Relationship between mid upper arm circumference and weight changes in children aged 6–59 months

BACKGROUND: The objectives of this study were to (i) describe the relationship between weight changes and MUAC changes in children aged between 6 and 59 months during treatment for SAM in CMAM programmes in three country contexts (Malawi, Ethiopia and Bangladesh) admitted using MUAC and (ii) describe the sensitivity of both MUAC and weight to episodes of disease experienced during the SAM treatment episodes in CMAM programmes in three country contexts (Malawi, Ethiopia and Bangladesh) admitted using MUAC. METHODS: Data collected under research conditions in Malawi were analysed for the correlation between MUAC and weight changes using the Pearson product–moment correlation coefficient (Pearson’s r). Further data from other CMAM programmes implemented under field conditions in Ethiopia and Bangladesh were similarly analysed. The association of growth failure following recent episodes of illness were assessed for MUAC and weight change using a two-by-two table, box-plots and Kruskal Wallis non-parametric rank sum test. RESULTS: MUAC and weight gain acheived over the entire treatment episode were strongly correlated in all three country contexts, Ethiopia (median Pearson's r = 0.816, 95 % CI = 0.782 - 0.845), Malawi (median Pearson's r = 0.843, 95 % CI = 0.802 - 0.876) and Bangladesh (median Pearson's r = 0.725, 95 % CI = 0.663 - 0.777). MUAC and weight changes at each outpatient visit were closely correlated (median Pearson’s r = 0.954, 95 % CI = 0.602 – 0.997) under research conditions. The field data from Ethiopia and Bangladesh showed similar correlation (median Pearson’s r = 0.945, 95 % CI = 0.685 – 0.998) and (median Pearson’s r = 0.939, 95 % CI = 0.705 – 0.994) respectively. MUAC and weight appear to respond rapidly and similarly to episodes of illness reported during outpatient treatment for SAM for MUAC, diarrhoea RR = 1.88 (95 % CI = 1.64 - 2.15), vomiting RR = 1.89 (95 % CI = 1.58 - 2.26), fever RR = 1.57 (95 % CI = 1.36 - 1.82) and cough1.42 (95 % CI = 1.22 - 1.65). Similar relative risks are seen for weight; diarrhoea RR = 2.03 (95 % CI = 1.77 - 2.31), vomiting RR = 2.09 (95 % CI = 1.77 - 2.47), fever RR = 1.76 (95 % CI = 1.53 - 2.03) and cough RR = 1.25 (95 % CI = 1.06 - 1.48). CONCLUSIONS: This study demonstrates a close relationship between MUAC and weight change during recovery from SAM under both research and operational field conditions. Furthermore, changes in both MUAC and weight are observed to occur similarly and rapidly during episodes of illness occurring during treatment with no lag effect on the part of MUAC. This presents the possibility for children undergoing outpatient treatment for SAM to be monitored using MUAC as an alternative to weight. Further research would be required to develop a tool which can be deployed safely and enable MUAC to be used as the sole anthropometric measure for admission, monitoring of recovery and discharge. This development would potentially allow the further decentralisation of the treatment of SAM thus improving programme coverage and child survival.