The Use of Tobit and Logistic Regression Models to Study Factors that Affect Blood Pressure in Cardiac Patients

Abstract

This research witnessed using the Tobit and logistic regression models to study the factors that affect blood pressure in cardiac patients. The data have been collected, from 500 patients with heart disease in hospital-heart center - Erbil. The two levels of blood pressure, low and high blood pressures, were taken from the patients, as dependent variables and some, independent variables (Gender, Age, Urea, Cholesterol, Creatinine, and Weight). The researcher found that the median of blood pressure by means of arterial pressure (MAP) equation contains each of high and low blood pressures differently because the threshold point was determined to be (99.33), which is equal to,12/8 mmHg, which is the normal amount of blood pressure. The aim of this research is to explain the main concepts and processes of Tobit regression analysis (Censored- and Truncated) and logistic regression analysis, which is used for knowing which factors of independent variables has more effect on the response variables (blood pressure), and to compare the outcomes of the two models (Tobit regression and logistic regression) in order to determine which of the models best fits our data in which AIC and BIC are used. The researcher reached a conclusion that the logistic regression model best fits our data compared with Tobit regression, and it arrived at the results obtained by utilizing the statistical packages in R programming, MATLAB and Statistical Packages of Social Sciences (SPSS) V.26, which was used to analyze data from our research