A Hybrid Logistic Regression Model with Harris Hawks Optimization (HHO) Algorithm of Hypertension Determinants among Iraqi Adults

Hypertension is a major chronic disease worldwide and in Iraq. Baghdad especially suffers from it due to rapid urbanization, dietary changes, and lifestyle shifts. We lack local data on hypertension factors in Baghdad compared to global studies. This cross-sectional study examined multiple risk factors in 1,050 adults (ages 18-70) in Baghdad during 2023-2024. We used multi-stage stratified sampling. Logistic regression usingthe Harris Hawks Optimization (HHO) algorithm was applied to select the most impactful variables. We studied 11 factors: age, gender, smoking, physical activity, BMI, cholesterol, salt intake, sleep quality, stress, education, and income. HHO was chosen because it handles high-dimensional data efficiently. Age (OR: 2.14) and obesity (BMI ≥30, OR: 3.26) emerged as the strongest predictors of hypertension in Baghdad. The hybrid model achieved 84.2% accuracy andan AUC of 0.87. Standard logistic regression hada lower AUC of 0.79. Age-targeted interventions are needed for hypertension control in Baghdad. Weight management programs are also essential. These results apply to other Middle Eastern cities facing similar epidemiological changes.