Predictors of psychological resilience trajectories in patients with knee arthroplasty: A longitudinal study

Abstract

Aims: To identify the different classes of total knee arthroplasty patients according to the heterogenous trajectories of psychological resilience and investigate the predictors for different patterns of resilience.

Design: A prospective cohort study.

Methods: A total of 210 patients with total knee arthroplasty from March to December 2021 were included. Baseline assessment (T0) data were collected before surgery and included demographic, biological (clinical characteristics), psychological (psychological resilience, self-efficacy, psychological distress, hope, medical coping mode) and social (social support) factors. Resilience measurements were repeated at 3 days after surgery (T1), the date of discharge (T2), and 1 month (T3) and 3 months (T4) after discharge. Latent growth mixture modelling was employed to define different resilience trajectories. Predictors of class membership were identified using multinomial logistic regression.

Results: Data from 198 patients were analysed. Three latent classes were identified with similar patterns in different intercepts, showing a significant decrease in resilience from admission (T0) to 3 days after surgery (T1) followed by an increase from T1 to T4. The three trajectories of psychological resilience were named the stable-resilience class (65.66%), high-resilience class (17.68%), and low-resilience class (16.66%). Multinomial logistic regression showed that compared with the stable-resilience class, the high-resilience class was predicted by having a higher level of hope, having higher education, living in urban areas and having more children, while the low-resilience class was predicted by having lower levels of self-efficacy and hope, living in semirural areas, and having more children.

Conclusions: The three trajectories indicated that surgery was the major stressor influencing patients' psychological resilience and that patients in the low-resilience class needed to be intervened.

Impact: Predictors of patients in different classes provide evidence for the identification of vulnerable populations and lay a foundation for future research contributing to the development of targeted interventions for improving patients' psychological resilience. No patient or public contribution but the time points of investigation were decided based on our interviews with 12 total knee arthroplasty patients.

Keywords: knee arthroplasty; latent class growth modelling; longitudinal studies; nursing; predictors; psychological resilience; trajectory.