English  Français

A Structured-Additive Modeling of Diabetes and Hypertension in Northeast India

Strong P Marbaniang, International Institute for Population Sciences (IIPS)
Holendro Singh, BBC World Service Trust
H Lhungdim, International Institute for Population Sciences

This paper examined the demographic and socio-economic risk factors as well as the spatial disparity of diabetes and hypertension among adults 15-49 years in Northeast India. Using the Indian Demographic Health Survey (2015-16), the study consist of men and women between the age 15-49 years. The cut-off point for diabetes and hypertension was based according to the survey definition. A Bayesian Geo-additive model was used to determine the risk factors which comprised a spatial effect for district respondent residence. Prevalence of diabetes and hypertension in Northeast India was 6.38% and 16.21% respectively. Male respondent, urban residence, widowed/divorced/separated individuals show a higher prevalence of diabetes and hypertension. The functional relationship between household wealth index and diabetes and hypertension depicts almost a U-shape. As the household wealth status increases, its effect on diabetes increases. However, interestingly the inverse is observed in the case of hypertension. The spatial variation of diabetes is mainly due to unmeasured district specific factors that do not transcend boundaries, while for hypertension the spatial variation is due to factors which are similar to close proximity districts. The diabetes and hypertension control measures should take into account the local factors and non-local factors which contributes to the spatial heterogeneity.

Keywords: Health and morbidity, Bayesian methods / estimation, Spatial analysis/regression, Geographic Information Systems (GIS)

See extended abstract.

  Presented in Session P14.