Factors associated with risk of developmental delay in preschool children in a setting with high rates of malnutrition: a cross-sectional analysis of data from the IHOPE study, Madagascar

Background: Madagascar has one of the highest rates of stunting and at-risk child development in the world:50% of Malagasy children have moderate to severe stunting. In 2016, a new 10 year National Nutrition Action Plan (PNAN III), supported by the World Bank, was initiated to help address stunting and developmental delay. PIVOT, a health care NGO operating in collaboration with the MOH, conducts a longitudinal study, the IHOPE, to assess the population level impact of their collaborative health system strengthening activities on the rural district of Ifanadiana in southeastern Madagascar. We report factors associated with risk to developmental delay in children ages 3 and 4 years old in Ifanadiana district in 2016. Methods: The household survey data are from a cross-sectional analysis of a one wave of panel data in 2016, at the time of the initiation of PNAN III. Women ages 15 to 49 were interviewed, using the Multiple Indicator Cluster Survey (MICS) Early Child Development Indicator (ECDI) module, which includes questions for physical, socio-emotional, learning and literacy/numeracy domains. ECDI data were analyzed using standardized z scores for relative relationships. We assessed 2 outcomes: at risk for delayed development compared to an international standard, and lower development than peers if ECDI z scores were <1 standard deviation below the international normed population and study mean, respectively. Covariates included demographics (maternal age, education level, household wealth, number of other children under 5 in the house), adult involvement (reading, singing, playing with child), household environment (number and type of childrens books or toys, whether child was left alone or in the care of another child <10 years), and child health (wasting, stunting, underweight, ill in last 4 weeks, vaccination status). We included variables significant in univariable analysis at an alpha of <0.1 in a multivariable model. We constructed the final models using backward stepwise regression on generalized linear models, clustered at the sampling level. Results: Of 432 children ages 3 and 4 years, 173 were at risk for delay compared to international norms and 68 children (16.0%) were considered lower-development compared to their peers. 50.5% of the children had moderate to severe stunting and 19.0% had severe stunting. Greater paternal engagement (OR 1.59(1.13, 2.21)) was associated with being at risk for delay based on ECDI z scores <-1 compared to international norms Factors associated with lower risk for delay included having some adult play with the child and greater engagement from a non-parental adult. Factors associated with low development compared to peers included: under age 5 in the household, having a teenaged mother (age <20 years--OR3.89 (1.32, 11.48)), and having an adult who recently took the child outside. Factors associated with less developmental risk included increased number of developmentally stimulating activities by someone other than a parent(OR 0.28(0.16, 0.50)), and primary child toys are found objects rather than store bought or homemade(OR 0.33 (0.14, 0.73)). Discussion: In this setting of high malnutrition, stunting in children is not associated with lower-development compared to their peers, when adjusting for other factors. Families with teenaged mothers are more likely to have children with lower-development. Non-parent adult involvement seems very relevant to child development in this setting. Further research in the IHOPE longitudinal study may help to clarify direction of association. Interventions targeting adolescent girls before they become parents may be of particular utility in this setting.