Analysis of intervariable relationships between major risk factors in the development of coronary artery disease: a classification tree approach.

OBJECTIVE

The purpose of this study is to determine how the major risk factors are related to each other in the development of coronary artery disease (CAD) using Chi-squared Automatic Interaction Detection (CHAID).

METHODS

All patients with suspected CAD seen in the cardiology clinic between January 1999 and February 2003 who underwent coronary angiography were included in the study. A retrospective analysis was performed in 1381 patients. In all patients' sex, age, type II diabetes mellitus, hypercholesterolemia, systemic hypertension, smoking status, family history of CAD, body mass index (BMI) were assessed.

RESULTS

According to classification tree, first-level split produced the two initial branches: female (unadjusted presence percentage = 48.07%) versus male (unadjusted presence percentage = 78.02%). For the male aged between 49-81 years and the female aged between 15-48, 49-60 and 61-71 years, diabetes mellitus was the most prominent risk factor. However, hypercholesterolemia was the best predicting variable for the females aged between 72-81 years. For the females of 15-48 years and 49-60 years age categories without diabetes mellitus, smoking status and family history of CAD had important contribution to the model.

CONCLUSION

Sorting the major risk factors of CAD from the most to least according to the classification importance was resulted as sex, age, diabetes mellitus, hypercholesterolemia, family history of CAD and smoking status.