Longitudinal association between eating alone and deterioration in frailty status: The Korean Frailty and Aging Cohort Study

Background: Among risk factors of developing frailty, dietary factor played an important role as a potentially modifiable risk factor. Eating alone is associated with malnutrition, depression, and social isolation, which are risk factors of frailty. We evaluated the longitudinal association between a change to eating alone and deterioration in frailty status in a cohort of community-dwelling elderly persons. Methods: The study subjects were 2072 non-frail Korean elderly persons aged 70-84 years who were recruited for the Korean Frailty and Aging Cohort Study (KFACS). The subjects were divided into 4 groups based on changes in eating with others or alone between the baseline survey (2016-2017) and the follow-up survey (2018-2019): group I (ate with others consistently), group II (ate with others at baseline but ate alone at follow-up), group III (ate alone at baseline but ate with others at follow-up), group IV (ate alone consistently). We assessed physical frailty using the Cardiovascular Health Study (CHS) frailty phenotype. The association between changes in eating with others or alone and frailty progression was assessed by multiple logistic regression analysis after adjusting for covariates. Results: The mean age of the study subjects was 76.2 (SD: 3.8) years old and 50.8 % were female. At follow-up, 364 new cases (34.5 %) of pre-frailty (n = 348) and frailty (n = 16) were identified among those who were robust at baseline (n = 1056), while 88 new cases (8.7 %) of frailty were identified among those who were pre-frail at baseline (n = 1016). Compared to group I, group II showed an increased risk of deterioration in frailty status after adjustments with multivariables including social isolation and malnutrition (adjusted odds ratio [aOR] = 1.61, 95 % confidence interval [CI]: 1.03-2.50). However, the association disappeared after further adjustment for depression. When we examined the longitudinal association between changes in eating with others or alone and changes in each frailty domain, group II showed an increased risk for the weight loss (aOR = 3.07, 95 % CI: 1.39-6.76) compared to group I. Group IV showed an increased risk for the weight loss (aOR = 2.39, 95 % CI: 0.95-6.00) and weakness (aOR = 2.07, 95 % CI: 1.16-3.68). Conclusions: A change from eating with others to eating alone was found to significantly increase the risk of deterioration in frailty status in elderly people, and the association seemed to be mediated by depression. These findings suggest that interventions to maintain eating partners and manage depression are needed to prevent frailty progression in elderly people.