Analysis Of Risk Factors And Construction Of Prediction Model For Sarcopenia Among Elderly People In Community

ObjectiveTo explore the incidence rate and hazard factors of sarcopenia in Chinese community elderly,build and verify the danger forewarning model of sarcopenia in community elderly,and develop a wechat mini program for practical application to predict the risk of sarcopenia.In order to quickly predict the occurrence probability of the elderly with sarcopenia according to the wechat mini program,provide reference for scientific prevention,treatment and efficient management of high-risk groups with sarcopenia.MethodsThis study gathered the physical examination information of 626 sample over 60 years old during community health examination in Dali City,Yunnan Province from December 2021 to July 2022.On the basis of the sarcopenia diagnostic criteria(Asian sarcopenia Working Group 2019),626 elderly people were screened to identify the status quo of sarcopenia among the elderly in the community.According to the ratio of construction model to verification model is 7:3,the sample was allotted into modeling group and authentication group at random.Logistic regression was applied to analyze the hazard factors of sarcopenia among the elderly in the modeling group,and constructed the risk forewarning model.H-L test,Calibration curve and ROC were used to judge the forewarning ability of the model.Youden index(YI)was used to evaluate the clinical usefulness of the risk prediction model for sarcopenia.The data of the verification group were substituted into the forewarning model to verify and evaluate the established danger forewarning model of community elderly people with sarcopenia.In order to facilitate community workers to screen sarcopenia in a timely and efficient manner,a wechat mini program was exploited to forecast the risk of sarcopenia.Result1.Basic Information626 elderly people in the community were included,and the total incidence of sarcopenia was 15.8%.In the modeling group of 438 cases,the incidence of sarcopenia was 16.2%.In the verification set of 188 cases,the rate of sarcopenia was 14.9%.2.Construct and verify the prediction model of community elderly people with sarcopenia(1)Binary Logistic regression analysis revealed that 6 affecting elements were closely related to sarcopenia of elderly people in the community: BMI(OR=0.496,95%CI=0.364-0.677),gender(OR=0.149,95%CI=0.033-0.671),frequency of physical exercise(occasionally)(OR=13.902,95%CI=2.538-76.155),frequency of physical exercise(more than once a week)(OR=13.722,95%CI=1.058-178.054),time of each exercise(OR=0.905,95%CI=0.849-0.964),albumin(OR=0.930,95%CI=0.866-0.999),total cholesterol(OR=0.348,95%CI=0.189-0.639).The formula of prediction model was established: Z=23.595-0.701×BMI-1.906×gender+2.632×frequency of physical exercise(occasionally)+2.619×frequency of physical exercise(more than once a week)-0.100×time of exercise-0.073×albumin-1.056×total cholesterol.H-L test(P=0.887)and Calibration curve indicate that the prediction model has good consistency.The AUC under ROC was 0.983(95%CI:0.971-0.995),sensitivity was 0.873,specificity was 0.992.It suggests that the model has better predictive power.(2)H-L test(P=0.927)and Calibration graph show a high consistency between the predicted probability and the actual probability.The AUC under ROC was 0.947(95%CI:0.894-0.998),the specificity was 0.944,the sensitivity was 0.929,suggesting that the prediction effect was in good agreement with the reality.3.Build a wechat mini program for model of community elderly people with sarcopeniaAccording to the Logistic regression model,a wechat mini program is exploited to forecast the risk of sarcopenia.The mini program can automatically calculate the risk probability of community elderly people with sarcopenia by inputting BMI,gender,frequency of physical exercise,time of each exercise,albumin and total cholesterol.The higher the predictive value,the higher the risk of sarcopenia.ConclusionLow BMI,female,low frequency of physical exercise,less time for each exercise,low albumin and low total cholesterol were independent hazardous elements for sarcopenia.The wechat mini program constructed in this study further simplifies the prediction process of sarcopenia.It has a certain degree of popularization and application,and can quickly predict the occurrence probability of elderly people with sarcopenia,providing a scientific basis for the prevention and intervention of community elderly people with sarcopenia.