# Everyday discrimination among middle-aged and older adults in India: a multilevel cross-sectional analysis from the Longitudinal Ageing Study in India

## TL;DR

This study examines everyday discrimination among older adults in India, finding significant geographic and demographic variations that affect health and well-being.

## Contribution

The study provides new insights into the distribution and correlates of everyday discrimination among Indian adults aged 45 and above.

## Key findings

- States like Nagaland and Mizoram had lower discrimination scores compared to the national average.
- Men, non-married individuals, migrants, and rural residents experienced higher everyday discrimination rates.
- Functional disability and physical or mental impairment were strongly associated with increased discrimination.

## Abstract

Everyday discrimination (ED) has adverse effects on health and well-being. This study highlights understudied trends in the distribution, correlates, and geographic variation of ED among Indian adults aged 45 and above. The analysis of 61,722 participants in the Longitudinal Ageing Study in India (2017-18) revealed significant state/union territory (UT)-level variation. While Nagaland, Tripura, Mizoram, and Lakshadweep had comparatively lower ED scores than the national average, Jammu and Kashmir, Delhi, Chhattisgarh, and Karnataka had higher scores. Additionally, using multilevel negative binomial regression, we found that men, non-married adults, migrant residents, and adults in rural areas had higher ED rates. In general, with increases in education level and household monthly per capita income, there was a reduction in ED rates. Notably, adults with a functional disability (Incident Rate Ratio (IRR) = 1.43 [95% confidence interval: 1.32, 1.55]) and physical or mental impairment (IRR = 2.15 [1.88, 2.45]) had significantly higher ED rates than those without. We also partitioned the geographic variation in ED and found that more geographic variance was explained by the community (village/ward) level than by the state/UT level, accounting for roughly 60% and 40% of the variation across models, respectively. Our findings suggest that community-based contextual factors necessitate further research.