Linear growth trajectories in the first two years of life predict attained linear growth and stunting at five years: Results from the MAL-ED multi-country birth cohort study.

Abstract

Background Linear growth faltering and stunting are associated with increased childhood mortality and morbidity, as well as impaired physical growth and cognitive development. Objectives We aimed to identify groups of children with distinct linear growth trajectories during their first two years of life and to determine whether these early-life trajectories were associated with attained linear growth and stunting at five years of age. Methods We used data from the MAL-ED birth cohort study for this analysis. Latent class growth modeling (LCGM) was applied to identify distinct classes of children who followed similar trajectories of length-for-age z-score (LAZ) from 0 to 24 months of age. Mixed-effects linear and logistic regression models were used to examine the associations between LCGM-derived trajectories and height-for-age z-score (HAZ) and stunting at age 60 months, respectively, with study site included as a random effect. Results We identified five LAZ trajectories among 1469 children aged 0-24 months, designated as follows: Class 1: severely attenuated linear growth (9%); Class 2: moderately attenuated linear growth (25%); Class 3: mildly attenuated linear growth (34%); Class 4: stable linear growth (25%); Class 5: improved linear growth (7%). In adjusted models, LAZ trajectories during the first two years of life were significantly associated with HAZ and stunting at five years. Compared with the stable linear growth class, children in the improved linear growth class had a predicted 0.86 higher HAZ at age five (95% CI: 0.67, 1.04), whereas those in the severely attenuated linear growth class had substantially lower HAZ at age five (β = -2.10; 95% CI: -2.26, -1.95). Similar associations were observed for stunting at five years. Conclusions Linear growth trajectories during the first two years of life are critical predictors of attained linear growth and stunting at five years. Greater emphasis should be placed on improving early-life linear growth through community-based interventions.