Evidence was derived from an updated Campbell systematic review of the effects of neonatal nutrition interventions on neonatal mortality and child health and development outcomes (204), and from an individual participant data (IPD) meta-analysis on the effects of early neonatal vitamin A supplementation on infant mortality (205). The IPD meta-analysis was used to further inform the evidence on the effects and safety of neonatal vitamin A supplementation, and to investigate the potential sources of heterogeneity. The Campbell review identified 16 trials (169 366 infants) that assessed the effect of neonatal vitamin A supplementation, of which 13 individually randomized or cluster RCTs (168 788 infants) contributed data to this evidence summary. All trials were conducted in LMICs. All 13 trials provided oral vitamin A as either a single dose of 50 000 IU, two doses of 24 000 IU (within 24 hours of one another), a single dose of 25 000 IU, or a single dose of 25 000 IU and 50 000 IU, compared with placebo.
The IPD meta-analysis included 11 trials assessing the effect of early neonatal vitamin A supplementation (25 000–50 000 IU intended to be given within 2–3 days after birth) compared with placebo, with infant follow-up through at least 6 months of age.
Comparison: Vitamin A supplementation compared with placebo or no vitamin A supplementation (IPD meta-analysis)16
Neonatal/infant mortality: Newborn vitamin A supplementation was associated with a variable effect from an 11% decrease to a 6% increase in the risk of death up to 6 months of age (11 trials; RR 0.97, 95% CI 0.89 to 1.06) and a variable effect from a 7% decrease to an 8% increase of death up to 12 months of age (10 trials; RR 1.00, 95% CI 0.93 to 1.08) when compared with placebo.
Univariate meta-regression and meta-analyses by study level characteristics
Meta-regression identified five variables as significantly associated with the effect of newborn vitamin A supplementation on mortality at six months, as presented below. Meta-analyses by study-level characteristics were subsequently conducted for these variables.
Geographic region: Until 6 months of age, newborn vitamin A supplementation was associated with a 13% lower risk of death in Asian countries (5 trials; RR 0.87, 95% CI 0.77 to 0.98), but in African countries the risk of death includes a variable effect from a 2% reduction to a 15% increase (6 trials; RR 1.06, 95% CI 0.98 to 1.15). Until 12 months of age, newborn vitamin A supplementation was associated with a variable effect from a 20% reduction to a 3% increase in the risk of death in Asian countries (4 trials; RR 0.91, 95% CI 0.80 to 1.03), and a variable effect from no effect to a 15% increase in African countries (6 trials; RR 1.07, 95% CI 1.00 to 1.15).
Maternal vitamin A deficiency: Until 6 months of age, and in study populations where the prevalence of maternal vitamin A deficiency was moderate or severe, newborn vitamin A supplementation was associated with a 13% lower risk of death (3 trials; RR 0.87, 95% CI 0.80 to 0.94). In study populations where there was no or mild maternal vitamin A deficiency, the risk of death includes a variable effect, from a 4% lower to a 15% higher risk of death (7 trials; RR 1.05, 95% CI 0.96 to 1.15). Until 12 months of age, and in study populations where the prevalence of maternal vitamin A deficiency was moderate or severe, newborn vitamin A supplementation was associated with a variable effect, from a 17% lower risk to no effect on the risk of death (2 trials; RR 0.91, 95% CI 0.83 to 1.00). In study populations where there was no or mild maternal vitamin A deficiency, the risk of death includes a variable effect, from a 2% lower risk to a 15% higher risk (7 trials; RR 1.06, 95% CI 0.98 to 1.15).
Early infant mortality: Newborn vitamin A supplementation was associated with a 9% lower risk of death up to 6 months of age in study populations where 6-month mortality in the control group was ≥ 30 per 1000 live births (6 trials; RR 0.91, 95% CI 0.85 to 0.98), but includes a variable effect from a 5% lower risk to a 24% higher risk of death in populations where 6-month mortality in the control group was under 30/1000 (5 trials; RR 1.08, 95% CI 0.95 to 1.24).
Ratio of 6-month to 12-month mortality in the control group: Newborn vitamin A supplementation was associated with a variable effect from a 16% lower to a 1% higher risk of death up to 6 months of age in study populations where 75% or more of infant mortality occurred in the first 6 months (6 trials; RR 0.92, 95% CI 0.84 to 1.01), but was associated with a variable effect from no effect to a 22% higher risk of mortality in populations where less than 75% of infant mortality occurred up to 6 months of age (4 trials; RR 1.11, 95% CI 1.00 to 1.22).
Maternal education: Newborn vitamin A supplementation was associated with a 12% lower risk of death up to 6 months of age in populations where ³ 32% of mothers had never been to school (4 trials; RR 0.88, 95% CI 0.80 to 0.96), but was associated with a variable effect from an 8% lower to an 18% higher risk of death in study populations where < 32% of mothers had never been to school (5 trials; RR 1.04, 95% CI 0.92 to 1.18).
Subgroup analysis by neonatal risk status (high risk or low risk)
Vitamin A given to < 2500 g newborns: Vitamin A supplementation in newborns whose birthweight was < 2500 g was associated with a variable effect from an 11% decrease to a 5% increase in the risk of death up to 6 months of age (8 trials; RR 0.97, 95% CI 0.89 to 1.05) and a variable effect from an 8% decrease to an 8% increase in the risk of death up to 12 months of age (7 trials; RR 0.99, 95% CI 0.92 to 1.08) when compared with placebo.
Vitamin A given to ≥ 2500 g newborns: Vitamin A supplementation in newborns whose birthweight was ≥ 2500 g was associated with a variable effect from an 11% decrease to a 4% increase in the risk of death up to 6 months of age (8 trials; RR 0.96, 95% CI 0.89 to 1.04) and a variable effect from a 7% decrease to a 7% increase in the risk of death up to 12 months of age (7 trials; RR 1.00, 95% CI 0.93 to 1.07) when compared with placebo.
Subgroup analysis by maternal risk of vitamin A deficiency
Vitamin A given to newborns of mothers with night blindness during pregnancy: Vitamin A supplementation in newborns whose mothers had night blindness during pregnancy was associated with a variable effect from a 35% decrease to a 33% increase in the risk of death up to 6 months of age (3 trials; RR 0.93, 95% CI 0.65 to 1.33) and a variable effect from a 36% decrease to a 37% increase of death up to 12 months of age (2 trials; RR 0.94, 95% CI 0.64 to 1.37) when compared with placebo.
Vitamin A given to newborns of mothers without night blindness during pregnancy: Vitamin A supplementation in newborns whose mothers did not had night blinding during pregnancy was associated with a 14% decrease in the risk of death up to 6 months of age (3 trials; RR 0.86, 95% CI 0.77 to 0.96) but a variable effect from a 23% decrease to no effect in the risk of death up to 12 months of age (2 trials; RR 0.88, 95% CI 0.78 to 1.00) when compared with placebo.
Subgroup analysis by initiation of breastfeeding
Vitamin A given to newborns initiating breastfeeding during the first hour after birth: Vitamin A supplementation in newborns who initiated breastfeeding during the first hour after birth was associated with a variable effect from a 6% decrease to a 14% increase in the risk of death up to 6 months of age (6 trials; RR 1.03, 95% CI 0.94 to 1.14) and a variable effect from a 5% decrease to a 12% increase in the risk of death up to 12 months of age (6 trials; RR 1.03, 95% CI 0.95 to 1.12), when compared with placebo.
Vitamin A given to newborns initiating breastfeeding 2–23 hours after birth: Vitamin A supplementation in newborns who initiated breastfeeding 2–23 hours after birth was associated with a variable effect from a 13% reduction to a 3% increase in the risk of death up to 6 months of age (6 trials; RR 0.94, 95% CI 0.87 to 1.03) and a variable effect from a 10% decrease to a 5% increase in the risk of death up to 12 months of age (6 trials; RR 0.97, 95% CI 0.90 to 1.05), when compared with placebo.
Vitamin A given to newborns initiating breastfeeding ≥ 24 hours after birth: Newborn vitamin A supplementation in newborns who initiated breastfeeding ≥ 24 hours after birth was associated with a variable effect from a 26% decrease to a 13% increase in the risk of death up to 6 months of age (6 trials; RR 0.92, 95% CI 0.74 to 1.13) and a variable effect from a 26% decrease to an 8% increase in the risk of death up to 12 months of age (6 trials; RR 0.90, 95% CI 0.74 to 1.08), when compared with placebo.
Subgroup analysis by newborn sex
Vitamin A given to male newborns: Vitamin A supplementation in male newborns was associated with a variable effect from an 8% decrease to an 8% increase in the risk of death up to 6 months of age (11 trials; RR 0.99, 95% CI 0.92 to 1.08) and a variable effect from a 7% decrease to a 7% increase in the risk of death up to 12 months of age (10 trials; RR 1.00, 95% CI 0.93 to 1.07), when compared with placebo.
Vitamin A given to female newborns: Newborn vitamin A supplementation in female newborns was associated with a variable effect from a 14% decrease to a 1% increase in the risk of death up to 6 months of age (11 trials; RR 0.93, 95% CI 0.86 to 1.01) and a variable effect from an 8% decrease to a 6% increase in the risk of death up to 12 months of age (10 trials; RR 1.01, 95% CI 0.92 to 1.06), when compared with placebo.
Additional considerations
From the NeoVita trials in Ghana, India and the United Republic of Tanzania (208–210), neonatal vitamin A supplementation has been accepted for research purposes by families and health workers. However, in the Bangladesh mechanistic study, it was reported that 11% (72/636) of pregnant women eligible for the study refused participation, not wishing to give their newborns supplements (211).
In a feasibility study conducted in Bangladesh and Nepal (212), mothers viewed neonatal vitamin A supplementation favourably with few reports of negative experiences. The negative experiences were largely self-limiting and not clearly attributable to the intervention. Health workers viewed the intervention favourably.
The potential for benefit in some contexts and the potential for harm in other contexts may lead to differences in acceptability among various population groups. If restricted to specific settings in which the intervention has shown significant benefit, the intervention would likely be acceptable.
Feasibility
A qualitative evidence synthesis of women’s experiences of postnatal care found no direct evidence relating to women’s views on the feasibility of using Vitamin A supplementation for their newborns (28). Indirect evidence from the same review indicates there may be feasibility challenges in some LMICs among women who don’t understand and/or value neonatal supplementation strategies (moderate confidence in the evidence). In addition, women and families in low resource settings may be less motivated to engage with health workers if they think health facilities are under-resourced or if they believe treatment will incur additional costs (moderate confidence in the evidence).
A qualitative evidence synthesis of health workers’ experiences of postnatal care found no direct evidence relating to views on vitamin supplementation for newborns (29). However, indirect evidence suggests lack of personnel, resources and training may limit provision of information and counselling on vitamin supplementation for newborns (moderate confidence in the evidence).