Integration of Bivariate Logistic Regression Models and Decision Trees to Explore the Relationship between Socio-Demographic and Anthropometric Factors with the Incidence of Hypertension and Diabetes in Prospective Athletes

Hypertension and diabetes are two medical conditions that are often associated with athletes’ health. Hypertension or high blood pressure is a condition where the blood pressure in the arteries becomes too high. Meanwhile, diabetes is a condition where the body cannot produce or use insulin properly, thereby causing high blood sugar levels. Athletes’ health is very important because they need optimal physical conditions to be able to compete effectively. Hypertension and diabetes can affect athletes’ health and their performance. Socio-demographic and anthropometric factors are believed to play an important role in the development of both conditions. The aim of this study is to determine the relationship between socio-demographic and anthropometric factors on the incidence of hypertension and diabetes in prospective athletes in athletics and determine whether prospective athletes pass the initial screening process. This study integrates bivariate logistic regression models and decision trees to analyze data collected from 200 athlete selection participants. The univariate logistic regression model showed that waist circumference, father’s occupation, and salary category 2 had a significant influence on hypertension, while BMI had a significant influence on diabetes. Meanwhile, the bivariate logistic regression model found that BMI and salary category 2 had a significant effect on hypertension. The optimal classification tree was formed using variables such as BMI, Salary Category 2, Hypertension, and Diabetes. The accuracy of the prediction data was 72%, indicating that the optimal tree is well-formed and suitable for classifying athletes’ data. This study concludes that there is a significant relationship between sociodemographic and anthropometric factors and the incidence of hypertension and diabetes in prospective athletes. This study provides valuable insight into physiological adaptation, fitness, recovery, and other factors that influence athlete performance.