Vitamin A supplements for preventing mortality, illness, and blindness in children aged under 5: systematic review and meta-analysis

Mayo-Wilson and colleagues have conducted a meta-analysis estimating
the overall mortality-reducing effect of vitamin A supplementation in
children between 6 and 59 months of age to be 24% (95% CI 17 to 31%). The
authors argue that further placebo-controlled trials of vitamin A
supplementation are not required and would be unethical (1). We disagree.

The meta-analysis does not differ substantially from previous meta-
analyses conducted in the early 1990s (2-4); a total of 8 later studies
have been added, but these trials only contribute 10% of the power. The
largest study conducted since the early 1990s is the DEVTA trial,
including 1 million Indian children. This trial found no beneficial effect
of vitamin A supplementation. For unknown reasons the trial has not been
published yet, but the preliminary result is known from a conference
presentation (5). The authors do not include DEVTA in the meta-analysis.
However, they include DEVTA in a sensitivity analysis and conclude that
though the result is significantly different from the results of the other
trials, the overall estimate of the vitamin A effect is still a
significant 12% (6 to 16%) reduction.

The authors note that there is considerable heterogeneity of effects
even without DEVTA, and test whether these could be due to differences in
effects over age groups, sex, location, dose and frequency of
supplementation. All these analyses are unfortunately based on very small
numbers (especially because the authors do not seem to have included the
Ghana VAST6 in their analysis by sex), and they are not conclusive. The
authors rest their case by explaining the heterogeneity with potential
underlying differences in vitamin A deficiency, the availability of other
nutrients or prevalence of disease, or simply faulty implementation of
some trials.

The authors make an analysis of secular trends without DEVTA and
conclude that there are no secular trends. However, if DEVTA is included
there is a significant secular trend. The data is compatible with the
notion that vitamin A supplementation was beneficial in the 1980s and
early 1990s, but has ceased to be beneficial.

We have put forward the hypothesis that the effect of vitamin A
supplementation may be modified by vaccination status (7). From that
perspective it is not surprising that the effect of vitamin A
supplementation may have changed during the last decades, in parallel with
the roll-out of the Expanded Programme on Immunization (EPI). That a
change in the effect of vitamin A over time could be linked to higher
vaccination intensity is supported by several other observations: First,
we reanalysed the Ghana VAST6 and found that the 19% (2 to 32%) reduction
in overall mortality was due to a significant reduction of 36% (12-53%)
among children without vaccination card, whereas there was no effect in
children with a vaccination card; this was due to a tendency for a
negative effect in girls who had a vaccination card (8). Second, when
vitamin A was provided with missing vaccines during a campaign in Guinea-
Bissau, vitamin A given with diphtheria-tetanus-pertussis (DTP) vaccine
was associated with significant negative effect compared with vitamin A
alone, and the effect was significantly different from the effect of
receiving vitamin A with measles vaccine (9). Third, during two recent
vitamin A campaigns in Guinea-Bissau, the effect of receiving vitamin A
compared to not participating was significantly different in children who
had DTP (negative) and measles vaccine (beneficial) as their most recent
vaccine before the campaign (manuscript). Fourth, due to the heterogeneity
in existing studies we have conducted a trial of vitamin A supplementation
with vaccines in children aged 6-23 months. As in the DEVTA trial, there
was no overall beneficial effect of vitamin A supplementation, and
furthermore the effect differed significantly in boys and girls
(manuscript).

Hence, there is accumulating evidence that the effect of vitamin A
supplementation after 6 months of age may be modified by vaccine status,
often in a sex-differential manner. This possibility is also supported by
trials of vitamin A supplementation in children less than 6 months of
age10-14. It is surprising that the authors do not discuss the
observations on vitamin A-vaccine interactions.

We think there is an urgent need to test the potential vaccine-and
sex-differential effects of vitamin A supplementation in other populations
to design the optimal vitamin A policy for all groups. From our
perspective it would be unethical not to conduct more studies and placebo-
controlled trials of vitamin A supplementation to children between 6 and
59 months of age, when there are indications that the existing policy may
cause harm in some groups. Vitamin A supplementation may be a cheap
intervention, but it is certainly a costly affair to deliver it biannually
to all children in 103 target countries if it confers no benefit and may
even harm subgroups of recipients. In a world that increasingly demands
evidence-based preventive policies it is embarrassing that our evidence
for such a widely-delivered intervention is more than 15 years old and
contradicted by all recent studies.

References

1. Mayo-Wilson E, Imdad A, Herzer K, Yakoob MY, Bhutta ZA. Vitamin A
supplements for preventing mortality, illness, and blindness in children
aged under 5: systematic review and meta-analysis. BMJ 2011;343:d5094.
doi: 10.1136/bmj.d5094

2. Beaton GH, Martorell R, Aronson KJ, et al. Effectiveness of Vitamin A
Supplementation in the Control of Young Child Morbidity and Mortality in
Developing Countries. Nutrition policy discussion paper No. 13. United
Nations Administrative Committee on Coordination/Subcommittee on Nutrition
1993.

3. Fawzi WW, Chalmers TC, Herrera MG, Mosteller F. Vitamin A
supplementation and child mortality. A meta-analysis. JAMA 1993;269:898-
903.

4. Glasziou PP, Mackerras DE. Vitamin A supplementation in infectious
diseases: a meta-analysis. BMJ 1993;306:366-70.

5. Awasthi S, Peto R, Read S, Bundy D. Six-monthly vitamin A from 1 to 6
years of age. DEVTA: cluster randomised trial in 1 million children in
North India. Presented at the Micronutrient Forum in Istanbul 1997.
Presentation available at www.ctsu.ox.ac.uk/projects/devta/istanbul-vit-a-
lecture.ppt.

6. Ghana VAST Study Team. Vitamin A supplementation in northern Ghana:
effects on clinic attendances, hospital admissions, and child mortality.
Lancet 1993; 342(8862):7-12.

7. Benn CS, Bale C, Sommerfelt H, Friis H, Aaby P. Vitamin A
supplementation and childhood mortality: Amplification of the non-specific
effects of vaccines? Int J Epidemiol 2003;32:822-8.

8. Benn CS, Aaby P, Nielsen J, Binka FN, Ross DA. Does vitamin A
supplementation interact with routine vaccinations? An analysis of the
Ghana Vitamin A Supplementation Trial. Am J Clin Nutr 2009; 90: 629-39.

9. Benn CS, Martins C, Rodrigues A, Fisker AB, Christoffersen D, Aaby P.
The effect of vitamin A supplementation administered with missing vaccines
during national immunisation days in Guinea-Bissau Int J Epidemiol
2009;38:304-11.

10. Benn CS, Diness BR, Roth A, et al. Randomised trial of the effect on
mortality of 50,000 IU vitamin A given with BCG vaccine to infants in
Guinea-Bissau, West-Africa BMJ 2008;336:1416-20.

11. Benn CS, Rodrigues A, Yazdanbakhsh M, et al. The effect of high-dose
vitamin A supplementation administered with BCG vaccine at birth may be
modified by subsequent DTP vaccination. Vaccine 2009;27:2891-98.

12. Benn CS, Fisker AB, Napirna BM, et al. Vitamin A supplementation and
BCG vaccination at birth in low birthweight neonates: two by two factorial
randomised controlled trial. BMJ 2010;340:c1101. doi: 10.1136/bmj.c1101.

13. Kirkwood B, Humphrey J, Moulton L, Martines J. Neonatal vitamin A
supplementation and infant survival. Lancet 2010;376(9753):1643-4.

14. Benn CS, Fisker A, Aaby P. Heterogeneous effects on child survival in
neonatal vitamin A supplementation trials. Lancet 2011;377:1314-5.