ABSTRACT Background: Renal failure is one among the slowly progressive diseases of kidney function characterized generally by low glomerular filtration (GRF). The replacement therapy of renal failure by hemodialysis involves the removal of excessive toxic fluids and toxic metabolic end products from the body. One continuous and five categorical predictors were included in the analysis. The mean baseline age of renal failure patients was 36.64 years. Out of 269 renal failure patients 118(43.87%) were female and 150(55.76%) were hypertensive. Joint models typically combine linear mixed effects models for repeated measurements and Cox models for censored survival outcomes. Thus, the aim of this study was to present joint modelling on longitudinal glomerular filtration rate measurement and time-to-death of renal failure patients treated under hemodialysis. Methods: Hospital based retrospective study was conducted among renal failure patients attending hemodialysis between 2016 and 2018 at Saint Paulo’s Hospital Millennium Medical College, Addis Ababa, Ethiopia. The longitudinal eGFR and the time to event (i.e. death) data with the separate modeling approach and the joint modeling approach was fitted. A total of 269 renal failure patients screened who were under hemodialysis follow-up at Saint Paulo’s Hospital Millennium Medical College. Results: The results for separate and joint models were quite similar to each other but not identical. However, the estimates of the association parameters in the joint analysis were significantly different from zero, providing evidence of association between the two sub-models. The relationship between kidney function as measured by eGFR and the hazard for death was negatively significant. Thus, death is less likely to occur in patients with higher eGFR. Conclusions: When evaluating the overall performance of both the separate and joint models in terms of model parsimony, goodness of fit, smaller total AIC, and the statistical significance of both the association parameters, the joint model performs better. Thus, authors concluded that the joint model was preferred for simultaneous analyses of repeated measurement and survival data. Key words: Hemodialysis, Chronic Kidney Disease, Joint Model, Longitudinal Data, Survival Data Cox PH model.