LAUSR

Objectives Living arrangement has been reported to have a significant influence on feelings of loneliness in older adults, but their living preferences may confound the association. This study aimed to investigate whether the associations of living arrangements with loneliness differ in community-dwelling older adults according to different living preferences. Methods In the 2008/2009 (baseline) and 2011/2012 (follow-up) waves of the Chinese Longitudinal Healthy Longevity Survey, living arrangements [living with children mainly (LWC), living with spouse only (LWS), and living alone (LA)], living preferences [preferring living with children (PreLWC) and preferring living alone/only with spouse (PreLA)], and feelings of loneliness were assessed. The effect modifications of living preferences in the associations of living arrangements with loneliness were estimated using logistic regression models, and corresponding odds ratios (ORs) were calculated. Results Living preferences significantly modified the associations of living arrangements with loneliness at baseline (p for interaction = 0.009 for LWS and = 0.015 for LA). Compared with LWC, LWS was protective for loneliness only in the PreLA older adults at baseline (OR = 0.53, 95% CI = 0.45–0.64, p < 0.001), and LA was significantly associated with loneliness especially in the PreLWC older adults, compared with their PreLA counterparts (at baseline, ORs = 2.89 vs. 2.15; at follow-up, ORs = 1.68 vs. 1.51). Conclusion Living preference modifies the associations of living arrangements with loneliness, and those who prefer living with children but live alone are more likely to feel lonely. It is recommended that living preferences should be considered when managing loneliness in community-dwelling older adults.