Abstract

India has one of the highest burdens of childhood undernutrition in the world. The two
principal dimensions of childhood undernutrition, namely stunting and underweight can be
significantly associated in a particular population, a fact that is rarely explored in the extant
literature. In this study, we develop and apply a copula geoadditive modelling framework to
assess the spatial distribution and critical drivers of the dual burden of childhood stunting
and underweight in India while accounting for this correlation. We apply the framework on
nationally representative data of 104,021 children obtained from the National Family Health
Survey 5 (NFHS 5). Prevalence of stunting, underweight and their co-occurrence among

under 5 children were 35.37%, 28.63% and 19.45% respectively with significant positive as-
sociation between the two (Pearsonian Chi square = 19346, p-value = 0). Some of the factors

which were significantly associated with stunting and underweight were child gender (Ad-
justed Odds Ratio (AOR) = 1.13 (1.12) for stunting (underweight)), birthweight (AOR =

1.46 (1.64) for stunting (underweight)), type of delivery (AOR = 1.12 (1.19) for stunting (un-
derweight)), prenatal checkup (AOR = 0.94 (0.96) for stunting (underweight)) and maternal

short-stature (AOR = 2.19 (1.85) for stunting (underweight)). There was significant spatial
heterogeneity in the dual burden of stunting and underweight with highest prevalence being
observed in eastern and western states while northern and southern states having relatively

lower prevalence. Overall, the results are indicative of the inadequacy of a “one-size-fits-
all” strategy and underscores the necessity of an interventional framework that addresses

the nutritional deficiency of the most susceptible regions and population sub-groups of the
country.