Joint Modeling of Longitudinal Systolic and Diastolic Blood Pressure Measurements of Hypertensive Patients Receiving Treatment

Abstract

Hypertension is a chronic disease that has a major health problem over the
centuries due to its signicant contribution to the global health burden. It
is also called high blood pressure, described by two numbers Systolic blood
pressure (SBP) and diastolic blood pressure (DBP). Hence, joint longitudinal
model was used to address how the evolution of SBP is associated with the
evolution of DBP. The objective was to investigate the joint evolution and
association of SBP and DBP measurements of hypertensive patients and
identify the potential risk factors aecting the two end points. In this this
study 354 hypertensive patients with age greater than or equal to 18 years,
who were on treatment, and who had measured at least three times were
included. For a close examination of the separate and joint models, rst, each
of the outcomes was analyzed separately using linear mixed model. Then,
a joint model was considered to study the joint evolution and identify the
potential risk factors aecting the two end points. Fit statistics showed that
the joint model resulted in better t to the data than the separate models.
Based on the joint model, sex, baseline age, and place of residence were the
signicant factors for the progression of blood pressure, while family history
and all the interaction term except age by time did not appear signicant.
The result from the joint model suggested a strong association between the
evolutions and a slowly increasing evolution of the association between SBP
and DBP.