Associations of loneliness, multidimensional self-care, and widowhood with health-related quality of life among older adults in Greece

Abstract Background Loneliness is consistently associated with poorer health-related quality of life in older adulthood. However, marital status alone may not adequately represent the subjective and everyday processes associated with health, and the role of multidimensional self-care in this context remains insufficiently understood. Objective This study examined the associations of loneliness and multidimensional self-care with mental and physical health-related quality of life among community-dwelling older adults and assessed whether self-care statistically accounted for part of these cross-sectional associations beyond widowed status and other relevant characteristics. Methods A cross-sectional observational study was conducted among community-dwelling older adults recruited across Greece between July 2024 and July 2025 through community centres, primary healthcare units, municipal social services, and local associations. Data on the principal study variables were available for 608 participants. Age data were available for 607 participants, who were aged 62–99 years (M = 77.61, SD = 8.53), and 63.5% of the sample were women. Fully adjusted regression and indirect-effect models included 607 complete cases. Participants completed the De Jong Gierveld Loneliness Scale, the Multidimensional Self-Care Scale for Older Adults, and the SF-36. Pearson correlations, independent-samples t tests, hierarchical multiple regressions, and cross-sectional indirect-effect analyses with 10,000 bootstrap resamples were performed. Post hoc sensitivity analyses were also conducted using outcome-specific self-care scores that excluded the domains most directly overlapping with the respective health outcomes. Results Greater loneliness was associated with poorer aggregated mental ( r = − .57) and physical health ( r = − .50), whereas higher multidimensional self-care was associated with better mental ( r = .69) and physical health ( r = .70; all p < .001). The final regression models explained 52.7% of the variance in mental health and 54.7% of the variance in physical health. In the fully adjusted models, self-care showed the strongest independent association with both mental (β = 0.494, p < .001) and physical health (β = 0.511, p < .001). Loneliness remained negatively associated with mental (β = −0.216, p < .001) and physical health (β = −0.085, p = .025), whereas widowed status was not independently associated with either outcome. Statistically supported indirect associations through self-care were observed for mental health (B = − 6.91, 95% bootstrap CI [− 8.33, − 5.59]) and physical health (B = − 9.14, 95% bootstrap CI [− 10.94, − 7.38]). In post hoc sensitivity analyses, the self-care associations were attenuated but remained statistically significant after the most directly overlapping domains were excluded (mental health: β = 0.311; physical health: β = 0.310; both p < .001), and the corresponding indirect associations remained statistically supported. Conclusions Loneliness and multidimensional self-care were independently associated with mental and physical health-related quality of life, whereas widowed status did not retain an independent association after adjustment. Self-care statistically accounted for part of the contemporaneous associations between loneliness and health. Sensitivity analyses indicated that conceptual overlap contributed to the magnitude of these associations but did not fully account for them. Because the data were cross-sectional, these findings do not establish temporal ordering or causal mediation. They nevertheless support the joint assessment of subjective loneliness and multidimensional self-care rather than reliance on marital status alone when examining health in later life.