Risk factor analysis of hypertension with logistic regression and Classification and Regression Tree (CART)

Abstract

Hypertension is one of the most common inherited diseases among Indonesians. This disease can affect the onset of various diseases, such as kidney failure, stroke, diabetic, and heart failure. Early detection is an effective way to control the incidence of hypertension by knowing risk factors such as age, sex, family history, genetics (irreversible/controlled risk factors), smoking habits, alcohol consumption habits, obesity, lack of physical activity that have significant effect. The methods that used to analyzed significant risk factor are logistic regression and Classification and Regression Tree (CART). This research compared the accuracy of two methods to select the best models to predict the risk of Hypertension. From the result, CART better than logistic for predict hypertension risk with AUC of 0,584.