Parametric Survival Models of Hemodialysis Patients in Relation with Patient-Related Factors

Abstract

Background: Survival analysis refers to analyzing of statistical data for which the outcome variable of interest is time until an event occurs. This research aimed at comparing different models of parametric Proportional Hazards (PH) models (Weibull, exponential, Gompertz) in patients with hemodialysis to determine the best model for assessing the survival of patient. Study consists of 325 hemodialysis patients who referred to public hospitals in Khartoum state in the period from December 2005 to December 2015. Data was used to estimate the survival function with view to identify risk factors influencing among end-stage renal disease (ESRD) population. Based on the Cox-Snell Residuals and AIC, BIC, and Gompertz (PH) model is an efficient model than other when the values of (AIC=662.21), (BIC=703.83) and (R2=0.211) where maintained Study assessed that the variables dealing with univariate models were signifi cant but had a signifi cant effect on hemodialysis survival. The Gompertz model had the smallest AIC and BIC value; therefore; it was selected as the most appropriate model. In multivariable analysis, the BIC had the lowest value and the highest value in each analysis. The study assessed that diabetes mellitus and hypertension, regular, and hospital, had a significant effect.

Keywords

• survival analysis
• parametric models
• hemodialysis
• cox-snell residuals
• aic