- April 27, 2018
- Dr Klaus Kraemer
- Most Recent, Perspectives
A recent report and podcast from the British Medical Journal1 has revived the controversy on when to scale back high-dose vitamin A supplementation (VAS) to reduce child mortality.2 We believe only compelling evidence can justify scaling back this intervention for the reasons pointed out below.
VAS is a Safe and Cost-effective Strategy for Reducing Child Mortality
Estimates from the World Health Organization (WHO) indicate that still 190 million children aged under five (U5) are vitamin A deficient (VAD).3 VAD increases the risk of disease and death from severe infections, and is the leading cause of preventable blindness in children U5. Well-controlled randomized trials in different regions with a high prevalence of VAD have conclusively demonstrated that high-dose vitamin A supplementation (VAS) given every four to six months to children U5 is an efficient and cost-effective strategy for reducing child mortality. VAS programs have largely contributed to the reduction in U5 mortality rates over the last decades.
A 2011 Cochrane review of 17 randomized controlled trials (RCTs) in 9 countries concluded that VAS results in a 24% reduction in U5 child mortality rate.4 This reduction in all-cause mortality confirmed an earlier meta-analysis from 1993, in which WHO’s5 original VAS guidelines were based upon and adopted by over 80 countries. There is also strong evidence, from both community and clinical trials, that VA prophylaxis and treatment can reduce the severity and fatality from measles, diarrhoea, and reduce the risk of hearing loss following middle ear infections. The most recent 2017 Cochrane reviewof 47 RCTs conducted in 19 countries concluded that vitamin A supplementation to U5 results in a significant 12% reduction for both all-cause mortality and mortality due to diarrhoea.6
DEVTA Contradicts the Evidence of VAS on Child Mortality
Cutting in half VAS effect on mortality between the 2011 and 2017 Cochrane reviews differ due to the latter’s inclusion of the DEVTA (Deworming and Enhanced Vitamin A) study,7 a large-scale program evaluation in the state of Uttar Pradesh, India. The ‘DEVTA study’ was an attempt to evaluate a large supplementation program providing VAS every six months through routine services. Approximately two million preschool children were reportedly enrolled through the Anganwadi Centres of the Integrated Child Development Scheme (ICDS) for the study and comparisons were made between usual care, 6-monthly VAS, 6-monthly albendazole (for deworming), or both. The intervention continued for 5 years and concluded that VAS did not reduce child mortality, although there was a 4% non-significant decline.
The DEVTA Study Had Limitations in the Program Design, Implementation, and Evaluation
Certain elements in the program design may shed light on the lack of effect in DEVTA. The research activity was vastly under-staffed with only 18 monitors overseeing the work of over 8,300 Anganwadi workers and the participation of two million children. This lack of human resources required the number of children and their levels of compliance to be determined only from a mid-study census using anon-random opportunistic sample of 2,106 children out of two million, therefore leaving the estimate open to bias. Thus, this number represented ‘compliers’ at the time of the census, rather than the numbers of children and capsules taken throughout the intervention. The authors reported an overall compliance of 86%, raising questions about accuracy of record keeping. Finally,the DEVTA evaluation also did not include younger children between 6-12 months, who normally account for one-third of the deaths in children U5 and would have benefited most from VAS.
The DEVTA Results Should Not Be Combined with Other Trials
Delivering VA to children every six months is a well-established intervention to reduce child mortality. The DEVTA study represented an earnest attempt to assess a large VAS program run by the Government of India with only a non-significant reduction in child mortality. The evaluation was expected to have revealed a greater program impact. The reasons for this ineffective program were due to how it was run and evaluated, problems that commonly afflict most large intervention strategies. There are many lessons to be learned from this undertaking, both with respect to VA delivery and program design, implementation, and resources needed for running and evaluating programs. We encourage the global scientific community to resist combining the DEVTA findings (as done in the 2017 Cochrane review) with those of previous, rigorously conducted trials to reset the overall “efficacy” of VA in reducing child mortality. Doing so could send the wrong message that mixing program evaluation results with those of well-controlled randomized human trials is an acceptable strategy, when it is not, especially when millions of children lives are potentially at stake.
Scaling Back of VAS Requires Compelling Evidence
As VAS saves lives, eyesight, and hearing of children only irrefutable evidence can justify scaling back this intervention. While continuing VAS programs, it is also critical to address the direct and underlying causes of VAD through effective interventions (e.g., dietary diversity, breastfeeding, fortification, hygiene, etc.). According to the Global Alliance for Vitamin A (GAVA),9it would only be justifiable to scale back VAS if VAD were no longer a public health issue (<5% biochemical VAD at population level). Only evidence from regular data collection in countries (at least every 10 years for VA intake and status data in children), or other data indicating a reduction in VAD, shall inform decisions to phase out of VAS.10
In settings where VAD is a public health problem (prevalence of night blindness is 1% or higher in children 24–59 months of age or where the prevalence of biochemical VAD (serum retinol 0.70 µmol/L or lower) is 20% or higher in infants and children 6–59 months of age), high-dose VAS (200,000 IU) is recommended for children 12–59 months of age (with a single half-dose for infants 6-11 months of age) every 4-6 months.5
1. Mason JB, Benn CS, Sachdev H, West KP Jr, Palmer AC, Sommer A. Should universal distribution of high dose vitamin A to children cease? BMJ. 2018 Mar 1;360:k927. doi: 10.1136/bmj.k927.
2. Mason JB, Sanders D, Greiner T, Shrimpton R, Yukich J. Vitamin A deficiency: policy implications of estimates of trends and mortality in children.Lancet Glob Health. 2016 Jan;4(1):e21. doi: 10.1016/S2214-109X(15)00246-6.
3. WHO. Global prevalence of vitamin A deficiency in populations at risk 1995–2005. WHO Global Database on Vitamin A Deficiency. Geneva, World Health Organization, 2009. (accessed 25 April 2018).
4. 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 Aug 25;343:d5094. doi: 10.1136/bmj.d5094.
5. WHO. Guideline: Vitamin A supplementation in infants and children 6–59 months of age. Geneva, World Health Organization, 2011. (accessed 25 April 2018).
6. Imdad A, Mayo-Wilson E, Herzer K, Bhutta ZA.Vitaminsupplementation for preventing morbidity and mortality in children from six months to five years of age. Cochrane Database Syst Rev. 2017 Mar 11;3:CD008524. doi: 10.1002/14651858.CD008524.pub3.
7. Awasthi S, Peto R, Read S, Clark S, Pande V, Bundy D; DEVTA (Deworming and Enhanced Vitamin A) team. Vitamin A supplementation every 6 months with retinol in 1 million pre-school children in north India: DEVTA, a cluster-randomised trial. Lancet2013 Apr 27;381(9876):1469-77. doi: 10.1016/S0140-6736(12)62125-4. Epub 2013 Mar 14.
8. Habicht JP, Victora C. Vitamin A supplementation in Indian children. Lancet. 2013 Aug 17;382(9892):592. doi: 10.1016/S0140-6736(13)61736-5.
9. The Global Alliance for Vitamin A (GAVA). Assessing vitamin A status in population based surveys. Draft statement, 2017.
10. Wirth JP, Petry N, Tanumihardjo SA, Rogers LM, McLean E, Greig A, Garrett GS, Klemm RD, Rohner F. Vitamin A Supplementation Programs and Country-Level Evidence of Vitamin A Deficiency.Nutrients. 2017 Feb 24;9(3). pii: E190. doi: 10.3390/nu9030190.