Weight-for-height Z score and mid-upper arm circumference as predictors of mortality in children with severe acute malnutrition

Abstract

Background: Mortality associated with severe acute malnutrition (SAM) could be reduced by screening malnourished children for those most vulnerable to death. We compared the weight-for-height Z score (WHZ) and mid-upper arm circumference (MUAC) as predictors of mortality in children with SAM. Methods: We conducted a retrospective study spanning over 8 years, using records of children aged 6-59 months, hospitalized for SAM and discharged alive or who died during hospitalization. Results: Area under the curve was greater for MUAC [0.809 (95% CI, 0.709-0.911, p = 0.001)] than WHZ [0.649 (95% CI, 0.524-0.774, p = 0.032)]. MUAC predicted death better [sensitivity: 95.5%, specificity: 25.0%, positive likelihood ratio (PLR): 1.27, negative likelihood ratio (NLR): 0.18] than WHZ (sensitivity: 86.4%, specificity: 21.4%, PLR: 1.10, NLR: 0.64). Best MUAC and WHZ cut-offs for predicting death (10.3 cm and -4, respectively) were most accurate in infants aged <12 months, the former being more accurate. Conclusion: MUAC predicts death better than WHZ in children with SAM.