Prediction of home-based Traditional Chinese Medicine nursing technology demand of elderly patients with chronic diseases
Objective:To investigate the current situation of home-based Traditional Chinese Medicine(TCM)nursing technology needs of elderly patients with chronic diseases and to predict their needs,so as to provide reference for clinical decision-making.Methods:From February to June 2024,a cross-sectional survey was conducted on 1026 elderly patients with chronic diseases in outpatient and inpatient departments of 9 TCM hospitals in Zhejiang province by using the self-made questionnaire on the needs of home TCM nursing technology for elderly patients with chronic diseases.The extreme gradient lifting algorithm is used to predict the demand,and the importance of the variables in the model is explained by the Shapley Additive Explanations.Results:The demand rate of home-based TCM nursing technology for elderly patients with chronic diseases was 70.27%,of which scraping(52.63%),moxibustion(51.85%),cupping(48.73%),ear point sticking(47.37%)ranked in the forefront.Overall importance ranking,TCM nursing technology cognition and attitude,specific symptoms of discomfort,per capita monthly family income,education level and other characteristics were identified as significant predictors.The AUC value of the model in the training set is 0.95-0.99,and the AUC value in the test set is 0.80-0.89,and the performance of the model is good.Conclusion:The demand for home traditional TCM nursing technology in elderly patients with chronic diseases is extensive and affected by many factors.In the future,the promotion and publicity of TCM nursing technology should be strengthened to improve public awareness.At the same time,we will increase personnel training,provide corresponding TCM nursing technical support for common discomfort symptoms of patients,and meet the personalized needs of elderly patients with chronic diseases.
elderlychronic diseasehome-basedTCM nursing technologymachine learningpredicting model
NETL
NSTL
万方数据
