G-3795

Latent Class Analysis of Risk Factors in hypertensive patients: result from baseline data of Fasa Adult Cohort Study

 Alireza Khajepour ¹ ©, Mobina Vatankhah ¹ ℗, Azam Namdar ², Mehdi Sharafi ³, Niloofar Choobin ⁴   

 
Abstract

Introduction: Hypertension is a complex, multifactorial health condition influenced by both behavioral and metabolic risk factors. While conventional approaches often evaluate these variables individually, they fail to capture the synergistic effects of risk factor clustering. This study aimed to identify distinct latent profiles of combined risk factors among hypertensive adults using baseline data from the Fasa Adult Cohort Study. Methods and Materials: This cross-sectional analysis included 2,026 hypertensive adults aged 35–92 years (mean age for men: 56.78 ± 8.93; women: 54.66 ± 8.89). Demographic, behavioral, and clinical variables—including BMI, waist-to-hip ratio (WHR), lipid profile, diabetes status, TyG index, smoking, and physical activity—were assessed. Latent class analysis (LCA) was performed to classify individuals into unobserved subgroups based on shared risk characteristics. Competing models from 1 to 6 classes were evaluated using AIC, BIC, and entropy values, with the most clinically meaningful model selected. Results: Significant sex-based differences were found in sociodemographic and clinical parameters, such as physical activity, drug use, LDL, WHR, and diabetes (P 0.05). LCA results indicated that a three-class model provided the best fit and interpretability (entropy = 7.087). Class 1 (48.8%) – Obesity dominant: Characterized by high BMI (≥25 in 63.8%) and elevated WHR (91.3%), but with relatively low prevalence of other metabolic or cardiovascular risks. Class 2 (22.0%) – Metabolic-risk obesity: Defined by severe metabolic dysfunction including elevated TyG index (99.9%), high TG/HDL ratio (92.2%), hypertriglyceridemia (53.8%), and obesity (BMI ≥25 in 80.8%). Class 3 (29.0%) – Cardiovascular and obesity cluster: Comprised of individuals with nearly universal prevalence of cardiac disease and CVD history (99.9%), in addition to high rates of obesity (BMI ≥25 in 69.0%; abnormal WHR in 94.9%). These findings suggest that hypertensive individuals exhibit heterogeneous risk profiles, with obesity as a common factor across all classes. Conclusion and Discussion: This study underscores the utility of latent class analysis in uncovering hidden subgroups within hypertensive populations. The identification of distinct phenotypes—especially those combining obesity with either metabolic dysregulation or cardiovascular comorbidities—offers insights for more tailored and effective prevention and management strategies. Early detection and stratification based on clustered risk patterns can play a crucial role in controlling the burden of hypertension and related chronic conditions.

Keywords: Hypertension, Latent Class Analysis, Obesity, Cardiovascular Risk, Metabolic Syndrome