Predictive Modelling of Hypertension Risk Using Lifestyle and Clinical Biomarkers

Hypertension remains one of the most widespread chronic conditions globally, yet it is often undiagnosed because it typically develops without noticeable symptoms. Many individuals only become aware of their condition after severe complications such as cardiovascular or kidney-related events occur. This study explores the use of supervised machine learning models to identify individuals at risk of hypertension based on demographic, clinical, and lifestyle factors. Two algorithms were implemented: Logistic Regression, chosen for its interpretability, and XGBoost, selected for its ability to capture complex nonlinear relationships. Both models were trained using an 80/20 stratified split and evaluated using accuracy, precision, recall, F1-score, and AUC-ROC. Results showed XGBoost consistently outperformed Logistic Regression, particularly in recall. Feature importance analysis revealed that modifiable lifestyle factors such as BMI, physical activity, diet, and smoking significantly influence prediction outcomes.