Effects of Metabolic Factors on Cognitive Function in Elderly Korean: A Prospective Community-Based Cohort Study

Abstract

Objective As geriatric population grows continuously, cognitive dysfunction becomes a major problem in quality of life of both patients and their families. It is important to understand whether metabolic factors such as hyperglycemia, insulin resistance, and obesity could be risk factors of cognitive dysfunction.

Method The study was conducted in the Ansung Cohort Study, the ongoing prospective community-based cohort study in Korea. A total of 1,387 participants aged over 65 years at baseline were followed up for a mean of 6 years. Of them, the final analysis was performed in 422 participants who were evaluated with Korean mini-mental status examination score (K-MMSE) at both baseline and follow-up. Pearsons correlation analyses and multivariate linear regression analyses for change of K-MMSE score were carried out.

Results Mean age at baseline was 69.3 ± 2.9 years, and 222 participants (52.6%) were men. Mean duration of education was 7.1 ± 3.6 years. During a mean follow-up of 6 years, K-MMSE score was significantly decreased with 1.0 (interquartile range [IQR] of 4.0), homeostasis model assessment of insulin resistance (HOMA-IR) was not significantly changed during follow-up. Participants with more decreased K-MMSE score (∆ K-MMSE) had shorter duration of education (p = 0.001), older age (p = 0.022), higher baseline K-MMSE score (p < 0.001), and increased insulin resistance (∆ HOMA-IR, p = 0.002). Participants with lower K-MMSE at their baseline had shorter education duration (r = 0.393, p <0.001), higher KDSQ score (r = -0.129, p =0.008), and higher GDS-K score (r = -0.128, p 0.008). Also, participants with lower K-MMSE at follow-up had older age (r = -0.119, p = 0.014) and shorter duration of education (r = 0.400, p < 0.001). The correlation between ∆ K-MMSE and ∆ HOMA-IR remained significant after adjustments for age, gender, baseline K-MMSE score, duration of education, baseline Korean geriatric depression scale, smoking status, history of diabetes and hypertension, body mass index, and apolipoprotein E4 genotype status (B = -0.201, p = 0.002). Correlations between ∆ K-MMSE and baseline HbA1c, ∆ K-MMSE and BMI were not significantly associated after the adjustments (B = -0.055, p = 0.257 for HbA1c, B = -0.008, p = 0.873 for BMI).

Conclusions During 6 years of follow-up in Korean elderly population, increased insulin resistance was significantly associated with decreased cognitive function.