A MULTIFACTORIAL MODEL FOR PREDICTING SUBCLINICAL ATHEROSCLEROSIS IN MILITARY PERSONNEL: THE CONTRIBUTION OF CRP, HOMA-IR, AND CAVI

The study aimed to build and evaluate a multifactorial diagnostic model for the probability of donozological vascular wall changes in military men. The model included age, abdominal obesity, apoB/apoA1 ratio, HOMA-IR, CAVI, occupational-environmental factor, and, in the second version, CRP. Adding CRP improved discrimination, increasing AUC from 0.85 to 0.88. At a probability threshold of 0.27, the model showed an overall accuracy of 79.0% and a high negative predictive value of 93.8%. These findings suggest that in organized military screening, the most useful model is one that reliably rules out structural vascular lesions in men with a low predicted probability.

Перевод пока недоступен