Global Health
Intersectional Inequalities in the Context of Rurality on Cognitive Functioning: A MAIHDA Analysis of the Harmonized Cognitive Assessment Protocol Network in Mexico, India, and China L. Paloma Rojas-Saunero* L. Paloma Rojas-Saunero Janette Vazquez Erika Meza Cristina A. F. Román Sneha Sarah Mani Ruijia Chen Yingyan Wu Lindsay Kobayashi Justina Avila-Rieger Elizabeth Rose Mayeda Marcia Pescador Jimenez
INTRODUCTION: There is limited evidence on how living in rural areas intersects with systems of oppression to influence cognitive health, particularly in non-Western countries. Guided by an intersectional framework, we examine how rurality intersects with sexism, ageism, and limited educational access to characterize cognitive performance in India, Mexico, and China.
METHODS: We use data from 3 participating countries of the Harmonized Cognitive Assessment Protocols (HCAPs): Mexico (MEXCog, n=2024; age: 54-104), India (LASI-DAD; n=2319; age: 60-105) and China (CHARLS-HCAP, n=9755; age: 60-108). We created 24 intersectional strata based on residence (rural/urban), sex/gender (women/men), age 75+ (yes/no) and education (no/early; primary; secondary/higher). Cognitive performance was measured cross-nationally with harmonized general cognitive function factor scores and z-scored. We performed a parallel multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA) to calculate the variance in cognitive performance due to intersectional strata. We fitted multilevel linear models with random effects for intersectional strata adjusting for age, and a second model with main effects for rurality, sex/gender, age 75+ and education. We calculated the proportional change in variance (PCV) to estimate the cognitive performance variance explained by the interaction of these social identities after accounting for main effects.
RESULTS: The variance between intersectional strata accounted for 41% of the total variance in cognitive function in MEXCog, 33% in LASI-DAD, and 34% in CHARLS-HCAP. After including main effects, the PCV was 9% in MEXCog, 19% in LASI-DAD and 5% in CHARLS-HCAP. Women ages 75+ in rural areas with no education had lower cognitive performance compared to other strata.
CONCLUSION: These results highlight that cognitive health is shaped by the interaction of multiple social identities, with varying impacts across countries.