Background: Recent several studies in both humans and animal models suggest that insulin resistance has negative effects on cognition, there were few evidences from large papulation study. In the current study, we evaluated the potential relationships between the triglyceride glucose index (TyG index), surrogate marker of insulin resistance, and dementia using a large-scale population dataset from the National Health Information Database (NHID).

Methods: We did a retrospective, observational, cohort study of 13,149,723 persons aged older than 40 years from 2009 to 2015 using NHID. We used the TyG index as a measures of insulin resistance and divided the subjects into quartiles groups based on TyG index. The incidence of demetia was assessed using the hazard ratios (HRs) estimated with cox proportional hazard modeling.

Results: During a median follow-up period of 5.26 years, dementia was diagnosed in 252,114 (1.92%) participants. Among them, Alzheimer disease (AD) were diagnosed 73% and vascular dementia (VD) were diagnosed 27%. Multivariate-adjusted hazard ratios (HRs) for patients in highest quartile of TyG index were higher for dementia (HRs =1.147; 95% confidence intervals [CI] 1.132-1.161), for AD (HRs =1.120; 95% CI 1.103-1.136), and for VD (HRs =1.220; 95% CI 1.177-1.264) compared with subjects in the lowest quartile of TyG index. These effects were independent of age, sex, smoking, physical activity, body mass index, systolic blood pressure, and total cholesterol.

Conclusions: In our large population study, TyG index, surrogate marker of insulin resistance was associated with an increased risk of dementia, even including AD, and it was independent of traditional cardiovascular risk factors.


S. Hong: None. K. Han: None. C. Park: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at