Comparison of Different Regression Analyses for Identifying Risk Factors in Obese and Nonobese Patients With Coronary Artery Disease

Abstract

This study was designed to assess conventional and novel risk factors in obese and nonobese patients with coronary artery disease (CAD) by using multivariate forward and univariate logistic regression analysis and to find the best model of analysis for identifying these risk factors. The study group consisted of 398 patients who consecutively underwent coronary angiography for the investigation of chest pain, except overweight patients. In univariate logistic regression analysis, high C-reactive protein and cigarette smoking were found to be the strongest variables in obese and nonobese patients with CAD, respectively. In multivariate forward logistic regression analysis, some risk factors were not found as predictors of CAD. Multivariate forward logistic regression analysis with the advantage of a high predictable ratio may be more useful for the analysis of risk factors in patients with CAD.