RR 0.99 (95% CI 0.97–1.02) at moderate certainty with zero heterogeneity. A dose-response analysis of 43 RCTs and 49,320 participants gave RR 0.99 (0.97–1.01). An earlier analysis found an apparent benefit for daily dosing that disappeared when restricted to high-quality trials.

The working

2Quantity

2Quality

2Consistency

9Directness

(2 × 0.25) + (2 × 0.35) + (2 × 0.25) + (9 × 0.15) = 3.1
Scored as support for the claim. The evidence base is enormous and high quality — and it no longer supports the claim, which is what the score has to reflect.

MetaJolliffe DA, Martineau AR, et al. Vitamin D supplementation to prevent acute respiratory infections: systematic review and meta-analysis of stratified aggregate data Lancet Diabetes Endocrinol · 46 RCTs, 64,086 participants · PROSPERO CRD42024527191
MetaJia H, Sheng F, Yan Y, Liu X, Zeng B. Vitamin D supplementation for prevention of acute respiratory infections in older adults PLoS One 2024 · 12 trials, 41,552 participants · RR 0.99, moderate certainty
MetaCho HE, Myung SK, Cho H. Efficacy of vitamin D supplements in prevention of acute respiratory infection Nutrients 2022;14(4):818 · benefit disappeared in high-quality subgroup

Claim 02

Reduces fracture risk in people with adequate levels

3.4/10

Weak·VITAL · null in a replete cohort

VITAL was run precisely because large randomised trials of daily supplemental vitamin D were lacking. It found no fracture risk reduction in a cohort whose baseline vitamin D levels were above 30 ng/mL.

Two features of that cohort matter for interpretation. The average participant started with a normal, if low-normal, level. And more than 40% were already taking vitamin D supplements at baseline. This is a trial about topping up people who are already sufficient — which is most supplement buyers — and it found nothing.

It says nothing about people who are genuinely deficient, which remains a different question with a different answer.

The working

2Quantity

3Quality

2Consistency

9Directness

(2 × 0.25) + (3 × 0.35) + (2 × 0.25) + (9 × 0.15) = 3.4
Scored as support for the claim in this specific population. A large dedicated trial found no benefit.

RCTLeBoff MS, et al., discussed in: Vitamin D supplementation: to D or not to D? Mayo Clin Proc 2024 · VITAL fracture outcomes in replete participants

Claim 03

Daily low doses work better than large intermittent doses

4.5/10

Emerging·Subgroup signal · survives in children, not overall

In a 2025 paediatric analysis of 17 RCTs and 18,372 participants, overall ARI incidence was not significantly reduced (RR 0.82, 95% CI 0.64–1.06). But the subgroup on daily low-dose regimens at ≤1000 IU/day showed RR 0.42 (95% CI 0.27–0.66) — and higher doses did not.

An earlier adult analysis found a similar pattern: daily supplementation RR 0.83 versus null overall. That signal did not survive restriction to high-quality trials, which is the reason to treat this as emerging rather than established.

The same paediatric analysis found supplementation did not improve outcomes once infection had already occurred.

The working

4Quantity

4Quality

4Consistency

7Directness

(4 × 0.25) + (4 × 0.35) + (4 × 0.25) + (7 × 0.15) = 4.5
A subgroup finding rather than a primary result, which is exactly the kind of finding that most often fails to replicate.

MetaThe role of vitamin D in prevention and treatment of acute respiratory infections in pediatric populations 17 RCTs, 18,372 participants · PMID 41387808
MetaCho HE, et al. Daily and short-term supplementation subgroups Nutrients 2022;14(4):818

Claim 04

Is safe at supplemental doses

6.6/10

Moderate·Fewer serious adverse events than placebo

Across 17 paediatric trials, supplementation was associated with fewer serious adverse events than control (RR 0.83, 95% CI 0.71–0.96). The reviewers concluded vitamin D appears safe.

This is the sensible way to hold the whole file. Vitamin D at ordinary supplemental doses is not doing harm. It is mostly not doing the things it is sold for either, in people who already have enough.

The working

6Quantity

6Quality

7Consistency

8Directness

(6 × 0.25) + (6 × 0.35) + (7 × 0.25) + (8 × 0.15) = 6.6
Unusually, the safety signal ran in the favourable direction rather than being merely absent.

MetaPaediatric ARI meta-analysis, serious adverse event outcome RR 0.83, 95% CI 0.71–0.96 · PMID 41387808

Dosage as studied

Most-studied regimen

≤1000 IU per day, where signal remains

The only subgroup with a surviving preventive signal in children used daily low-dose regimens. Higher doses did not show it.

Daily low dose: ≤1000 IU/day is the regimen with the surviving subgroup signal in paediatric ARI prevention.
Bolus dosing: High intermittent doses did not show preventive benefit in subgroup analyses.
Baseline matters: VITAL found no fracture benefit in participants whose baseline levels were already above 30 ng/mL — and over 40% were already supplementing.
Safety: Supplementation was associated with fewer serious adverse events, not more (RR 0.83, 95% CI 0.71–0.96, paediatric analysis).