District Level Assessment of Spatial Clustering and Determinants of Diabetes Mellitus among Older Adolescents and Young Adults in India

Abstract

Globally non-communicable disease is the leading cause of death. Amongst all, diabetes mellitus (DM) represents the most prevalent non-communicable disease in India. However, the burden of DM shifting towards the lower ages has received relatively little attention in India. Thus, this study aims to assess the district level spatial clustering and determinants of DM among older adolescents and young adults. The study used data from the fourth round of the National Family Health Survey. A person in the age group 15-29 years of age is considered to be older adolescents and young adults. Moran’s I and LISA were used to examine the presence of spatial autocorrelation and clustering respectively. Simultaneous Autoregressive Model was used for the assessment of determinants. Prevalence of DM across 640 districts ranged from 0.15 percent to 35 percent. Global Moran’s I is 0.61 implying significant spatial correlation. There is significant high-high clustering of the prevalence of DM in the districts of northeastern states, eastern states, southern states and Delhi National Capital Regions. And, significant low-low clustering of the prevalence of DM in the districts of northern states and central India. Overweight and professionals who are physically inactive are at higher risk of acquiring DM. As the older adolescent and the young adults are the potential assets to the nation, reduction in the prevalence of DM among older adolescents and young adults is important and would be possible only if area-specific measures would be adopted on those clusters of districts where DM is high considering the other intervening covariates.