JOINT MODELLING OF LONGITUDINAL CHANGE OF PULSE RATE AND TIME TO DEFAULT FROM TREATMENT DURING TREATMENT PERIOD OF CONGESTIVE HEART FAILURE PATIENTS IN CASE OF FELEGE HIWOT REFERAL HOSPITAL, BAHIR DAR, ETHIOPIA

BERHANIE, ADDIS

dc.contributor.author	BERHANIE, ADDIS
dc.date.accessioned	2020-09-08T11:36:07Z
dc.date.available	2020-09-08T11:36:07Z
dc.date.issued	2020-09-07
dc.identifier.uri	http://hdl.handle.net/123456789/11197
dc.description.abstract	ABSTRACT Background: Like all other organs the Heart is susceptible to disease. The main objective of this study was to joint modeling of longitudinal change of pulse rate and time-to default from treatment of congestive heart failure patients at Felege- Hiwot Referral Hospital, Bahir Dar, Ethiopia. Method: A retrospective cohort study design was conducted on congestive heart failure patients selected by inclusion and exclusion criteria at FHRH under a follow-up period from January 1st 2016 to December 31th 2019. Exploratory data analysis (profile) analysis was used to investigate various associations, structures and patterns exhibited in the data. Linear Mixed Model for Longitudinal Pulse Rate, Cox-Proportional hazard model for survival part time to default and joint model for the two response variables simultaneously linked by shared random effect, were used to identify factors that jointly affect the longitudinal change of the pulse rate and the time to default from the hospital Results: On the New York Heart Association Classification (NYHAC), 5.2% of the patients were class I and 14.9%, 36.1%, and 43.7% were found in classes II, III, and IV respectively. Patients with co morbidities like anaemia, pneumonia, chronic kidney disease were less likely to default as compared to patients without co morbidities. Based on the result of AIC, BIC and likelihood ratio tests joint model of longitudinal change and time to default was seen to perform better than separate models. Patients’ New York Heart Association Class, marital status, weight of the patient, Place residence, Hypertension, Anaemia, TB status and Type of CHF were found to be significant determinants of pulse rate.The estimated association parameter (α) in joint model is 1.3462 and statistically significance with (p-value<0.0001). This indicates that there is strong evidence of association between the effect of longitudinal change of PR and risk of defaulting. Conclusion: the variable NYHAC, TB co-infected status and LVEF were common factors for longitudinal measure of PR and time to default of CHF patients. The risk of defaulting for TB co-infected patients were higher as compared to non-TB co-infected patients with HR=8.24. Hence in hospital health professionals should be needed give special attention to the patients who had the co-morbidity disease. And government should allocate appropriate budget to hospital for hospitalization of CHF patients for the long time until recovery. Keywords: Linear Mixed Model, PR, time to default, Cox-Proportional hazard model, joint	en_US
dc.language.iso	en_US	en_US
dc.subject	STATISTICS	en_US
dc.title	JOINT MODELLING OF LONGITUDINAL CHANGE OF PULSE RATE AND TIME TO DEFAULT FROM TREATMENT DURING TREATMENT PERIOD OF CONGESTIVE HEART FAILURE PATIENTS IN CASE OF FELEGE HIWOT REFERAL HOSPITAL, BAHIR DAR, ETHIOPIA	en_US
dc.type	Thesis	en_US