A Recursive Bayesian Approach to Estimation of Prevalence of High Blood Pressure Among Different Age Groups

Abstract

High blood pressure (or hypertension) is a major public health issue affecting old aged adults in many countries and is a major risk factor in the development of stroke, cardiovascular and chronic kidney disease. High blood pressure of recent is becoming an important area of research due to its high prevalence among lower aged groups (i.e. below 30 years old). Many studies have been conducted on prevalence of high blood pressures amongst adults at local and national levels, and at urban or rural areas and all pointing to the fact that there has been increasing prevalence of hypertension across the globe. Since blood pressure tends to rise with age, there is the need to investigate the prevalence of high blood pressures at different age groups in order to describe the process which generates a particular signal or set of observations. In this study, staff data from the University of Lagos medical Centre were used for the research. A recursive Bayesian approach to dynamic state space estimation was developed to model the prevalence of high blood pressure together with the use of analytic solutions based on the Kalman filter. Based on model diagnostic criteria adopted, the result generated a best fit autoregressive model for the number patients with hypertension. The Kalman filter provided an optimal estimate of the linear state space approach to modelling dynamic system. The state space approach to modelling dynamic system in this study focused on discrete time formulation by using different equations to model the evolution of the system with time, and measurements assumed to be available at discrete time. It provided a generic and flexible framework for modelling the prevalence of high blood pressure.