SSM: Population Health (Sep 2022)
Investigating the direct and indirect associations between birth intervals and child growth and development: A cross-sectional analysis of 13 Demographic and Health Surveys
Abstract
There is considerable literature on the associations of short birth intervals with adverse perinatal outcomes. However, less is known about the associations with child growth and development. In this study, we investigated the associations between birth intervals and child growth and development and examined child illness, child diet, and maternal stimulation as potential mechanisms. We pooled Demographic and Health Survey data on 8300 children aged 36–59 months from 13 countries (Benin, Burundi, Cambodia, Cameroon, Chad, Congo, Haiti, Honduras, Rwanda, Senegal, Timor-Leste, Togo, and Uganda). Longer birth interval was defined as a preceding birth interval ≥33 months. Child growth was assessed using height-for-age Z-score (HAZ). Child cognitive and socio-emotional development were measured using the Early Childhood Development Index. Child morbidity was defined as any illness in the past two weeks. Child diet was assessed using dietary diversity score and maternal stimulation by the number of stimulation activities. We used generalised linear models to estimate associations between longer birth intervals and child growth and development. Structural equation modelling was used to assess direct and indirect effects. In our sample, 44% of children had a preceding birth interval ≥33 months, 42% were stunted, 25% were cognitively off-track, and 33% socio-emotionally off-track. Longer birth intervals were associated with higher HAZ (mean difference 0.23 (95% CI 0.14, 0.32)) and socio-emotional development (relative risk (RR) 1.04 (95% CI 1.00, 1.09), but not cognitive development (RR 1.02 (95% CI 0.98, 1.06). We observed no significant indirect effects via child illness, child dietary diversity, or maternal stimulation. Although longer birth intervals were beneficial for child growth and socio-emotional development, we found no empirical support for the biological and behavioural mechanisms we explored. Additional research is needed to investigate alternative mechanisms to elucidate underlying processes and inform future interventions.
