Letter on inadequate coverage of SAM in Lancet Undernutrition Series, by Susan Shepherd

SAM inadequately addressed in the Lancet Undernutrition Series

Dear Editor,

In 2003, The Lancet captured and focused attention on saving children’s lives with the publication of a five-part series on child survival. In 36 succinct pages, this series described the "where" and "why" children are dying and highlighted the contributing role of undernutrition to approximately 50% of all deaths in children under 5 years of age¹. With the recent Maternal and Child Undernutrition series, The Lancet continues its commitment to shine the spotlight on nutrition as "a desperately neglected aspect of maternal, newborn, and child health." The toll exacted by childhood undernutrition is felt every day by Médecins Sans Frontières (MSF) field workers in more than 60 countries around the world. Development of nutrition programmes is an increasingly important focus of our work. MSF has gone from treating 100,000 malnourished children during the period from 2000-2005, to reaching 150,000 such children in 2006 and again in 2007.

MSF welcomes this series and commends the efforts to emphasise the vital importance of nutrition for vulnerable populations, particularly children under 2 years of age. The series also highlights the glaring under-allocation of resources to this area. However, we think the series misses the mark in some important areas; underestimation of disease burden of severe acute malnutrition and failure to acknowledge the tremendous potential of new treatment strategies.

Underestimation of the disease burden of severe acute malnutrition and the associated mortality In paper 1 of the series, the burden of disease is estimated, by prevalence, to be 19 million (3.5% of 555,729,000) children suffering from severe acute malnutrition (SAM), and 178 million (32% of same population) stunted children². As SAM is a transitory condition, with known seasonal variations in many parts of the world, a crosssectional survey ‘snapshot’ at any one point in time will certainly lead to an underestimation of the caseload. In addition, no mention is made in paper 1 (or any of the other papers in the series) of kwashiorkor, a common form of SAM in central and eastern Africa, further underestimating the burden of SAM.

According to the series, SAM is identified as a significant cause of death, responsible for 449,160 deaths per year among children in developing countries. This estimate is substantially lower than the 1 million SAMattributable deaths reported in the UN Joint Statement on Community-based Management of Severe Acute Malnutrition³ and again raises questions about the validity of calculating mortality from prevalence surveys.

The risk of death associated with SAM is reported as 9.4 times higher than for a nonwasted child, yet the global case fatality rate calculated using the authors’ data is only 2% - this with only a minute fraction of the 19 million children receiving treatment for SAM. In 2007, MSF’s nutrition project in Maradi, Niger, treated 22,250 children for SAM using WHO standards for admission criteria. This programme reported an 83% recovery rate, while case fatality and defaulter rates were 2.7% and 8.8%, respectively (G. Harczi, personal communication). Thus, even with effective and proven treatments, case-fatality rates are not as low as suggested in paper 1.

Despite the increased risk of death associated with SAM, however, the undernutrition series focuses its attention on stunting, and interventional strategies targeting the 36 countries that account for 90% of this condition. Countries such as Turkey are included in this list, while Chad, Somalia, Central African Republic, Sierra Leone, Liberia, and Haiti are not. This type of analysis prioritises the needs of the less undernourished over those most immediately at risk of death.

Failure to acknowledge the major impact of Community-based Treatment of SAM

Treatment of SAM is not a ‘key message’ in paper 3, despite its huge potential to save lives. Relying solely on data from randomised controlled trials (RCTs), the authors limit their recommendation to treat SAM in facility-based programmes instead of endorsing communitybased management. This fails to accurately reflect current reality of practice and does not acknowledge community-based care strategies using ready-to-use therapeutic foods (RUF) as an enormous advance in the ability to treat this lethal condition and its potential to reach vast numbers of children.

In paper 3, the authors give several reasons for not widely recommending communitybased care. Firstly, their intent is to discuss intervention effects in national populations, not in "special circumstances of crisis." This differentiation ignores the fact that, although initially developed and implemented for more effective response in emergency relief⁴, outpatient management of SAM is gaining favour with national governments. Ethiopia, Malawi, Sri Lanka, and Niger are institutionalising community-based SAM management protocols, with many more countries following their lead. The vast majority of children affected by SAM do not live in areas of conflict or "special circumstances," but rather live in families that are fundamentally food insecure.

Secondly, the authors of paper 3 cite the lack of robust, randomised studies as a reason for not endorsing community-based care. In web table 5, the authors present their search criteria, which led to the selection of RCTs, including those with historical controls. Their search methods returned 19 articles addressing the use of RUF in community management of acute malnutrition, of which only 5 were considered suitable for analysis. Of the 14 studies excluded, 5 were observational studies, one of which enrolled 2,131 severely malnourished children⁵. However, large-scale observational studies can provide equivalent or better evidence, especially when randomised studies are small and inadequately control for confounding variables⁶.

Thirdly, the authors of paper 3 focus on reduction of case fatality rate as the basis for making their recommendations. Thus, facilitybased care according to WHO protocols is recommended for implementation (Table 1 of paper 3) because it is deemed effective at reducing the case fatality rate for children with SAM, compared with those not treated according to this protocol. Yet none of the five published studies cited and used for this pooled analysis (Ahmed, Ashworth, Deen, Falbo, Wilkinson) for supporting improved facility-based care used randomisation. These studies used historical controls from the same facility or non-random controls from case series. Furthermore, no mention is made of the coverage or defaulter rates associated with facility-based treatment. In MSF’s experience in Niger, exclusive facility-based treatment has been associated with substantially higher default rates compared with combined or strictly outpatient care: 28% vs 16.8% and 5.6%, respectively⁷. Thus, actual mortality from stand-alone facility-based programmes is likely to be underestimated.

Compared with resource-intensive, facilitybased inpatient management, communitybased care is a practical, large-scale intervention, as evidenced by ~60,000 children treated in Niger in 2006⁸, and >26,000 treated in Malawi, Ethiopia, and Sudan in 2001-2006⁹. Coverage of >70% with community-based treatment, versus <10% in facility-based programmes, suggests community care has a much greater impact at the population level¹⁰, with improved outcomes¹¹. Furthermore, it is important to remember that community-based treatment includes WHO protocol facilitybased care; it is an effective strategy because it offers inpatient care for complicated SAM while achieving high patient coverage through an outpatient network, and minimising defaulter rates by decreasing opportunity costs for mothers.

While the authors of the undernutrition series are to be commended for their rigor in establishing a solid evidence base for making recommendations, the fundamental question remains: what rules of evidence should be applied? Of the 10 original papers cited as references for the UN Joint Statement on Community-based Management of Severe Acute Malnutrition, four were expressly excluded from consideration in the Systematic Review of Management of Childhood Severe Malnutrition for the undernutrition series. Among these, there is a controlled trial with systematic allocation, a clinical trial, a retrospective study and a review (paper 3, webtable 5). It may be time to re-examine the rules of what constitutes valid evidence.

The three studies from the UN Joint Statement that are included in the paper 3 SAM review are RCTs. Although RCTs have become the gold standard for clinical decisionmaking, the same standard is not sufficient for public health decision-making where the pathways from intervention to impact are multiple and complex¹² Evidence-based public health decisions must therefore rely on a variety of data types, not just from RCTs, ranging from highly controlled efficacy trials, to observational studies with control or comparator groups, to the reporting of results obtained from large-scale programmes in differing contexts.

Lastly, many of the interventions recommended for the 36 countries in Table 1 of paper 3, e.g. hand washing and behaviour change communication, are not supported by studies showing an effect on nutritional status or mortality. Different standards thus seem to be used to recommend some interventions and not others.

Moving forward

The Lancet Undernutrition series has taken important steps forward in acknowledging and beginning to define the problems with nutrition for women and children under 2. But some of the recommended interventions- exclusive breastfeeding, complementary feeding practices and hand washing - rely on behaviour change, presuming that mothers can easily change their busy, overburdened lives. Furthermore, Vitamin A and zinc supplements address only part of the nutritional deficits.

As the Lancet series points out, funding is woefully inadequate. If the nutrition community is serious about addressing the crisis of maternal and child undernutrition, it must not be limited by inexpensive solutions that intervene on the margins. Funding must increase and programming developed to ensure that energy-dense, nutrient rich foods or supplements get into the mouths of young children. Scaling up community-based treatment of SAM is a major step in the right direction, but it is not sufficient. A child should not have to deteriorate to the point of severe wasting to "qualify" for nutritious food.

Susan Shepherd, MD
Nutrition Advisor
Access Campaign for Essential Medicines

¹Black, RE et al. Where and why are 10 million children dying every year ? Lancet 2003; 361: 2226-2234.

²Black, RE et al. Maternal and Child Undernutrition : Global and regional exposures and health consequences. Lancet 2008

³WHO, WFP, UNSCN, UNICEF. Community-based management of severe acute malnutrition. Geneva: World Health Organisation, 2007. http://www.who.int/nutrition/topics/statement_commbased_malnutrition/en/index.html (accessed on May 12, 2008).

⁴Collins S, Sadler K. Outpatient care for severely malnourished children in emergency relief programmes: a retrospective cohort study. Lancet; 2002. 360: 1824-30.

⁵Linneman Z, Matilsky D, Ndekha M, A large-scale operational study of home-based therapy with ready-to-use therapeutic food in childhood malnutrition in Malawi. Matern Child Nutr. 2007 Jul;3(3):206-215.

⁶Shrier I, Boivin J-F, Steele RJ, et al. Should meta-analyses of interventions include observational studies in addition to randomized controlled trials? A critical examination of underlying principles. Am J Epidemiol 2007; 166: 1203-9.

⁷Gaboulaud V, Dan-Bouzoua N, Brasher C, Fedida G, Gergonne B, Brown V. Could nutritional rehabilitation at home complement or replace centre-based therapeutic feeding programmes for severe malnutrition? J Trop Pediatr. 2007; 53: 49-51.

⁸Defourny I, Seroux G, Abdelkader I, Harczi G. Management of moderate acute malnutrition with RUTF in Niger. Field Exchange 2007; 31: 2-4.

⁹Collins S, Sadler K, Dent N, et al. Key issues in the success of community-based management of severe malnutrition. Food Nutr Bull 2006; 27 (suppl 3): S49-82.

¹⁰Collins S, Dent N, Binns P, Bahwere P, Sadler K, Hallam A. Management of severe acute malnutrition in children. Lancet 2006; 368: 1992-2000.

¹¹Ciliberto MA, Sandige H, Ndekha MJ, et al. Comparison of home-based therapy with ready-to-use therapeutic food with standard therapy in the treatment of malnourished Malawian children: a controlled, clinical effectiveness trial. Am J Clin Nutr 2005; 81: 864-70.

¹²Victora CG, Habicht, JP and Bryce J. Evidence-Based Public Health: Moving beyond randomized trials. Am J Public Health; March 2004; 94, 3. 400-405.

¹³Duffield A et al (2004). Review of the Published Literature for the Impact and Cost-effectiveness of Six Nutrition Related Emergency Interventions. Report prepared by the ENN; December.