Table of Contents

ABSTRACT

Hypertension is becoming a global public health issue leading to a global burden of morbidity and mortality worldwide and in sub-Saharan Africa, South Africa is one of the countries experiencing a dramatic increase in the burden of hypertension. The ultimate goal of this study was to develop a reasonably robust model meant to establish the risk factors of hypertension amongst adults in South Africa attributable to extreme systolic and diastolic blood pressure values. This was a cross-sectional community-based study using data acquired from a survey distributed to patients visiting or referred to a public hospital in Pretoria, South Africa. The high prevalence of hypertension in South African adults (17.1% to 25.9%) was found to be predicted by age, BMI, gender, highest education level, race, physical exercises, family history of hypertension, depression, diabetes, stress and marital status after fitting the Bayesian binary quantile regression models. Convergence was achieved using the Gelman-Rubin diagnostic as indicated by the estimated potential scale reduction factors of less than 1.1 for both systolic blood pressure and diastolic blood pressure risk factors suggesting the accuracy of the Bayesian binary posterior estimated upper quantiles and the associated 95% credible intervals. Aggressive collaborative efforts between the general public, health authorities, non-governmental organisations, private sector and the government meant to minimise lifestyle causes of hypertension are a priority.

Keywords: Hypertension, Bayesian binary quantile regression, South Africa