Vitamin A supplementation involves giving Vitamin A to children at risk of deficiency to prevent death and other negative health impacts. We’d be interested in supporting a charity to carry out this program, but so far we have not found one we’d like to recommend.
The evidence on the effectiveness of the program raises a number of questions that we’d need a charity to answer, and we haven’t found one that can satisfactorily answer them. This post lays out the state of the evidence regarding Vitamin A, and the questions charities would need to answer to receive our recommendation.
This is a summary of our full report on Vitamin A Supplementation as a charitable program. As with our recent post about water quality interventions, we’re interested in providing more accessible summaries of our research to illustrate the challenges of identifying the most effective charities.
The key points of this post:
- Vitamin A supplementation has a mixed evidence base that seems to suggest that the program is particularly effective in certain circumstances.
- Because of the mixed evidence base, we have a set of questions any Vitamin A charity would need to answer before we would be willing to make them one of our top charities.
What is the problem?
Vitamin A deficiency (VAD) is a common, potentially deadly, condition in the developing world. Symptoms include:
- dry eyes (the leading cause of preventable childhood blindness)
- susceptibility to infection
Providing children with doses of Vitamin A two to three times per year can combat Vitamin A deficiency, and is typically relatively inexpensive. Doses cost less than a dollar per person per year, although distribution can be more costly.
Does Vitamin A supplementation work?
Studies of Vitamin A supplementation primarily focus on whether giving vitamin A pills to children can reduce their risk of dying.
Many large, randomized controlled trials (which we consider a particularly strong method of evaluating global health programs) have been conducted to determine the impact of Vitamin A supplementation, in which children are randomly chosen to receive the pills or not, and the mortality rate between recipients and non-recipients can be compared.
Most of the results look very promising: a Cochrane Collaboration review of seventeen randomized studies, mostly conducted in the 1980s and 1990s, found that Vitamin A supplementation reduces all-cause mortality by 24%.
However, one major study, with four times as many participants as all the studies included in the Cochrane review combined, contradicts these results.
Reconciling DEVTA and earlier trials
What should we think about whether Vitamin A is an effective program to support given that 17 trials found a large, significant effect of giving children Vitamin A supplements, while DEVTA found no statistically significant effect? There are a few possibilities:
- The world changed between the time the initial 17 studies were conducted and DEVTA, and Vitamin A supplementation is no longer as effective. Vitamin A supplementation may only be effective in areas with extremely high Vitamin A deficiency or child mortality – if so, worldwide improvements in health may mean Vitamin A supplementation is not as impactful as it once was, on average.
- The best guide to impact is averaging the effects of DEVTA and the other 17 trials. If we believe DEVTA’s results were due to chance, we could assume that the average effect of Vitamin A supplementation will be a 12% reduction in childhood mortality, based on the Cochrane meta-analysis that includes DEVTA. This is about half the size of the effect that was estimated prior to DEVTA.
- DEVTA’s lack of results were due to specific features of DEVTA, and charities conducting work under conditions more similar to the original 17 trials are likely to be highly impactful. Some differences between DEVTA and the original trials include (read more):
- DEVTA may have failed to reach enough children. DEVTA reported treating 86% of the children in the study, close to the rate achieved in previous trials, but some researchers have called this number into question, believing the study was not implemented as rigorously as previous trials and that it is implausible to achieve such high coverage at such low cost.
- DEVTA may have treated a population with less severe or less prevalent Vitamin A deficiency than in previous trials. However, DEVTA reports that a similar percentage of children had Vitamin A deficiency as in previous trials, although reliable, comparable data on Vitamin A deficiency is scarce.
- The population treated by DEVTA may have had better overall health than previously studied populations. Deaths prevented by Vitamin A Supplementation may be due to reduced mortality from diarrhea or measles, so if DEVTA participants were less vulnerable to dying from these diseases than participants in other studies, we would expect Vitamin A to have a smaller effect on mortality. The mortality rate in the control group in DEVTA was lower than control group mortality in 4 of the 5 trials that account for 80% of the weight in the Cochrane review, although the prevalence of diarrhea and measles does not appear to be very different.
We would guess that the best available explanation for the discrepancy between DEVTA and the 17 earlier trials is the lower baseline child mortality rate – and possibly better overall health – among DEVTA participants.
Finding a Vitamin A charity to recommend
For this reason, we would be interested in recommending a charity that could demonstrate that the conditions of the area it targets are similar to those of the original 17 studies considered in the Cochrane review, and so we have a list of questions that a charity would have to answer in order for us to consider them as a top charity.
Our list of questions includes:
- Is the charity working in an area with high child mortality?
- To what extent do children in the targeted area suffer from Vitamin A deficiency?
- Can the charity provide evidence that it successfully reaches the children it targets?
Our current conclusion
Considering all of these factors, we believe that Vitamin A supplementation may be one of the most cost-effective ways to save lives when the program is high quality and delivered in locations with high child mortality rates. Before recommending a charity, we’d need to see compelling evidence to satisfy these concerns about the conditions under which providing children with Vitamin A supplements is likely to be effective. Our concerns have not been satisfactorily addressed for any charity we’ve spoken with so far.
95% confidence interval: 17% to 31%
The study found a 4% reduction in child mortality, but with a 95% confidence interval between a 3% increase in child mortality and an 11% decrease in child mortality, so the study leaves open the possibility that giving children Vitamin A supplements has no effect on their risk of death.