Joint analysis of blood pressure changes over time and survival of hypertensive patients under treatment at Sawula hospital in Southern Ethiopia

Abstract

Abstract Introduction Hypertension is among the most significant non-communicable public health issues worldwide. High blood pressure, or hypertension, has been associated with severe health consequences, including death, aneurysms, stroke, chronic renal disease, eye damage, heart attack, heart failure, peripheral artery disease, and vascular dementia. Consequently, this study aimed to investigate the predictors linked to survival time and the progression of blood pressure measurements in hypertensive patients. Methods This study on retrospective cohort analysis was conducted among hypertensive patients receiving follow-up treatment. A total of 200 hypertensive patients were screened during their follow-up at Sawula General Hospital. Multivariate joint models typically combine a multivariate linear mixed-effects model for repeated measurements with a Cox model for time-to-event outcomes. Result The baseline characteristics of patients indicated that, out of 200 hypertensive patients, the majority were male, totaling 116 (58%). Regarding alcohol consumption, 72 (36%) of the hypertensive patients reported using alcohol, while 128 (64%) did not. The multivariate joint model emerged as the most effective model in this study. The analysis revealed that observation time, economic status, sex, place of residence, baseline FBS, and age significantly influenced log FBS, while visiting time, age, sex, pulse rate, place of residence, and baseline DBP were identified as significant factors for log diastolic blood pressure. Conclusion The multivariate joint model exhibited superior performance. Hypertension continues to be a non-communicable disease and a public health issue in Sawula and Ethiopia, deserving greater attention.

Keywords

• Cox PH
• Hypertensive
• Longitudinal measurement
• Joint Model