Anthropometric prevalence indicators such as stunting, wasting, and underweight are widely-used population-level tools used to track trends in childhood nutrition. Threats to the validity of these data can lead to erroneous decision making and improper allocation of finite resources intended to support some of the world's most vulnerable populations. It has been demonstrated previously that aggregated prevalence rates for these indicators can be highly sensitive to biases in the presence of non-directional measurement errors, but the quantitative relationship between the contributing factors and the scale of this bias has not been fully described. In this work, a Monte Carlo simulation exercise was performed to generate high-statistics z-score distributions with a wide range of mean and standard deviation parameters relevant to the populations in low- and middle-income countries (LMIC). With the important assumption that the distribution's standard deviation should be close to 1.0 in the absence of non-directional measurement errors, the shift in prevalence rate due to this common challenge is calculated and explored. Assuming access to a given z-score distribution's mean and standard deviation values, this relationship can be used to evaluate the potential scale of prevalence bias for both historical and modern anthropometric indicator results. As a demonstration of the efficacy of this exercise, the bias scale for a set of 21 child anthropometry datasets collected in LMIC contexts is presented.
Copyright: © 2024 Grange et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.