Abstract
BACKGROUND: Childhood is a period marked by dynamic growth. Evidence of the association between childhood linear growth and school achievement comes mostly from cross-sectional data. We assessed associations between birth length, childhood linear growth velocities, and stunting with school achievement. METHODS: Newborns were recruited into the Ethiopian infant Anthropometry and Body Composition (iABC) birth cohort and followed from birth to 10 years. Velocities from 0 to 6 years were computed using linear spline mixed effect modelling. Stunting (height-for-age < -2 z-scores) was assessed at the ages of 1, 2, 4, 5, and 6 years. School achievement was measured by having a high (≥ 80%) combined high math, English, and science (MES) score and being at appropriate grade-for-age. Logistic regression models assessed associations between birth length, linear growth velocities and stunting with school achievement. RESULTS: Children's mean age was 9.8 years (standard deviation, SD 1.0, range 7-12 years). A 1 SD greater birth length increased the odds of achieving a high MES combined score by 1.42 (95% CI: 0.99, 2.03). A 1 SD increase in linear growth velocity from 6 to 24 months increased the odds of being in the appropriate grade-for-age by 1.66 (95% CI: 1.14, 2.43). Stunting at ages 4 and 6 years was associated with lower odds of achieving a high MES combined score: 0.43 (95% CI: 0.20, 0.93) and 0.31 (95% CI: 0.11, 0.89), respectively. Faster post-natal linear growth was not associated with school achievement. CONCLUSION: Greater birth length and higher growth velocity from 6 to 24 months were associated with higher school achievement and being in the appropriate grade-for-age, respectively. Children who experienced growth failure were less likely to achieve a high MES score. Interventions aimed at improving school achievement should address maternal and fetal nutrition and health, and monitor post-natal growth.