Vitamin A: lifesaving where children lack it, useless-to-risky where they don't

Vitamin A is two different stories depending on who is taking it: in children who are deficient, supplements are one of the most solidly proven survival interventions in global health, cutting all-cause deaths by about 12% [s1]. In people who already get enough — most of the rich world — adding more vitamin A confers no such benefit and, in the wrong form and dose, has raised the risk of harm in a large trial [s3]. Treating those two situations as one question is how a genuine public-health triumph gets turned into a pointless, occasionally dangerous, supplement pitch.

Where it saves lives

Vitamin A deficiency remains a major problem in low- and middle-income countries, affecting an estimated 190 million children under five and contributing to blindness, infection and death [s1]. On the strength of the evidence, the World Health Organization recommends vitamin A supplementation for children aged 6 to 59 months in settings where deficiency is a public-health problem [s2].

The evidence behind that recommendation is unusually strong. A Cochrane review pooled 47 studies in roughly 1,223,856 children across 19 countries, most in Asia and Africa [s1]. In the meta-analysis of all-cause mortality — 19 trials covering 1,202,382 children — supplementation reduced the risk of death by 12% (risk ratio 0.88, 95% confidence interval 0.83 to 0.93), which the reviewers rated high-certainty evidence [s1]. Deaths from diarrhoea fell by a similar 12% (risk ratio 0.88, 95% confidence interval 0.79 to 0.98) [s1]. Supplementation also cut the incidence of diarrhoea (risk ratio 0.85), measles (risk ratio 0.45, 95% confidence interval 0.30 to 0.69), night blindness (risk ratio 0.32) and Bitot's spots, the eye lesions of deficiency (risk ratio 0.42) [s1]. For a cheap, twice-yearly capsule, that is a remarkable return, and it is why vitamin A campaigns sit near the top of the child-survival toolkit.

The picture is not uniformly positive, which is itself a mark of honest evidence. The same review found no clear effect on deaths from measles (risk ratio 0.88, 95% confidence interval 0.69 to 1.11) or respiratory disease, and a raised risk of vomiting within 48 hours of a dose (risk ratio 1.97, 95% confidence interval 1.44 to 2.69) [s1]. The benefit is real and large where deficiency is common; it is not a cure-all even there.

Where it does nothing — or harm

The case collapses when the same logic is exported to well-nourished populations, where vitamin A intake is already adequate and there is no deficiency to correct. The starkest warning comes from the Beta-Carotene and Retinol Efficacy Trial (CARET), which gave smokers, former smokers and asbestos-exposed workers a daily combination of 30 mg of beta carotene and 25,000 IU of retinol — vitamin A as retinyl palmitate [s3]. Rather than preventing cancer, the supplement group had a relative risk of lung cancer of 1.28 and a relative risk of death from lung cancer of 1.46 (95% confidence interval 1.07 to 2.00); the trial was stopped early, and the authors concluded the combination had no benefit and might have caused harm [s3]. The lesson generalises: a nutrient that rescues the deficient can injure the replete. This is the same reversal documented for the beta-carotene half of that trial, covered in beta-carotene and vitamin A supplements and cancer.

Preformed vitamin A also accumulates, unlike water-soluble vitamins, so chronic high intake carries its own toxicity risks — another reason the blanket "more is better" framing fails outside deficiency.

How to read this

The dividing line is deficiency, not the molecule. Where children lack vitamin A, supplementation is among the best-evidenced interventions in medicine, with high-certainty trials showing fewer deaths and less blindness [s1][s2]. Where intake is already sufficient, the survival benefit disappears and the risk ledger can turn negative, as CARET showed in people who were not short of it [s3]. That is the opposite of how vitamin A is often sold in wealthy countries — as a general immune or eye "booster" — and it fits the wider pattern that single-nutrient supplements rarely help people who are not deficient, seen again with selenium and cancer. The public-health value is enormous, and precisely targeted; the consumer-supplement version is neither. The measles findings also connect to why outbreaks still kill where deficiency and low vaccination overlap, as in the 2026 Bangladesh measles outbreak.

This article is informational and not medical or dietary advice; decisions about supplementation should be made with a qualified clinician.