The association between undiagnosed diabetes and cognitive function: findings from the China health and retirement longitudinal study

Background The cognitive function of people with diabetes has gained an increasing interest in recent years, and this study focuses on exploring the relationship between undiagnosed diabetes and cognitive function among the middle-aged and elderly people in China. Methods The data came from the China Health and Retirement Longitudinal Study (CHARLS) which was conducted between July and October 2015. 9855 subjects were enrolled in the study. Executive function and episodic memory were used to assess cognitive function. The subjects were divided into three groups: no diabetes, diagnosed diabetes, and undiagnosed diabetes, and weighted multiple linear regression models were established to evaluate the association of undiagnosed diabetes with cognitive function. Results After controlling for covariates, undiagnosed diabetes was statistically associated with executive function (β = −0.215, P < 0.01). In the age group of ≥65 years, undiagnosed diabetes was statistically associated with executive function (β = −0.358, P < 0.01) and episodic memory (β = −0.356, P < 0.01). When adjusting for confounders, no statistically significant associations were found between diagnosed diabetes and cognitive function except in 45-54 age group (β = 0.374, P < 0.05). Conclusions The cross-sectional study suggested that undiagnosed diabetes was linked to poor cognitive function, especially in the elderly population. Timely diagnosis and active treatment of diabetes are important to reduce the occurrence of cognitive impairment. Further prospective cohort studies are required to articulate the association between undiagnosed diabetes and cognitive function.