Abstract
OBJECTIVES: To investigate nutritional status and body composition in cognitively impaired older persons living alone. METHODS: This cross-sectional study included 1051 older adults (633 women and 418 men, mean age: 77.1 years) from the Takashimadaira study. The study participants were categorized according to whether they lived alone, which was confirmed via questionnaire, and had cognitive impairment, which was defined as having a Mini Mental State Examination-Japanese score ≤23. Nutritional status was evaluated using the serum albumin level. The fat-free mass index (FFMI) was calculated based on anthropometric and body composition measurements. A logistic regression model with the outcome of a low serum albumin level (serum albumin <4 g/dL) and low FFMI (<16 kg/m2 in men and <14 kg/m2 in women) were used to analyze the data. RESULTS: The percentages of participants in the living alone (-)/cognitive impairment (-) group, the living alone (+)/cognitive impairment (-) group, the living alone (-)/cognitive impairment (+) group, and the living alone (+)/cognitive impairment (+) group were 54.8%, 37.3%, 5.6%, and 2.3%, respectively. Compared to the living alone (-)/cognitive impairment (-) group, the living alone (+)/cognitive impairment (+) group was more likely to have a low serum albumin level (adjusted odds ratio = 3.10, 95% confidence interval = 1.31 to 7.33) and low FFMI (adjusted odds ratio = 2.79, 95% confidence interval = 1.10 to 7.06) after adjusting for potential confounders. CONCLUSION: Cognitively impaired older adults living alone had poorer nutrition than cognitively normal and cohabitating persons in this study. Our results highlight the importance of paying extra attention to nutritional status for this group of community-dwelling older adults.