CLUSTERING OF BEHAVIORAL RISK FACTORS AND UNDIAGNOSED HYPERTENSION AMONG ADULTS IN SOUTHERN AFRICA ZAMBIA: A SECONDARY ANALYSIS OF WHO STEPS SURVEY DATA

Abstract

This study examines the epidemiological configurations of cardiovascular risk in Zambia by evaluating the clustering patterns of modifiable behavioral risk factors and their direct relationship with undiagnosed hypertension. Performing a secondary cross-sectional analysis on a nationally representative sample of 4,813 adults aged 18 to 69 from the WHO STEPwise Approach to Non-Communicable Disease (NCD) Risk Factor Surveillance registry, the research utilizes Latent Class Analysis (LCA) alongside multivariable logistic regression modeling to map hidden behavioral typologies within the population. The statistical analysis reveals a severe public health crisis, identifying an overall hypertension prevalence of 29.4%, with a striking 63.1% of these individuals completely unaware of their condition. Physical inactivity (34.6%) and unhealthy diets low in fruit and vegetable intake (31.7%) emerged as the most prevalent individual behavioral liabilities. Through LCA modeling, the study identifies four distinct behavioral classes: Low Risk (31.9%), Sedentary-Diet Risk (26.9%), Alcohol-Tobacco Risk (23.8%), and Multiple Risk (17.3%). Multivariable models demonstrate a robust, independent, and graded association between risk accumulation and non-diagnosis. Compared to the low-risk baseline, adjusted odds ratios for undiagnosed hypertension scale from 1.64 in the Sedentary-Diet class to 2.73 in the Multiple Risk class. Demographic stratification shows that the Sedentary-Diet class is heavily populated by urban females, while the Alcohol-Tobacco class is predominantly rural and male. Population Attributable Fraction (PAF) estimations suggest that migrating high-risk groups to a low-risk behavior matrix could prevent 48.3% of the nation's undiagnosed hypertension cases, validating calls for targeted, culturally responsive community screening.