LAUSR

Performancebased assessment is considered the gold standard for measuring child neurodevelopment (ND). However, such assessment is often not feasible in lowresource settings (LRSs) because it is timeconsuming, resourceintensive, and available assessment tools are often not validated in the local language or lack regional or local norms. In the absence of direct developmental assessment, estimates of children at increased risk of ND impairment (‘ND risk’) and health policies that aim to improve child ND are usually based on populationbased metrics of child linear growth (i.e., stunting defined as <2 SD below the mean in heightforage or lengthforage, WHO). However, there is growing evidence that while stunting continues to be a robust proxy for child physical health and nutritional status, it is likely not the best proxy for ND risk, as the causal pathways of stunting and ND are complex and incompletely overlap. We propose that occipitofrontal head circumference (OFC) may be a more direct and appropriate proxy, as well as a feasible and costeffective method of identifying children at highest ND risk in LRSs. Research from highincome countries (HICs) suggests that OFC is a robust anthropometric measurement of brain volume, and several studies have demonstrated this relationship on neuroimaging. An extensive body of research from HICs has shown that when an OFC is 2 SD or more below the mean on the WHO growth charts, consistent with a diagnosis of microcephaly, there is an elevated risk of adverse ND outcome. Microcephaly is a relatively rare occurrence in HICs. For example, in the USA and Europe, the reported prevalence of microcephaly ranges from 2.0 to 14.7 per 10 000 live births. 4 5 The Zika epidemic both heightened and highlighted the concern about microcephaly in children in LRSs and led to further research showing that its prevalence may be substantially higher than in HICs. Studies from Guatemala and Brazil conducted prior to the Zika epidemic reported prevalences of microcephaly of 1216 and 350 per 10 000 live births, respectively. 7 Additionally, several studies in LRSs have shown increasing prevalence of microcephaly as children age, similar to what is commonly observed with stunting. In a study conducted in India, 33% of children were reported to meet criteria for microcephaly at birth, increasing to 50% by the end of first year of life. A similar phenomenon of increasing prevalence with age was reported in rural Nepal, where over half of children had an OFC equal to 2 SD or below WHO growth standards by age 4 years. While the specific aetiology of microcephaly is often not known, the higher prevalences reported in LRSs compared with HICs, and known associations, such as lower socioeconomic status and parental OFC, strongly implicate the intergenerational impact and cumulative adverse risks of living in poverty on child growth. Some have disregarded these reports of elevated prevalence of microcephaly in LRSs, arguing that the numbers must be inflated due to an inappropriate application of a global growth standard in groups of people who are genetically smaller in stature. This premise is challenged by the 2006 WHO Multicentre Growth Study, in which very small differences in growth were found among children globally when nutritional and other risks associated with living in poverty were minimised. A pair of studies from Nepal showed that healthy children had OFCs that were within normal range when using the WHO growth standards but that 50% of children living in more rural, impoverished areas met criteria for microcephaly using the same metrics. While it is well known that many risk factors associated with living in poverty, including elevated rates of preterm birth, intrauterine growth restriction and congenital infections, have adverse impacts on the growth and development of the brain, OFC has been historically understudied in LRSs. One reason is the belief in the global health community that OFC is spared under conditions of nutritional stress. However, recent data from LRSs suggest that OFC might be a sensitive marker of the ND risks of living in poverty, may be more directly affected by repeated illness and undernutrition than previously appreciated, and likely incompletely overlaps with stunting, suggesting both shared as well as separate aetiologies. 12 13 In order to understand whether or not clinically significant microcephaly is being overidentified with global growth standards, we must study the direct association between child OFC and ND outcomes in all resource settings. This association has been identified in HICs, and research in LRSs is beginning to reveal similar associations. In the multicountry MALED Study, OFC was associated with ND outcome and noted to be a stronger predictor of ND than stunting status. Several other studies around the world, including our studies in rural Guatemala, have also reported an association between small OFC and poor ND outcome in children. 14 Therefore, the clinical and public health community should consider that global growth standards may be applied to OFC, can serve as an indicator of the risks associated with living in poverty and of ND risk, and that the prevalence of microcephaly may be equivalently high and even more clinically concerning than stunting. More research focusing on the association between child OFC and ND risk in LRSs is needed; child OFC should be routinely collected at all medical visits (with global harmonisation of measurement methods to minimise the potential for errors) in the context of nutritional monitoring and in relevant research studies involving young children. These additional data will help us determine if the measurement of OFC can serve as a more accurate proxy of ND risk than stunting, and whether OFC is feasible to measure accurately. Because research suggests that the <2 SD cutoff to define stunting may be arbitrary and continuous heightforage analysis may better define risk, a similar analysis to determine whether OFC should be assessed as a continuous metric or specific cutoff should be conducted. In addition, tracking OFC trends throughout early Children’s Hospital Colorado, Aurora, Colorado, USA Department of Physical Medicine and Rehabilitation, University of Colorado Denver School of Medicine, Aurora, Colorado, USA Center for Global Health and Department of Epidemiology, Colorado School of Public Health, Aurora, Colorado, USA