TY - JOUR T1 - Factors associated with head circumference and indices of cognitive development in early childhood JF - BMJ Global Health JO - BMJ Global Health DO - 10.1136/bmjgh-2020-003427 VL - 5 IS - 10 SP - e003427 AU - Laura Nicolaou AU - Tahmeed Ahmed AU - Zulfiqar Ahmed Bhutta AU - Pascal Bessong AU - Margaret Kosek AU - Aldo A M Lima AU - Sanjaya Shrestha AU - Ram Chandyo AU - Estomih R Mduma AU - Laura Murray-Kolb AU - Brooks Morgan AU - Matthew R Grigsby AU - William Checkley A2 - , Y1 - 2020/10/01 UR - http://gh.bmj.com/content/5/10/e003427.abstract N2 - Background While head circumference (HC) has been related to intracranial volume and brain size, its association with cognitive function remains unclear. We sought to understand the relationship among various biological and socioeconomic risk factors, HC and cognitive development.Methods We analysed data across resource-poor settings in Bangladesh, India, Nepal, Peru, South Africa and Tanzania from the Etiology, Risk Factors and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development longitudinal birth cohort study. Participating children were enrolled and followed up between 2009 and 2014. A final sample of 1210 children aged 0–24 months were included in the analyses. The main outcomes were HC for age Z-score and cognitive, gross motor and language scores from Bayley Scales of Infant Development-III tests. Length, weight and HC were measured monthly, and cognitive tests were administered at 6, 15 and 24 months of age. To disentangle the associations between risk factors and HC from linear growth and to distinguish the direct and indirect effects of these risk factors on cognitive function, we conducted mediation analysis using longitudinal models to account for all data measured during follow-up.Results Average HC-for-age Z-score (HCAZ) was −0.54 (95% CI −0.47 to −0.62) near birth and −1.01 (95% CI −0.94 to −1.08) at 24 months. Children with higher enrolment weight (p<0.0001), higher socioeconomic score (p=0.00037) and taller mothers (p=0.00084) had higher HCAZ at all ages, while enteropathogen infection (p=0.013) and more febrile episodes (p=0.013) were associated with lower HCAZ. The associations between HCAZ and enrolment weight-for-age, maternal height, socioeconomic status or pathogen burden were partly mediated through their associations with length-for-age. HCAZ showed no association with cognitive, gross motor or language skills at 6, 15 and 24 months of age.Conclusions The main risk factors associated with HC are similar to those associated with body length, and HC is not related to cognitive function. ER -