Development of a Binary Logistic Regression Model for Predicting Hypertension Risk Using Anthropometric and Lifestyle Variables

The purpose of the present study was to develop a binary logistic regression model for predicting hypertension risk using selected anthropometric and lifestyle variables among adults. The dependent variable of the study was hypertension status, categorized as hypertensive and non-hypertensive individuals. The independent variables selected for the study included anthropometric variables such as Body Mass Index (BMI), waist circumference, hip circumference, waist-hip ratio, and body fat percentage, along with lifestyle variables including physical activity level, smoking habit, alcohol consumption, and sleep duration. A total of 200 adults aged between 30 and 60 years were selected as participants for the study using purposive sampling technique. Anthropometric measurements were obtained using standardized procedures, while lifestyle-related information was collected through structured questionnaires. Blood pressure measurements were recorded to classify participants into hypertensive and non-hypertensive groups according to standard clinical guidelines. The collected data were analysed using descriptive statistics and binary logistic regression analysis. Odds ratios and confidence intervals were calculated to identify significant predictors of hypertension risk. The findings of the study revealed that BMI, waist circumference, body fat percentage, smoking habit, alcohol consumption, and reduced physical activity level were significant predictors of hypertension. The developed logistic regression model effectively predicted hypertension risk among adults. The study concluded that both anthropometric and lifestyle variables play an important role in the prediction of hypertension risk and that binary logistic regression modeling can be effectively used as a statistical approach for identifying individuals at greater risk of hypertension.