Survival Modelling of Haemodialysis Patients on Covariates of Clinical and Demographic Factors

Ramkumar Balan

Background and Objective: Most of the patients of CKD in Tanzania due to lack of knowledge and fear of the treatment cost keep away from modern medicine and are trapped to death even in younger years. The aim of this study was to develop survival models for hemodialysis patients by determining cofactors influencing the mortality of dialysis patients. Materials and Methods: A sample of 171 dialysis patients admitted to Muhimbili Hospital in 2015 and followed up to 2018 were studied. Basic prevalence was determined and the survival model on parametric semi-parametric and non-parametric methods was found. The Cox CPH and Kaplan-Meier model are used in analysis to identify the significant survival curves on smoking, alcohol habit and HIV status. Results: Out of 171 patients, 148 survived between 0-500 days, 20 survived between 501-1000 days and only 3 patients survived in 1000+ days. Factors affecting survival are sex, increased number of dialysis, blood transfusion and alcohol consumption. Log-normal distribution was the best parametric fit for the data and the average survival time was 268 days, while the CPH model exhibits alcohol habit and the number of dialysis as significant covariates. The KM curve and rate of mortality curve depict the significant difference under smoking, alcohol consumption and HIV-infected patients and log rank tests validated it. Conclusion: The CKD and dialysis treatment are more common in males in Tanzania and few survive after three years of treatment and follow-up. Increased number of dialysis, lack of hygienic blood transfusion and alcohol intake are leading many CKD dialysis patients to death.

View Fulltext Back