Diabetes and Cognitive Health in India: A Nationally Representative Survey of Adults Aged 45 Years and Older

Abstract

Background: Few studies have assessed the relationship between diabetes and cognitive health outside of high-income countries. We described this relationship in India, a middle-income country of 1.4 billion people. Methods: Data were analyzed from 58,027 adults aged 45 years or older in the baseline wave of the nationally representative Longitudinal Aging Study in India, from 2017 to 2019. Outcomes were scaled cognitive scores (mean of 0 and standard deviation [SD] of 1) and cognitive impairment, defined as a cognitive score 1.5 standard deviations or below the age- and education-matched mean. Diabetes was defined as a self-report of a prior diabetes diagnosis made by a health professional or having a measured hemoglobin A1c ≥6.5% at the time of the LASI survey. To describe the cross-sectional relationship between cognitive health and diabetes, we constructed survey-weighted multivariable linear and logistic regression models controlling for prespecified covariates, including age, sex, sociodemographic characteristics, and rural versus urban residence. Results: In age- and sex-adjusted models, people with diabetes had cognitive scores that were 0.25 SD higher (95% CI: 0.23 to 0.27) and had a 1.3% (95% CI: 0.7 to 1.8) lower prevalence of cognitive impairment than people without diabetes. Differences persisted even when adjusting for demographic, socioeconomic, and geographic characteristics. Rural versus urban residence modified the relationships of diabetes with cognitive score (P = 0.001) and cognitive impairment (P = 0.004). In fully adjusted models, rural respondents with diabetes had 0.05 SD (95% CI: 0.03 to 0.07) greater cognitive scores and 1.7% (95% CI: 0.9 to 2.5) lower prevalence of cognitive impairment than those without diabetes. In urban areas, respondents with and without diabetes had similar cognitive scores and prevalence of cognitive impairment. Conclusion: Urban-rural health differences, the nutrition transition, and social conditions likely influence the observed cross-sectional relationship between diabetes and cognitive dysfunction in India, leading to different associations than previously reported in many high-income countries.