Anthropometric measurements for detecting low birth weight

Abstract

Background In several provinces of Eastern Indonesia, the
majority of births take place at home (60%) and are assisted by
traditional birth attendants. Most of these newborns do not have
their birth weight recorded, due to lack of available weighing scales
or lack of skill to perform the measurement, especially in rural
areas. As such, an early iden tification oflow birth weight cases is
needed to prevent infan t morbidity and mortality.
Objective To assess anthropometric measurements including calf,
chest, and h ead circumferences as a method of choice for detecting
low birth weight, as substitute for actual weighing.
Methods This cross-sectional study was performed at Banjar
Barn, South Kalimantan, Indonesia, from January to March 2012.
Subjects were full term, singleton, and live-born infants during
the study period, and obtained from private clinics by a purposive
sampling procedure. Calf, chest, and h ead circumferences were
measured to identify the most suitable substitute for birth weight
using Pearson's correlation, ROC, sensitivity, and specificity.
Results In this study, a correlation was shown between birth
weight and all anthropometric measurements. Optimal calf,
chest, and head circumference cutoff points to identify low birth
weight infants were 10.3 cm, 30. 7 cm, and 3 1.2 cm, respectively.
The area under the curves (AUC) showed good accuracy for all
measuremen t types. Calf circumference had the closest estimated
true prevalence to the true prevalence (8.52% and 8.6%, respectively)
compared to the other measurement types.
Conclusion Calf circumference is the most suitable measurement
as a substitute for birth weight, due to its estimated true
prevalence.