Handling missingness and specify the dropout models in haematocrit levels over time in patients who received renal graft

Abstract

Chronic Kidney Disease (CKD) or renal failure is a public global health problem with an estimated prevalence of as 8 to 16% worldwide. This study was conducted in order to Specify Dropout Models the haematocrit levels over time in renal patients after their transplant. This is a longitudinal study that consisted of 1160 patients who received a renal transplant. Level of Haematocrit level as a response, time (log (years)), gender and age of the patients just to mention a few were as a covariate. Different statistical methods, such as dropout models were employed. Direct likelihood and MI based approach of Linear Mixed Model revealed that all the covariates included in the final models were significant. It was observed that the evolution of haematocrit levels follows a quartic time trend and the effect of time on the evolution varies according to gender and age. The effect of time followed a significant 5th power time trend under MI GEE and quadratic time trend under WGEE. However, both models are valid under MAR. In addition, age, gender and rejection symptoms have a significant effect on the evolution of haematocrit levels in renal patients, however their effect varies with time.