Assessing associational strength of 23 correlates of child anthropometric failure: an econometric analysis of the 2015-2016 National Family Health Survey, India

Despite the broad consensus that investments in nutrition-sensitive
programmes are required to reduce child undernutrition, in practice empirical
studies and interventions tend to focus on few nutrition-specific risk factors
in isolation. The 2015-16 National Family Health Survey provides the first
opportunity in more than a decade to conduct an up-to-date comprehensive
evaluation of the relative importance of various maternal and child health
and nutrition (MCHN) factors in respect to child anthropometric failures in
India. The primary analysis included 140,444 children aged 6-59 months with
complete data on 20 MCHN factors, and the secondary analysis included a
subset of 25,603 children with additional paternal data. Outcome variables
were stunting, underweight and wasting. We conducted logistic regression
models to first evaluate each correlate separately in age- and sex-adjusted
models, and then jointly in a mutually adjusted model. For all anthropometric
failures, indicators of past and present socioeconomic conditions showed the
most robust associations. The strongest correlates for stunting were short
maternal stature (OR: 4.39; 95%CI: 4.00, 4.81), lack of maternal education
(OR: 1.74; 95%CI: 1.60, 1.89), low maternal BMI (OR: 1.64; 95%CI: 1.54, 1.75),
poor household wealth (OR: 1.25; 95%CI: 1.15, 1.35) and poor household
air quality (OR: 1.22; 95%CI: 1.16, 1.29). Weaker associations were found
for other correlates, including dietary diversity, vitamin A supplementation
and breastfeeding initiation. Paternal factors were also important predictors
of anthropometric failures, but to a lesser degree than maternal factors.
The results remained consistent when stratified by children’s age (6-23
vs 24-59 months) and sex (girls vs boys), and when low birth weight was
additionally considered. Our findings indicate the limitation of nutritionspecific
interventions. Breaking multi-generational poverty and improving
environmental factors are promising investments to prevent anthropometric
failures in early childhood.