Analysing concordance between MUAC, MUACZ, and WHZ in diagnosing acute malnutrition among children under five in
Sydney Garretson1, Shelley Walton1, Kemish Kenneth Alier1
1Johns Hopkins University, Department of International Health, Baltimore, Maryland, USA.
Insights
Mid-upper arm circumference (MUAC) alone misses many cases of child wasting. Age-adjusted MUAC z-scores (MUACZ) and specific thresholds show promise for improving malnutrition screening in children under five.
Area of Science:
- Pediatrics
- Global Health
- Nutrition Science
Background:
- Acute malnutrition (wasting) affects millions of children under five globally.
- Mid-upper arm circumference (MUAC) is a common screening tool in low-resource settings.
- MUAC alone may miss a significant proportion of wasted children identified by weight-for-height z-score (WHZ).
Purpose of the Study:
- To evaluate the diagnostic accuracy of MUAC and MUAC z-score (MUACZ) compared to WHZ in identifying wasting in children under five.
- To explore alternative MUAC and MUACZ thresholds for improved wasting detection.
- To analyze the concordance between different wasting indicators.
Main Methods:
- Secondary analysis of 1408 children under five from a cluster-randomized controlled trial in Somalia.
- Comparison of wasting prevalence using MUAC, MUACZ, and WHZ.
- Linear regression modeling and receiver operating characteristic (ROC) analysis using WHZ as the reference standard.
Main Results:
- Wasting prevalence was 1.5% (MUAC), 8.5% (MUACZ), and 14.8% (WHZ).
- MUAC alone missed 94% of WHZ-defined wasted children; MUACZ showed fair concordance (κ=0.385).
- Optimized MUAC and MUACZ thresholds identified over 82% of wasted children aged 9-23 months and over 76% aged 24-59 months.
Conclusions:
- Current MUAC screening guidelines may need revision due to poor concordance with WHZ.
- Age-specific MUACZ thresholds could improve the detection of wasting and treatment coverage.
- Policymakers should consider the impact of updated screening guidelines on healthcare facilities and treatment availability.
Background:
Acute malnutrition, also known as wasting, affected an estimated 45 million children under five (CU5) globally in 2023. Wasting is measured using a child's mid upper arm circumference (MUAC), weight-for-height z-score (WHZ), or nutritional oedema. In low-resource contexts, MUAC is often the only measurement used to regularly screen for malnutrition, but recent research suggests MUAC alone fails to diagnose 25-80% of WHZ-wasted children. Mid upper arm circumference, an age-adjusted MUAC z-score, may identify additional cases missed by MUAC alone.
Methods:
This was a secondary analysis of 1408 CU5 enrolled in a cluster-randomised controlled trial in two high-wasting regions of Somalia. This analysis explored wasting prevalences, concordance between indicator pairs, linear regression modelling, and receiver operating characteristic (ROC) analysis using WHZ as a reference standard to examine alternative MUAC and MUACZ thresholds.
Results:
Wasting prevalence was 1.5% by MUAC, 8.5% by MUACZ, and 14.8% by WHZ. Mid upper arm circumference alone failed to identify 94% of WHZ-wasted children. There was slight concordance between MUAC and WHZ (kappa (κ) = 0.089) and fair concordance between MUACZ and WHZ (κ = 0.385). Linear regression indicated that MUAC, age, sex, and stunting were all statistically significant variables for estimating child WHZ. Using WHZ as a reference standard, a MUAC of 13.7cm and MUACZ of -0.9 both accurately diagnosed >82% of wasted children ages 9-23 months. A MUAC of 14.4 cm and MUACZ of -1 both accurately diagnosed >76% of wasted children ages 24-59 months.
Conclusions:
Poor concordance between MUAC and WHZ should prompt review of wasting measurement guidelines for this population. Stratified analyses and regression modelling showed increased MUAC with age among CU5. MUACZ or age-specific MUAC thresholds may identify more truly wasted children, improving coverage of treatment interventions. Policymakers should examine how adapting screening guidelines impacts health facilities and treatment availability. Future studies should consider long-term outcomes and mortality associated with increased thresholds of MUAC and MUACZ.
Registration:
The cluster-RCT is registered at ClinicalTrials.gov, ID: NCT06642012.
Related Concept Videos
01:13Multiple Comparison Tests
It would be easy to compare two samples using a significance alpha level of 0.05. In other words, there is only one sample pair to be compared. However, it would be difficult to identify a significantly different sample if the number...
01:23Test for Homogeneity
01:17z Scores and Area Under the Curve
01:21Wilcoxon Rank-Sum Test
01:23McNemar's Test