A spatial analysis of disparity in the prevalence of stunting rates among children aged under five between rural and urban areas in Peru

The main objective of this article is to assess disparities in the prevalence of stunting rates among children aged 6–59 months between rural and urban areas in Peru, through the spatial analysis of the evolution of stunting rates in small geographic areas (districts and provinces), over the 2013–2023 period. The data from the National Demographic and Health Survey (ENDES) carried out in 2013, 2018 y 2023 was analyzed using the model-assisted approach. Specifically, in this research, the spatial small area modeling is employed in order to address the problem of a very low representation of rural areas in the survey, that results in missing or unreliable direct estimates of the prevalence of stunting rates in these areas. An inadequate representation of the rural areas where the problem of stunting is more critical, compared to the urban areas, may present an obstacle for identifying the regions where the situation is more pressing, and could result in an incorrect assessment of the nationwide magnitude of the childhood stunting and its evolution over time. The approach utilized in this article permits the derivation of the prevalence of stunting rates estimates in out-of-sample areas, and improving the direct estimates obtained in the areas with small sample sizes. This allows a more insightful look into the disparity between the urban and rural areas over the last decade. In particular, we have shown that despite a substantial reduction in the prevalence of stunting rates observed in all regions over the studied period, the disparities between the urban and rural areas remain large.