Description of children identified as suffering from MAM in Bangladesh: Varying results based on case definitions

Full text: Background: There is a wide discrepancy between the proportion of children classified as acutely malnourished when MUAC criteria are used compared to weight for height. This has greatly complicated setting targets for the coverage of SAM and MAM programs in Bangladesh. This difference is much larger for children identified with MAM than for those with SAM, largely because identification as MAM can overlap both with SAM and with children not identified as acutely malnourished. Objective: To review existing data sets in order to determine the relationship between MUAC and other anthropometric measures, helping to provide a better understanding of the implications of different admission criteria to therapeutic and supplementary feeding programs. Methodology: This study uses data collected through national nutritional surveillance projects over multiple seasons in Bangladesh. For the years 1990 to 2006, sub-samples of data from the Nutritional Surveillance Project were pulled from areas of the country that remained constant over a set period. Data from 2010 to 2012 was pulled from the Food Security and Nutrition Surveillance Project. Case definition: Cases of moderate acute malnutrition were identified using MUAC- for-age z-scores (-3>z-score>-2), MUAC cut-offs (115mm>MUAC>125mm), and weight-for-height z-scores (-3>z-score>-2). Results: In all years more than 50% of all children identified as moderately malnourished were classified as such by only one measure (1990 selected sub-districts: 52%, 2012 national sample: 69%) In 1990 a higher proportion of children were categorized as moderately malnourished based on MUAC-for-age z-scores than by weight for height z-scores, but since 2000 the opposite has been true. This change is closely tied to the increasing height of children sampled, due to the declining rates of stunting in the country. After controlling for age and weight-for-height z-scores, an increase in height of one cm was associated with an increase in MUAC slightly less than 1 mm. For children of less than two years of age, this relationship was ever larger. Discussion: Due to the large differences in which children are identified as moderately malnourished depending on the measure used, treatment protocols can be expected to differ greatly. Further research into these differences may allow programs to run at a reduced cost which still serving those most in need. Funding: Support for this review was provided though the Millennium Development Goals Fund in Bangladesh and facilitated by WFP. The Food Security and Nutrition Surveillance Project is funded by the European Union. The Nutritional Surveillance Project was funded by multiple partners during its 17 years of operation, notably including the United States Agency for International Development and The Embassy of the Kingdom of the Netherlands/Bangladesh. (author)