Construction and validation of a predictive model for reversible cognitive frailty in elderly people in nursing homes

To investigate the factors affecting reversible cognitive frailty in older adults in nursing homes, to construct a risk prediction model, and to validate its prediction performance. Older adults from five nursing homes in Liaoning Province, China, were selected as the study subjects from April to October 2024, and the data were randomly divided into a training set and a validation set according to 8:2. Logistic regression analysis was used to analyze the factors affecting reversible cognitive frailty in older adults in nursing homes, and a column-line graph model was constructed using R software (4.4.1). A total of 311 cases of older adults were included, of which 86 cases (27.7%) had reversible cognitive frailty. The AUCs of the training and validation sets were 0.852 and 0.805, respectively. The Hosmer-Lemeshow goodness-of-fit test values were P = 0.513 and 0.796 (both > 0.05). The risk prediction model constructed in this study has good predictive performance, which is beneficial for institutional workers in providing early warning of the risk of reversible cognitive frailty for older adults and intervention. 1. A literature search has shown that there is no research on the construction of an RCF prediction model for older adults in nursing homes. 2. RCF is a significant predictor of all-cause mortality and the occurrence of dementia in older adults. 3. Older adults in nursing homes, as a special group, are different from older adults in the community in terms of their living environment, social support, psychological status, and activity participation.