Cognitive Reserve in rural and urban populations: Insights from two aging cohorts in southern India

BACKGROUND: There are no Indian studies estimating Cognitive Reserve (CR) across rural and urban aging populations. METHODS: We estimated CR from two ongoing aging studies in rural (CBR-SANSCOG, n = 4459) and urban (CBR-TLSA, n = 663) southern India. We used years of education (YOE), job skill level (JSL), social network diversity (SND) and multilingualism (ML) as factors and assigned weights based on their capability to predict cognitive performance (assessed using a culturally adapted cognitive test battery). We evaluated several candidate machine learning models and chose the linear regression based on its fit. FINDINGS: In the rural cohort, YOE, ML, and SND contributed significantly (Rural CR = 0.085×YOE + 0.184×ML + 0.030×SND), whereas YOE, ML and JSL were significant contributors for the urban cohort (Urban CR = 0.064×YOE + 0.184×ML + 0.197×JSL). INTERPRETATION: The contribution of CR factors differs across rural and urban Indian populations. Targeted interventions to enhance population-specific CR factors could reduce dementia risk.