Assessment of Ten Insulin Resistance Surrogate Indexes Predicts New-Onset Cardiovascular Disease Incidence in Patients with Prediabetes or Diabetes: Insights from CHARLS Data with Machine Learning Ana
3 hours ago
- #Insulin Resistance
- #Machine Learning
- #Cardiovascular Disease
- Study evaluates ten insulin resistance (IR) surrogate indexes for predicting new-onset cardiovascular disease (CVD) in Chinese patients with prediabetes or diabetes.
- Longitudinal analysis of 3,532 participants from CHARLS data, with 24.7% developing CVD during follow-up.
- eGDR (estimated glucose disposal rate) associated with reduced CVD risk, while CVAI (Chinese visceral adiposity index) linked to increased risk.
- Highest eGDR quartile showed 47.3% lower CVD risk; highest CVAI quartile had 33.1% higher risk.
- Machine learning models, especially KNN, improved CVD prediction when incorporating eGDR and CVAI (AUC = 0.936).
- eGDR and CVAI identified as the most effective IR indexes for CVD risk stratification in this population.