@article{Ma2026, 
author = {Yanjun Ma and Chaofan Geng and Zhibin Wang and Yiwei Zhao and Lixin Ma and Pengpeng Ye and Leilei Duan and Guoping Peng and Yi Tang},
title = {A Prediction Model for One‐Year Disability Risk Among Community‐Dwelling Older Adults in China: Integrating Physical, Metabolic, and Psychosocial Factors},
year = {2026},
journal = {Health Care Science},
volume = {5},
number = {4},
pages = {320-329},
keywords = {community‐based cohort, disability risk prediction, emotion regulation, gait speed, LASSO regression, nomogram, older adults},
url = {https://www.sciopen.com/article/10.1002/hcs2.70080},
doi = {10.1002/hcs2.70080},
abstract = {BackgroundDisability in older adults, characterized by progressive limitations in performing activities of daily living, poses a significant public health challenge in aging societies. Early identification of individuals at risk is critical for implementing targeted interventions to mitigate functional decline. However, existing prediction models often prioritize disease‐related physiological indicators and overlook psychosocial dimensions, limiting their practicality and scalability in community‐based settings. This study aimed to develop and validate a practicable prediction model for 1‐year disability risk by integrating multidimensional indicators.MethodsA prospective community‐based cohort in Beijing, functionally independent at baseline, was followed for 1 year. Disability was defined as a decline in the Barthel Index score. Potential predictors included demographic, lifestyle, clinical, physical, and psychosocial measures. The least absolute shrinkage and selection operator (LASSO) regression was used for variable selection, followed by logistic regression to construct the final model. Model performance was evaluated through fivefold cross‐validation repeated 100 times, with assessment of discrimination and calibration.ResultsAmong 1003 community‐dwelling older adults (mean age 68.4 ± 5.2 years; 58.3% female), 163 (16.3%) developed disability at follow‐up. The LASSO regression identified seven predictors: age, dyslipidemia, gait speed, waist circumference, difficulty in lifting weights, appetite, and emotion regulation ability. The model demonstrated moderate discrimination, with an area under the receiver operating characteristic curve of 0.716 (95% CI: 0.712–0.720). Calibration curves indicated good overall agreement between predicted and observed risks. A nomogram was developed to facilitate individualized risk prediction in clinical practice.ConclusionsThis study presents a practical disability risk prediction model incorporating physical, metabolic, and psychosocial factors. The model exhibits acceptable discrimination and calibration, supporting its potential for early screening and stratified management of community‐dwelling older adults. Future multicenter validations are warranted to enhance generalizability and explore dynamic interventions targeting modifiable factors like emotion regulation and gait speed.}
}