Short Form were collected for analysis. A generalized estimating equation model was used after it was adjusted for age, educational level, five-item Geriatric Depression Scale, and problem of communication difficulty to identify potential modifiable risk factors for cognitive decline.
Results:The mean age of the participants was 85.7 ± 5.2 years, with a mean follow-up period of 41 ± 21.6 weeks. The prevalence of cognitive impairment (defined by MMSE <24) was 65.6%, whereas 34% of the study participants were positive for depressive symptoms. Approximately one-fifth of the study participants were using psychotropic agents, which was higher among participants with cognitive impairment (23.6% vs 15.6%, P < 0.05) than those without. In the generalized estimating equation model, physical function, nutritional status, depressive symptoms, ex-drinker, multimorbidity and stool incontinence were positively correlated with MMSE score; whereas advanced age, low educational level (<6 years), presence of communication difficulty and use of psychotropic agents were inversely associated with the MMSE score.Conclusions: Physical function and nutritional status were positively associated with the MMSE score, and use of psychotropic agents was negatively correlated with cognitive function. Further intervention study is required to improve the cognitive health of older adults living in the veterans' retirement communities. Geriatr Gerontol Int 2017: 17 (Suppl. 1): 7-13.