LAUSR

OBJECTIVES Previous literature has reported that loneliness is the strongest predictor of frailty, which is also closely associated with activity engagement. Yet, to date, the path from loneliness to frailty has not been investigated. In this study, we evaluated whether activity engagement mediates the relationship between loneliness and frailty among older adults living in nursing homes. DESIGN We employed the use of a cross-sectional descriptive survey. SETTING AND PARTICIPANTS Residents (N = 370; aged ≥60 years; 220 female, 150 male) were from 33 nursing homes in China. MEASURES Information was collected regarding loneliness, activity engagement, frailty, sociodemographic characteristics, nursing home characteristics, self-rated health, insomnia, and depression. The mediation analyses, comprising regression and bootstrap analyses, were performed to test both direct and indirect effects of loneliness on frailty (ie, the mediating role of activity engagement). RESULTS The prevalence of frailty was 29.2% among Chinese older adults living in nursing homes. Activity engagement mediated the association between loneliness and frailty [95% confidence interval (CI) = 0.011, 0.070]; however, loneliness was not significantly related to frailty when covarying for activity engagement (95% CI = 0.028, 0.190). CONCLUSION AND IMPLICATIONS Results suggest that it is not loneliness in general that affects frailty but actively engaging in activities. Therefore, preventing frailty by employing interventions that promote older adults' activity engagement may improve nursing home residents' quality of life. The main implication is that activity engagement should be an effective indicator for care planning and quality evaluation in nursing homes.