Associations among resilience, post-traumatic growth, and quality of life in older adults hospitalized with hip fracture: a cross-sectional mediation analysis
Abstract
Objective This study described post-traumatic growth (PTG), resilience (RI), and quality of life (QOL) in older adults hospitalized with hip fracture and examined their cross-sectional associations, including whether PTG statistically accounted for part of the association between RI and QOL. Methods A cross-sectional convenience sample of 308 adults aged 60–90 years with hip fracture was recruited from an orthopedic ward, and 273 valid questionnaires were analyzed (effective response rate, 88.6%). Participants completed the General Information Questionnaire, the 20-item Chinese Post-traumatic Growth Inventory, the Connor-Davidson Resilience Scale, and the Short-Form 12 Health Survey. Associations were evaluated using Spearman correlation, covariate-adjusted linear regression with heteroskedasticity-consistent HC3 standard errors, and a 5,000-resample bootstrap mediation analysis. Results Median (interquartile range) scores were 56 (49–60) for PTG, 71 (64–74) for RI, and 41.34 (37.90–44.22) for QOL. RI was positively correlated with PTG (rs=0.511) and QOL (rs=0.682), and PTG was positively correlated with QOL (rs=0.633; all p<0.001). In the fully adjusted QOL model, PTG (standardized β=0.302, p<0.001) and RI (standardized β=0.622, p<0.001) remained independently associated with QOL (R²=0.760; adjusted R²=0.745). The covariate-adjusted indirect association through PTG was 0.131 (bootstrap 95% CI, 0.086–0.175), representing 25.4% of the total RI–QOL association. Conclusion RI, PTG, and QOL were positively associated in this cross-sectional sample of older adults hospitalized with hip fracture. PTG statistically accounted for part of the RI–QOL association; however, the simultaneous measurement of all variables precludes causal or temporal interpretation. Longitudinal studies with detailed clinical covariates are required before intervention effects can be inferred.