Cost-effectiveness and budget impact analysis of peritoneal dialysis and haemodialysis in South Africa.

Background: More than 800 million people are affected by chronic kidney disease (CKD) worldwide. In South Africa, the prevalence of CKD increased by 67% between 1999 and 2006. Haemodialysis (HD), peritoneal dialysis (PD), and kidney transplant are the three main modalities used for managing end stage kidney disease. The cost of these therapies poses a significant burden to the health care system in South Africa. The aim of this study is to determine the cost-effectiveness and budget impact of peritoneal dialysis versus haemodialysis from the societal perspective in South Africa.

Methods: A Markov model was constructed to estimate the cost-effectiveness of peritoneal dialysis versus haemodialysis. The model was developed in excel and populated with clinical evidence and cost data synthesized from the literature. The costs and outcomes were estimated over a 5-year time-horizon. The outcomes were presented as quality-adjusted life years. Cost effectiveness was estimated using the incremental cost-effectiveness ratio and the incremental net monetary benefit (INMB). Probabilistic sensitivity analysis was also conducted to assess the robustness of the results. A budget impact model was constructed to estimate the impact of PD and HD over a 5 year period.

Results: The total discounted costs per patient over 5 years were R788 384 for PD versus R1 227 708 for HD. The incremental cost for providing PD was estimated at -R438 875. The net QALYs for delivering PD compared to HD were estimated at -0.09. Cost effectiveness ratio for PD versus PD was R5 096 154/QALY. At a threshold of R38 500, PD provision has a 79% probability of being cost-effective relative to HD. The INMB was estimated at R328 574 for PD and R322 194 for HD indicating the cost-effectiveness of PD. The budget impact analysis showed that it would cost government approximately R25 billion over 5 years to treat all individuals eligible for KRT under the current scenario of 88% HD and 12% PD.

Conclusions: In South Africa, PD is shown to be cost-effective at a willingness to pay threshold of less than R38 500. A PD-preferred policy that considers clinical appropriateness and patients' values should be considered.