Development and validation of a nomogram to predict frailty risk in Chinese older adults combined with physical fitness test indicators.

Abstract

This study aimed to develop and validate a nomogram combined with the indicators of the physical fitness test to predict frailty risk in Chinese older adults. We recruited 344 participants from a community in Hebei Province, China. Data were collected on 57 candidate factor variables from sociodemographic factors, lifestyle factors, clinical factors, body composition test, and physical fitness test. Ultimately 6 factor variables were included in this predictive model: age, nutritional risk, hypertension, multimorbidity, depression and 2-Minute step test. The area under the curve (AUC) value in the training set and validation set is 0.866 and 0.854, which indicates that the model has a good ability to discriminate. The results of the H-L test indicate that the model is well calibrated. The calibration curves also indicate a good model fit. The model provides older adults with risk indicators to identify and prevent the onset of frailty as early as possible.