Trajectories and predictors of frailty in older patients undergoing radical prostatectomy: A longitudinal study

Yongcai Liu; Jieru Zhou; Yijuan Huang; Xin Yao; Xiaoyu Zhang; Jian Cai; Haihong Jiang; Wei Chen; Haiyan Li

doi:10.1016/j.apjon.2025.100741

Trajectories and predictors of frailty in older patients undergoing radical prostatectomy: A longitudinal study

Asia Pac J Oncol Nurs. 2025 Dec 6:12:100741. doi: 10.1016/j.apjon.2025.100741. eCollection 2025 Dec.

Authors

Yongcai Liu¹, Jieru Zhou¹, Yijuan Huang¹, Xin Yao¹, Xiaoyu Zhang², Jian Cai¹, Haihong Jiang¹, Wei Chen¹, Haiyan Li¹

Affiliations

¹ Department of Urology, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
² The Second School of Clinical Medical College, Zhejiang Chinese Medical University, Hangzhou, China.

Abstract

Objective: To identify the longitudinal heterogeneous trajectories of frailty in older patients with prostate cancer 6 months after radical prostatectomy and to explore the predictors of different trajectories.

Methods: A longitudinal design was conducted, and a total of 248 patients were recruited at the urology department in a tertiary-grade A hospital between June 2024 and March 2025. Data on predictive variables were collected at baseline using General information questionnaire, Geriatric Nutritional Risk Index, Health Literacy Management Scale, Kessler Psychological Distress Scale, and Medical Outcome Study Social Support Survey. Frailty assessments were subsequently performed at four time points using Tilburg Frailty Indicator: hospital discharge, and 1-, 3-, and 6 months postoperatively. Growth Mixture Modeling was employed to examine frailty trajectories over time. Multiple logistic regression and decision tree models were used to explore the predictors of heterogeneous trajectories of frailty.

Results: Three distinct trajectories of frailty were identified in old patients with prostate cancer: frailty rapid improvement group (54.4%), frailty progressive deterioration group (14.5%), and frailty persistent high group (31.1%). Multiple logistic regression analysis identified health literacy, nutritional status, psychological distress, and the presence of two or more comorbidities as independent predictors of frailty trajectories. The decision tree model further highlighted health literacy as the most influential predictor, followed by psychological distress, nutritional status, and social support.

Conclusions: Frailty trajectories in older prostate cancer patients exhibit substantial heterogeneity. These findings provide insights into the predictors of frailty in the process of diagnosis and treatment, and can be used in the clinical identification and monitoring of high-risk patients. Health care providers should develop personalized tailored to these predictive variables to mitigate frailty progression.

Keywords: Decision tree; Frailty; Prostate cancer; Trajectory.