Cognitive function and its risk factors in a national survey of older adults in Taiwan: A latent class analysis

[[abstract]]Background: This study aimed to categorize older adults into subgroups according to their health behaviors and compare the risk of cognitive impairment of these groups. Methods: Cross-sectional data were from the 2013 National Health Interview Survey in Taiwan. A total of 2,817 older adults were analyzed using the latent class analysis (LCA) method to categorize their health behaviors. Logistic regression was used to identify the potential risks and protective factors of cognitive impairment. Results: Latent class analysis (LCA) identified six classes. The results of logistic regression showed that the physically and socially inactive group faced a 68% higher risk of cognitive impairment (odds ratio, 1.68; 95% confidence interval, 1.20-2.35) and the physically and socially active group faced a 68% lower risk of cognitive impairment (odds ratio, 0.32; 95% confidence interval, 0.19-0.55) than the reference (sound-health-status) group. Conclusions: The findings offer insights and implications that are useful for the future planning of related interventions to reduce the risk and incidence of cognitive impairment