A nomogram to predict the risk of sarcopenia in older people

Abstract

The burden of sarcopenia is increasing worldwide. However, most cases of sarcopenia are undiagnosed due to the lack of simple screening tools. This study aimed to develop and validate an individualized and simple nomogram for predicting sarcopenia in older adults. A total of 180 medical examination populations aged ≥60 years were enrolled in this study. Sarcopenia was diagnosed according to the Asian Working Group for Sarcopenia 2019 consensus. The primary data were randomly divided into training and validation sets. Univariate logistic regression analysis was performed to select the risk factors of sarcopenia, which were subjected to the least absolute shrinkage and selection operator for feature selection. A nomogram was established using multivariate logistic regression analysis by incorporating the features selected in the least absolute shrinkage and selection operator regression model. The discrimination and calibration of the predictive model were verified by the concordance index, receiver operating characteristic curve, and calibration curve. In this study, 55 cases of sarcopenia were available. Risk predictors included age, albumin, blood urea nitrogen, grip strength, and calf circumference. The model had good discrimination and calibration capabilities. concordance index was 0.92 (95% confidence interval: 0.84–1.00), and the area under the receiver operating characteristic curve was 0.92 (95% confidence interval: 0.83–1.00) in the validation set. The Hosmer-Lemeshow test had a P value of .94. The predictive model in this study will be a clinically useful tool for predicting the risk of sarcopenia, and it will facilitate earlier detection and therapeutic intervention for sarcopenia.

Machine Learning Applications in Sarcopenia Detection and Management: A Comprehensive Survey

“…Logistic regression has emerged as a powerful statistical technique for predicting sarcopenia risk and onset amongst aging populations. By incorporating multiple predictors like age, sex, body composition, and physical activity, logistic models can estimate the individual likelihood of developing sarcopenia [ 100 , 101 , 102 ]. These predictive models enable more targeted screening and early interventions.…”

“…With 80% sensitivity and 70% specificity, this model shows promise as a practical screening tool. Expanding on this, Yin et al [ 102 ] combined logistic regression and nomogram visualization to accurately predict individual sarcopenia risk, highlighting clinical utility.…”