Clustering metabolic and behavioral risk factors for hypertension: evidence from a longitudinal cohort in western Iran

Hypertension (HTN) is a major modifiable risk factor for non-communicable diseases, shaped by both metabolic and behavioral determinants. This study investigated the combined impact of key lifestyle and clinical variables on HTN incidence using a clustering approach within the Ravansar Non-Communicable Disease (RaNCD) Cohort. A total of 8082 adults aged 35-65 years from the RaNCD follow-up phase were included. Cluster analysis was performed using four variables body mass index (BMI), dyslipidemia, dietary inflammatory index (DII), and smoking status to identify distinct risk profiles. Associations between clusters and HTN incidence were assessed primarily using logistic regression. Sensitivity analyses were conducted by age and sex. Nelson-Aalen cumulative hazard estimates were calculated. Over a median follow-up of 9.3 (IQR 8.6-9.8) years, totaling 70977.024 person-years 1171 new cases of HTN were identified corresponding to an incidence rate of 16.50 per 1000 person- years. Three clusters emerged: Cluster 1 (Low Risk), Cluster 2 (Smokers), and Cluster 3 (Dyslipidemia-Overweight). Adjusted models revealed significantly higher odds of HTN in Cluster 3 (OR = 1.17; 95% ci 1.02-1.35) and Cluster 2 (OR = 1.27; 95% CI 1.01-1.59) compared to Cluster 1. Age-stratified sensitivity analysis confirmed the stability of these associations, while gender-stratified results showed significance only among men. Individuals with clustered metabolic and behavioral risk factors particularly smoking, overweight, and dyslipidemia are at elevated risk for developing HTN whereas DII distribution did not meaningfully differentiate risk across clusters. These findings support the need for targeted screening and prevention strategies focused on high-risk phenotypes identified through multidimensional profiling.