Abstract
OBJECTIVES: This study aimed to explore the bidirectional association between frailty and social relationships in older adults while distinguishing between interpersonal and intrapersonal effects. METHODS: A prospective cohort study of community-dwelling older adults was conducted in Japan in three waves spanning six years with follow-ups in every three years. Random intercept cross-lagged panel model was used to explore temporal associations between frailty and social relationships. RESULTS: Data for 520 participants (mean age 73.02 [SD 6.38] years, 56.7% women) were analyzed. Across individuals, frailty was associated with social relationships (β = -0.514, p < 0.001). At the interpersonal level, frailty was cross-sectionally associated with social relationships separately at T1(β = -0.389, p < 0.01), T2 (β = -0.343, p < 0.001) and T3 (β = -0.273, p < 0.05). Moreover, social relationships were associated with subsequent increases in symptoms of frailty in all measurement waves (β = -0.332, p < 0.001; β = -0.169, p < 0.01) and vice versa (β = -0.149, p < 0.05; β = -0.292, p < 0.001). CONCLUSIONS: The results suggest that frailty was associated with lower levels of social relationships. Frailty improvement programs can be combined with interventions to enhance social relationships, which will be beneficial in preventing frailty. The results emphasize the importance of combining clinical treatments of frailty with interventions to improve social relationships.