Last modified: 2020-02-25
Abstract
Introduction: WHO classifications of adult underweight, overweight and obesity are based on body mass index (BMI). That commonly raise logistical problems in large scale community based surveys in low- and middle-income settings. Indices based on body circumferences have not been validated in relation to BMI low-income settings in Sub-Saharan Africa.
Objective: The aim of this study was to evaluate the use of MUAC, waist-, hip- and neck circumference measurements against BMI as options for screening of women’s nutritional status.
Methods: A community based cross-sectional study was conducted in Addis Ababa, Ethiopia where randomly selected 4936 women with children under five year of age were involved. Anthropometric measures were collected using standard methods. Receiver operating characteristic (ROC) analysis was used for each anthropometric variable to determine their sensitivity and specificity for predicting the risk of under or overweight according to Youden index using BMI as gold standard.
Results: To define underweight, MUAC at a cut-off value of <25 cm derived with the highest Youden index of 0.748 (Sensitivity= 93.7%, Specificity=81.1% and area under curve 0.934). For overweight, a MUAC 28.0 – 30.0 cm derived with the highest Youden index of 0.644 (Sensitivity = 83.2%, Specificity= 81.2%, area under curve=0.908) while for obesity a MUAC > 30.0cm derived with the highest Youden index of 0.701(Sensitivity = 87.0%, Specificity =83.2%, area under curve = 0.925).
Discussion: We found the MUAC measurement to have a fairly good sensitivity and specificity for screening underweight and overweight/obese. Therefore, MUAC could be a pragmatic nutritional screening instrument when measuring BMI in low-income settings is logistically and economically not feasible