Effect of early neonatal vitamin A supplementation on mortality during infancy in Ghana (Neovita): a randomised, double-blind, placebo-controlled trial

Randomized, double-blind, placebo-controlled trial in newborn infants in rural Brong Ahafo, Ghana. Infants were identified at birth or within 2 days of birth and followed for at least 6 months; 11,474 infants were randomized to the vitamin A arm, within a total trial population of 22,955 newborns.

Single oral dose of 50,000 IU vitamin A as retinol palmitate was given on the day of birth or within the next 2 days. The capsules were taken by newborns orally; placebo capsules matched the formulation and contained soybean oil with minute vitamin E.

Neonatal vitamin A did not reduce mortality through 6 months. In the vitamin A group, 278 deaths occurred versus 248 in placebo, with RR 1.12 (95% CI 0.95-1.33; p=0.183); the Kaplan-Meier hazard ratio was also 1.12 (95% CI 0.95-1.33; p=0.181). Early deaths within 3 days were numerically higher with vitamin A (70 vs 52; RR 1.35, 95% CI 0.94-1.93), and bulging fontanelle also trended higher (32 vs 21; RR 1.53, 95% CI 0.88-2.64). Fever, vomiting, diarrhoea, inability to suck or feed, and convulsions within 3 days were not meaningfully different between groups, while hospital admission to 6 months was higher in the vitamin A arm (793 vs 718; p=0.044).

Although the trial was large and well blinded, it was conducted in a single rural Ghana region with low HIV prevalence, which limits generalizability. Some secondary outcomes had modest event differences and imprecise estimates, and the hospital-admission finding should be interpreted cautiously because the reported effect estimate was unusual and likely sensitive to analysis details.