Development and validation of a predictive model for the risk of possible sarcopenia in middle-aged and older adult diabetes mellitus in China.

BACKGROUND

People with diabetes mellitus (DM) have a significantly increased risk of sarcopenia. A cross-sectional analysis was performed using nationally representative data to evaluate possible sarcopenia in middle-aged and older adults with diabetes mellitus, and to develop and validate a prediction model suitable for possible sarcopenia in middle-aged and older adults with diabetes mellitus in the Chinese community.

METHODS

Data from the China Health and Retirement Longitudinal Study (CHARLS), which focuses on people 45 years of age or older, served as the basis for the prediction model. CHARLS 2015 participants were used in the study, which examined 53 factors. In order to guarantee model reliability, the study participants were split into two groups at random: 70% for training and 30% for validation. Ten-fold cross-validation and Least Absolute Shrinkage and Selection Operator (LASSO) regression analyses were used to determine the best predictors for the model. The factors associated with sarcopenia in DM were researched using logistic regression models. Nomogram were constructed to develop the predictive model. The performance of the model was assessed using area under the curve (), calibration curves and decision curve analysis ().

RESULTS

A total of 2,131 participants from the CHARLS database collected in 2015 passed the final analysis, and the prevalence of sarcopenia was 28.9% (616/2131). Eight factors were subsequently chosen as predictive models by LASSO logistic regression: age, residence, body mass index, diastolic blood pressure, cognitive function, activities of daily living, peak expiratory flow and hemoglobin. These factors were used in the nomogram predictive model, which showed good accuracy and agreement. The values for the training and validation sets were 0.867 (95%CI: 0.8470.887) and 0.849 (95%CI: 0.8160.883). Calibration curves and DCA indicated that the nomogram model exhibited good predictive performance.

CONCLUSION

The nomogram predictive model constructed in this study can be used to evaluate the probability of sarcopenia in middle-aged and older adult DM, which is helpful for early identification and intervention of high-risk groups.