The editors

Short communication: WN October commentary

Broader vision is needed

The debate about vitamin A supplementation initiated by Michael Latham in this journal (1) needs to be seen with a broad vision. The issue is not just one of a nutrient, its needs and availability, and its specific meaning for child survival. To treat people suffering from vitamin A deficiency with the nutrient is clearly essential. This therapeutic approach is beyond debate, if only for ethical reasons.

But prevention of vitamin A deficiency is another matter. This is more than an extension of the therapeutic approach to children at risk. Supplementation as one means for preventing micronutrient deficiencies has its inherent challenges and problems, and limited potential. In emergency situations it may be the only alternative, as long as vitamin A-fortified foods are not available. In an urban slum, fortification is much more appropriate and effective in reaching the population at risk.

The principal weakness of any supplementation strategy for the prevention of vitamin A deficiency, is that in most cases inadequate diets are not short of or deficient in one nutrient only (2). Much child mortality is not caused just by vitamin A deficiency. It follows that various dietary approaches to nutritional problems are needed.

The potential and benefits from the distribution of vitamin A capsules have been demonstrated in many studies over a long time. But all these success stories and achievements cannot be related only to the reduction of vitamin A deficiency by supplementation. For examples, more extensive immunisation and improved basic primary health care provision have also contributed.

Also, are resources used for supplementation programmes optimally used, and could more have been achieved? It’s hard to say. In most cases no well designed baseline surveys, nor post-intervention surveys, have been done. If they had been, they might have helped the proponents of supplementation programmes a lot. But this was not part of the plan.

Supplements do not reach children in greatest need

One inherent weakness of supplement provision is that it is generally restricted to areas easy to reach by plane, ship, train and vehicle. Children outside such areas generally cannot benefit. The costs of distribution in remote areas need to be taken into account. Effective medicines have no impact on disease in patients having no access to them (3). Also, studying vitamin A deficiency is more readily done in areas that are easy to reach. How far have investigators gone to investigate or even to identify those in greatest need ?

The colleagues who have designed and implemented the supplementation programmes have done vital work. Originally, with the challenge of the observed and assumed single nutrient deficiency, supplementation was probably an approach without any real alternative. Now though, considerations of driving forces, and the development of the market for artificial nutrient providers, are of real concern. How free are organisations like the Global Alliance for Improved Nutrition (GAIN) and the Micronutrient Initiative, from the companies that manufacture synthetic vitamin A?

Supplementation suppresses food-based approaches

As my predecessor in my post at the university Claus Leitzmann has pointed out: ‘Constipation does not indicate a lack of laxatives’. Vitamin A deficiency needs to be prevented by dietary means. In this way, complex deficiencies can be addressed, local resources can be developed and used, and local food systems can be strengthened.

Because many children with the greatest dietary needs have multiple nutrient and also energy deficiencies, there is no fully rational alternative to dietary approaches, while accepting fortification as an intermediate stage, but always aiming at balanced wholesome food systems and supplies and thus diets. This ‘developmental approach’ to the prevention of vitamin A deficiency is surely more appropriate than any narrowly focused approach. Supplementation programmes have weakened the ‘Alma Ata’ primary health care philosophy, and do not contribute to the development of stable food systems.

Yes, supplementation is effective in reducing vitamin A deficiency. But why are the Millennium Development Goals far from being reached in most African countries? Do these countries not receive enough supplements? Or do they need research into and development of functioning local and regional food systems? And do we need to pay more attention to unexpected effects of the provision of antioxidants on a population basis? (4).

Conclusion

The nutrition policy-making and scientific community would be wise not to focus on single nutrient approaches, but instead on people’s dietary needs (5). The main problem of serious research on the potential of diverse diets still is scientific neglect and severe underfunding. Who protects the funding needs of food-based policies and programmes? Most funding agencies want quick results that seem to show the success of narrowly focussed interventions. Rational nutrition policies are still to be developed. These need to be interdisciplinary and used to identify the need for sustainable preventive nutrition interventions; which is to say, for the nutritional, societal and other development of those in greatest need (6).

Since 2008, WHO has refocused on primary health care. The nutrition community is well advised to adopt the developmental approach to food systems, nutrition and diets, and to leave single nutrient focussed interventions to nutrition therapists. First steps in this direction are already being made (7).

References

1. Latham M. The great vitamin A fiasco. World Nutrition 2010; 1, 1:12-45.
2. Keding G. Krawinkel M. Food Diversity from Plough to Plate - Linking Agrobiodiversity, Dietary Diversity and Micronutrient Supply. Sight and Life 2008; 3: 23–2
3. Akhter N, Witten C, Stallkamp G, Anderson V, de Pee S, Haselow N. Children aged 12–59 months missed through the national vitamin A capsule distribution program in Bangladesh: findings of the Nutritional Surveillance Project. Field Actions Science Reports 2008; 1:33-41.
4. Ristow M, Zarse K, Oberbach A, Klöting N, Birringer M, Kiehntopf M, Stumvoll M, Kahn CR, Blüher M. Antioxidants prevent health-promoting effects of physical exercise in humans. Proceedings of the Natlional Academy of Sciences (USA) 2009; 106, 21: 8665-8670.
5. Ogle BM, Hung PH, Tuyet HT. Significance of wild vegetables in micronutrient intakes of women in Vietnam: an analysis of food variety. Asia Pacific Journal of Clinical Nutrition. 2001; 10, 1:21-30.
6. Beauman C, Cannon G, Elmadfa I, Glasauer P, Hoffmann I, Keller M, Krawinkel M, Lang T, Leitzmann C, Lötsch B, Margetts BM, McMichael AJ, Meyer-Abich K, Oltersdorf U, Pettoello-Mantovani M, Sabaté J, Shetty P, Sória M, Spiekermann U, Tudge C, Vorster HH, Wahlqvist M, Zerilli-Marimò M.The principles, definition and dimensions of the new nutrition science. Public Health Nutrition 2005;8(6A):695-698.
7. Wahlqvist ML, Keatinge JD, Butler CD, Friel S, McKay J, Easdown W, Kuo KN, Huang CJ, Pan WH, Yang RY, Lee MS, Chang HY, Chiu YW, Jaron D, Krawinkel M, Barlow S, Walsh G, Chiang TL, Huang PC, Li D; FIHS Roundtable Participants. A Food in Health Security (FIHS) platform in the Asia-Pacific Region: the way forward. Asia Pacific Journal of Clinical Nutrition. 2009; 18(4):688-702.

Please cite as: Krawinkel M. Broader vision is needed. [Short communication] World Nutrition, November 2010, 1, 6: 272-275. Obtainable at www.wphna.org

Short communication: WN October commentary

The case for universal

supplementation is not well made

I will begin by expressing my pleasure with some of the points made by Professor West et al, in their October WN commentary (1), and in particular their call for movement on the important and neglected food-based approaches to undernutrition. Evidently they, like me, believe these deserve equal attention to the other approach they advocate, (originally called an ‘emergency’ approach in the early 1970s when WHO missions began advocating to South Asian governments) of universal semi-annual supplementation of young children with large-dose vitamin A capsules.

I also agree that vitamin A deficiency is still a serious public health problem in many areas, including in much of India. I believe Professor Latham’s point, in his May WN commentary (2), was rather that, from the trend data available from a few countries, the prevalence of serious clinical deficiency tended to decline substantially before national vitamin A capsule programmes with high coverage rates could plausibly ‘take credit’ for such a change.

Prevalence rates have continued declining since the late 1990s, when vitamin A capsule distribution programmes reached the kinds of coverage levels required before any such impact could be expected. But the rate of improvement appears to have been more rapid before then in Bangladesh, one of the few countries where enough national surveys have been conducted to allow an estimate of such a difference.

Exclusive focus on universal vitamin A supplementation has
harmed nutrition programme and policy development

West et al mention only low vitamin A capsule coverage rates as a likely explanation for the world not having done better in reducing vitamin A deficiency. I suggest that a larger contribution is made by the fact that such programmes only cover children during part of the year, and often fail to reach the worst affected populations. The children who least need supplementation, are likely to be the ones who first receive it. For example, in India, the 20 per cent who recently received it had less malnutrition and came from families with lower mortality rates in young children. This is a possible cause of spurious attribution of lower mortality to receipt of the capsules until coverage rates achieve very high levels (3). Another likely reason is that donors have listened only to proponents of the capsule programme as the sole approach. The Micronutrient Initiative, for example, has been required by a major donor not to spend more than 15 per cent of its budget on food-based approaches. And of course supplementation does not reach a majority of the population.

Like Michael Latham, I am convinced that the widespread implementation of vitamin A capsule programmes has also had the effect of reducing government and donor interest and motivation in implementing basic approaches to improve diets. Executives and other professionals working for governments and donors have told me as much. In the case of vitamin A fortification, I have been told by policy makers in several countries that it cannot even be considered, as this would make things even more precarious for the young children routinely being given huge doses. Others have told me they have heard the same thing. No, there are no clinical trials to back up these statements. But in my experience, valuable as clinical trials may often be for policy-making, limiting oneself to peer-reviewed published trials is likely to make policy-making appear to be a mystical activity. Of the 21 factors listed as causes of nutrition policy change in lower-income countries by a World Bank/UNICEF review (4), only two related to such evidence.

As stated (2), at the 1993 International Vitamin A Consultative Group meeting, when it was clear that the findings from the mortality trials was likely to raise interest among donors, I called for an integrated response—as West et al now seem to do. I suggested donors say ‘yes’ to requests for support to the short-term approach, while asking countries at the same time to present proposals for an equal amount of funds for longer-term, more sustainable and self-reliant approaches and for a simple system of monitoring that would allow them to know when universal capsule distribution was no longer needed. Given that semi-annual capsule distribution would improve serum retinol for only a few months a year, I agree with West et al that serum monitoring during the other months would be better than the dietary approach I suggested, though the use of retinol-binding protein would be just about as effective, and much cheaper, than serum retinol (5).

However, no balanced response emerged. Michael Latham and I, as well as many of the others who have commented on his initial article in WN, find this highly questionable. We believe governments and funding agencies should be held accountable for this serious policy failure..

Measles may be confounding the trial results

One major point Michael Latham makes has been expressed similarly in an paper co-authored by a colleague of West et al (6): ‘It was reasonable to expect, given the large reductions in mortality, that vitamin A would modify the incidence and severity of the principal causes of child mortality—that is, lower respiratory tract infections, diarrhoea, and malaria.’

But something doesn’t add up, since the majority of studies that have looked for such effects have failed to find them. Latham went on to hypothesise that measles may hold the key to understanding this anomaly. There is no doubt about the impact of vitamin A on measles, and measles commonly kills children in low income settings. At the time when the original trials were done, measles vaccination coverage was likely to have been low in the areas studied, while it might have been higher years later when the huge DEVTA trial, which showed no mortality impact of vitamin A supplementation, was done.

West et al say that, contrary to what Latham said, measles was not identified as the cause of most of the mortality prevented in the trials they refer to. But how children died in those studies was not determined by a physician at the child's bedside, nor by a real autopsy. Cause of death was rather simply assumed, based on asking a series of questions, weeks or months later, to a parent. While useful for some purposes, such a ‘verbal autopsy’ is completely unable to discriminate deaths caused by measles from deaths caused by a range of other causes, because measles can and often does cause similar symptoms. Certainly it often causes diarrhoea and typically a few days of fever before the rash appears. These may kill a seriously vitamin A deficient child. Such a death may be reasonably attributed to diarrhoea/dehydration, even though investigators cannot know how often measles was the actual cause of the diarrhoea. By analogy, ‘verbal autopsies’ often attribute to malaria, deaths occurring while a child is suffering from fever, which certainly cannot be considered to be more than a reasonable guess.

West et al agree that vitamin A has no impact on morbidity in general, but quote a few findings that suggest it may attenuate diseases or help the body cope better with severe disease. Perhaps this is so important that vitamin A supplementation would reduce mortality even in well-vaccinated populations. Let’s hope such research will be forthcoming. But for the moment we cannot rule out the possibility that achieving high coverage of measles vaccination will largely remove the impact of vitamin A supplementation on young child mortality.

Further adding to the plausibility of Latham’s hypothesis is the fact that measles tends to occur in local epidemics or in waves, with very few cases showing up in between. If indeed the well-documented link between measles and vitamin A was responsible for much if not most of the mortality that vitamin A supplementation prevented in most of the trials, then that would explain why two of the published trials found no impact. West et al cite data from a total of 165,000 children, but the DEVTA trial failed to find any impact among a million children in a low-income state in India – where West et al make a specific point that vitamin A deficiency is still a problem. What surely is most likely, is that measles happened to be relatively quiescent in those populations during the study periods. Alternatively, measles vaccination coverage may have been greater, or vaccinations done more effectively, than in the other study areas.

If diarrhoea not associated with measles had been responsible for much of the seemingly substantial mortality-protective effect of vitamin A supplementation, then why was this effect not operating in two of the study sites? One might expect some variation, but if vitamin A really eliminated 20-30 per cent or even more of mortality, wouldn't those studies, and the huge DEVTA trial, have shown at least say a 10-15 per cent impact? Plenty (though far from all) well-conducted studies have shown no impact of vitamin A on diarrhoea morbidity; whereas few if any well-conducted studies have failed to find a link between measles morbidity and vitamin A.

Vitamin A supplementation may often assist in fighting infection, among children deficient in vitamin A. But so do many other nutrients, many of which are commonly provided in food-based interventions but obviously not in vitamin A capsules. And if vitamin A alone has such a huge impact on mortality, via non-measles mechanisms, virtually all studies should also show that it also has an impact on morbidity as well, as it does for measles.

The soft underbelly of vitamin A supplementation: morbidity

West et al do not dispute that vitamin A supplementation has failed to be shown to have a beneficial impact on the other major causes of morbidity and mortality in low-income countries. They do say that ‘vitamin A prophylaxis and treatment can reduce the severity and fatality from measles and diarrhoea, among other less-well defined infections. No peer reviewed, published data has emerged in recent years to contradict these findings.’ Research on the issue is of course hardly a hot topic any longer, except for neonatal supplementation which, regarding diarrhoea, is certainly failing to live up to expectations (7.8).

Nevertheless, contrary to what West et al say, some trials with older infants and young children have in recent years failed to find any impact of vitamin A on diarrhoea (9). As cited by Latham, a recent meta-analysis of 8 trials found no effect. Half suggested a positive impact and half were negative (10). The impact is variable even in children exposed to HIV (11). And some recent studies suggest that other nutrients like zinc may have a much more powerful impact on diarrhoea than vitamin A (12). Again, food-based approaches could certainly provide zinc. Vitamin A capsules do not.

Contrary to what the West et al commentary would lead readers to believe, the capsule programme is associated with more than minor temporary side effects. It often appears to do harm, particularly in children who are not deficient (13). Of the 8 studies in the recent meta-analysis (10), 6 found adverse effects. None found statistically significant positive effects. The meta-analysis resulted in an overall significant finding of adverse respiratory outcomes.

Why no proper evaluations?

West et al emphasise that universal vitamin A capsule programmes cost $US 1-2 per child per year. But the food-based programme supported by the Swedish International Development Cooperation Agency and others in Bangladesh, in which I was involved, achieved substantial increases in household production and consumption of green leafy vegetables by children under 5 on a huge scale, at a cost of $US 0.13 per child per year, while providing at the same time numerous other benefits (14).

It’s odd, to say the least, that after spending hundreds of millions of dollars over a 15-year period, donors who normally call for impact evaluations have apparently not funded any in this case – at least not any that have been published.

West et al cite what they describe as ‘program evaluations’ that provide evidence of impact. The first is a comparison of national surveys in Nepal, which might indicate that vitamin A coverage at 85 per cent was effective in reducing mortality rates, but of course many other relevant changes have occurred there. Such findings in a majority of the dozens of countries involved would be needed to make a strong case using such data. Next they cite a study from 1984 finding that vitamin A supplementation reduces xerophthalmia. This refers neither to a programme nor to mortality. Next is a US CDC study estimating overall impact and cost based on ‘assumed mortality reduction.’ Next is a study done in one slum of India, comparing data with another randomly selected slum that did not receive supplements. This quasi-experiment involved monthly visits to homes, which hopefully resulted in sick children receiving attention, making impact somewhat distant from that likely to be seen in real life supplementation programmes. There were no significant differences in morbidity but a decrease in mortality. The next examined the case fatality effect of vitamin A supplementation on hospitalised children, again hardly relevant. The final citation is of an estimate of the burden of disease attributable to vitamin A deficiency in South Africa. Again, it assumed mortality impact, it did not measure it.

Thus, in contrast to what West et al would have the reader believe, none of these were actually programme evaluations. In contrast to what they correctly say about the status of publication of the original clinical trials, most of these studies were published in low-impact national journals.

As West et al point out, such evaluations are extremely challenging to do. But large scale interventions with comparable costs, like oral rehydration therapy, breastfeeding promotion, and folic acid fortification, have been accompanied by evaluations that have done a much better job. Any objective reviewer of this situation must be left wondering whether they – or the programme donors – actually want any real impartial impact evaluation to be done.

Food-based programmes are the way forward

Food-based programmes designed to reduce under-nutrition are often criticised for doing a patchy job. So what? Leafy green vegetables, and most of the other foods promoted in food-based programmes, are extremely nutrient-dense foods. They provide a wide range of nutrients, more even than Sprinkles, and many of the other currently popular multi-micronutrient supplementation approaches.

Helping poor people eat more nutritious food does not require justification by proofs that this totally solves deficiencies of one nutrient for one group in the population. Justification for spending hundreds of millions of dollars in ways that distract from food-based programmes, certainly does require strong proof. Without such justification, such programmes should be phased out. Where the vitamin A capsule programme is demonstrably acting as a barrier and impeding fortification or other food-based programmes, then the phasing out process may need to be rapid.

Editor’s note. Professor Greiner has been a colleague of Professor Latham for many years, and supported him as he prepared his May WN commentary.

References

1. 1 West KP Jr, Klemm RDW, Sommer A. Vitamin A saves lives. Sound
   science, sound policy. [Commentary]. World Nutrition 2010; 1, 5: 211-229. Obtainable at: www.wphna.org
2. Latham M. The great vitamin A fiasco. [Commentary] World Nutrition May 2010, 1, 1: 12-45. Obtainable at www.wphna.org.
3. Semba RD, de Pee S, Sun K, Campbell AA, Bloem MW, Raju VK. Low intake of vitamin A-rich foods among children, aged 12-35 months, in India: association with malnutrition, anemia, and missed child survival interventions. Nutrition 2010; 26 (10):958-962.
4. Gillespie S, McLachlan M, Shrimpton R, editors. Combatting Malnutrition. Washington DC: World Bank and UNICEF, 2003.
5. Gorstein JL, Dary O, Pongtorn, Shell-Duncan B, Quick T, Wasanwisut E. Feasibility of using retinol-binding protein from capillary blood specimens to estimate serum retinol concentrations and the prevalence of vitamin A deficiency in low-resource settings. Public Health Nutrition. 2008; 11(5): 513-520.
6. Kosek M, Oberhelman RA. Unraveling the contradictions of vitamin A and infectious diseases in children. Journal of Infectious Diseases. 2007; 196: 965-967.
7. Diness BR, Christoffersen D, Pedersen UB, Rodrigues A, Fischer TK, A A, et al. The effect of high-dose vitamin A supplementation given with bacille Calmette-Guérin vaccine at birth on infant rotavirus infection and diarrhea: a randomized prospective study from Guinea-Bissau. Journal of Infectious Diseases. 2010 Sep 1; 202(Suppl): S243-51.
8. Gogia S, H.S. S. Neonatal vitamin A supplementation for prevention of mortality and morbidity in infancy: systematic review of randomised controlled trials. British Medical Journal. 2009 March 27; 338: b919.
9. Chen K, Zhang X, Li TY, Chen L, Wei XP, P Q, et al. Effect of vitamin A, vitamin A plus iron and multiple micronutrient-fortified seasoning powder on infectious morbidity of preschool children. Nutrition. 2010; Epub ahead of print.
10. Grotto I, Mimouni M, Gdalevich M, MIMOUNI D. Vitamin A supplementation and childhood morbidity from diarrhea and respiratory infections: A meta-analysis Journal of Pediatrics. 2003; 142(3): 297-304.
11. Humphreys EH, Smith NA, Azman H, McLeod D, Rutherford GW. Prevention of diarrhoea in children with HIV infection or exposure to maternal HIV infection. Cochrane Database Systematic Reviews. 2010 June 16;6:CD008563.
12. Chhagan MK, Van den Broeck J, Luabeya KK, Mpontshane N, Tucker KL, Bennish ML. Effect of micronutrient supplementation on diarrhoeal disease among stunted children in rural South Africa. European Journal of Clinical Nutrition 2009; 63(7): 850-857.
13. Chen H, Zhuo Q, Yuan W, Wang J, Wu T. Vitamin A for preventing acute lower respiratory tract infections in children up to seven years of age. Cochrane Database Systematic Reviews 2008; 23(1): CD006090.
14. Greiner T, Mitra SN. Evaluation of the impact of a food-based approach to solving vitamin A deficiency in Bangladesh. Food and Nutrition Bulletin. 1995;16(3):193-205.

Please cite as: Greiner T. The case for universal supplementation is not well made. [Short communication] World Nutrition, November 2010, 1, 6: 275-282. Obtainable at www.wphna.org

Letter: WN October commentary

No good evidence that supplementation

actually saves many lives

Sir: West et al. (1) have again claimed, based on studies conducted in the 1980s, that vitamin A supplementation reduces under-5 mortality rate in children in the range of 23-34 per cent in undernourished settings. They also state that ‘No peer reviewed, published data have emerged in recent years to contradict this finding’. However, the most recent and a robust trial conducted on 1 million children in deprived settings in India (the DEVTA trial), documented no impact of vitamin A supplementation on under-5 mortality (2). It remains a mystery as to why even after completion of the trial in 2006, it has yet not been published or apparently even submitted for publication. Probably forces other than science are at work.

The trends in routinely collected under-5 mortality data from lower-income countries (3) with more than a 90 per cent mega-dose vitamin A supplementation coverage also do not support a child survival effect of vitamin A. The table shown here illustrates under-5 mortality rates in 24 countries for three time points (1960, 1990 and 2002).

There is no evidence of a roughly 30 per cent decline rate after vitamin A supplementation was introduced. What these figures do show, is a consistent and sometimes impressive decline in the period 1960-1990. This was before introduction of the mega-dose vitamin A supplementation programme. In the period 1990-2002, after the vitamin A capsule programme was instituted, there were practically no changes in mortality in 9 countries, a reduction in 11 countries, and in 4 countries an increase in mortality. Such results can have a number of causes. They do not support the claims made by advocates of the vitamin A capsule programme.

____________________________________________________________

Trend of under 5 mortality among children in countries with 90% or more coverage with vitamin A supplementation in 2001



t indicates countries that have achieved a second round of vitamin A supplementation to children within any one year.
Source: UNICEF. The State of the World’s Children 2004. Obtainable at http://www.unicef.org/sowc04/  Accessed 5 October 2010.

____________________________________________________________

Countries with limited financial resources and competing health priorities cannot afford the luxury of initiating interventions to raise serum biochemistry alone. Unambiguous demonstration of prevention of health consequences below a serum retinol cut-off is imperative to consider vitamin A supplementation, based on these considerations. Extrapolations based on simple cross-sectional correlations have no value in this context.

In addition, the magnitude of vitamin A deficiency is grossly inflated by the serum retinol cut-offs proposed by West et al. (1). The authors provide no convincing evidence of benefit of supplementation below these cut-offs. Further, in the Indian context, these levels have not been corrected for simultaneous C-reactive protein levels.

Umesh Kapil
All-India Institute of Medical Sciences
New Delhi 110029, India
Email: umeshkapil@yahoo.com 

HPS Sachdev
Sitaram Bhartia Institute of Science and Research, New Delhi 110016, India
Email: hpssachdev@gmail.com

References

1. West KP Jr, Klemm RDW, Sommer A. Vitamin A saves lives. Sound
   science, sound policy. [Commentary]. World Nutrition 2010; 1, 5: 211-229. Obtainable at: www.wphna.org
2. 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. www.ctsu.ox.ac.uk/projects/devta/istanbul-vit-a­lecture.ppt
3. The State of the World's Children 2004. http://www.unicef.org/sowc04/ . Accessed 5 October 2010.

Please cite as: Kapil U, Sachdev HPS. No evidence that supplementation actually saves many lives. [Letter] World Nutrition, November 2010, 1, 6: 282-285. Obtainable at www.wphna.org.