Deceased donor kidney allocation: an economic evaluation of contemporary longevity matching practices

Abstract

Background: Matching survival of a donor kidney with that of the recipient (longevity matching), is used in some kidney allocation systems to maximize graft-life years. It is not part of the allocation algorithm for Australia. Given the growing evidence of survival benefit due to longevity matching based allocation algorithms, development of a similar kidney allocation system for Australia is currently underway. The aim of this research is to estimate the impact that changes to costs and health outcomes arising from 'longevity matching' on the Australian healthcare system.

Methods: A decision analytic model to estimate cost-effectiveness was developed using a Markov process. Four plausible competing allocation options were compared to the current kidney allocation practice. Models were simulated in one-year cycles for a 20-year time horizon, with transitions through distinct health states relevant to the kidney recipient. Willingness to pay was considered as AUD 28000.

Results: Base case analysis indicated that allocating the worst 20% of Kidney Donor Risk Index (KDRI) donor kidneys to the worst 20% of estimated post-transplant survival (EPTS) recipients (option 2) and allocating the oldest 25% of donor kidneys to the oldest 25% of recipients are both cost saving and more effective compared to the current Australian allocation practice. Option 2, returned the lowest costs, greatest health benefits and largest gain to net monetary benefits (NMB). Allocating the best 20% of KDRI donor kidneys to the best 20% of EPTS recipients had the lowest expected incremental NMB.

Conclusion: Of the four longevity-based kidney allocation practices considered, transplanting the lowest quality kidneys to the worst kidney recipients (option 2), was estimated to return the best value for money for the Australian health system.

Keywords: Cost utility analysis; Kidney allocation; Longevity matching; QALY; Transplant.

Fig. 1

Markov model used. The Markov model has four health states: waitlisted for a kidney, kidney transplanted, post graft-failure dialysis and death. The cohort starts at the “waitlisted for a kidney” health state and patients in the cohort will be in this health state until they are transplanted or until die. When a patient transitions to the “kidney transplanted” health state they can experience either graft failure or death, or continuing successful transplantation

Fig. 2

Mean KDRI and EPTS values according to different allocation practices. a Distribution of the mean KDRI values with 95% confidence interval according to different age groups and different allocation options. b Distribution of the mean EPTS values with 95% confidence interval according to different age groups and different allocation options

Fig. 3

Mean and range of the incremental cost, QALY and NMB for each allocation options compared with current practice; a Incremental cost, b Incremental QALY, c Incremental NMB. The black vertical line in all three graphs indicate the range of values generated from the 20,000 iterations in PSA. Negative incremental cost (a) indicates a cost saving compared to current practice. The option with the highest probability of negative incremental cost has the most probability of being cost saving compared to current practice. Negative incremental QALY (b) indicates less effectiveness compared to the current practice and the option with the lowest probability of negative incremental QALY has the most probability of being effective compared to current practice. Negative incremental NMB (c) indicates the option is not cost effective compared to current practice. Probability of error indicated the probability of the option not being the cost-effective option compared to current practice. Therefore, the option with the lowest probability of error is the most suitable option. ^*Results presented for 1000 patients for 20-year time horizon; ^# NMB - Net Monetary Benefit. Option 1: Best 20% of KDRI donor kidneys transplanted to best 20% of EPTS recipients. Option 2: Worst 20% of KDRI donor kidneys transplanted to worst 20% of EPTS recipients. Option 3: Youngest 25% of donor kidneys transplanted to youngest 25% of recipients. Option 4: Oldest 25% of donor kidneys transplanted to oldest 25% of recipients.