RI: Small: New Computational Techniques and Market Designs for Kidney Exchanges and Other Barter Markets

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

In the US alone, 35,000 patients with terminal kidney disease get added to the deceased-donor wait list each year, and 100,000 await for a kidney transplant - more than all other solid-organ transplants combined. That demand far exceeds the supply of deceased-donor kidneys. It is possible for a living person to donate a kidney, but it is unlikely that a given donor can donate to a given patient due to blood type and tissue type incompatibilities. This opens the door for kidney exchange where such willing donor-patient pairs "swap" donors. In modern kidney exchanges these "swaps" are conducted via cycles and altruist-donor-initiated chains. Prior work in developing the methods, algorithms, and software for kidney exchanges has already been used by two of the largest regional kidney exchanges in the US. More importantly, it runs the United Network for Organ Sharing (UNOS) nationwide kidney exchange, which includes 153 transplant centers. This work has uncovered a myriad of open problems in barter exchanges, which this grant will address by developing computational techniques, conducting computational experiments, and fielding the fruits of the research. The research under this grant has the potential to save hundreds of lives annually in the US alone, and the integration with the deceased-donor waiting list has the potential to save thousands. The approach also leads to dramatic improvements in the quality of life by moving patients off dialysis and back into the productive work force. Societal benefits come also from transplants being less expensive than dialysis; this effect has been independently estimated to be $750 million in the US over five years. This project will focus on four prongs for kidney exchanges and other barter markets, as follows: 1. developing computational techniques for automatically deriving market-clearing policies for dynamic problems; 2) developing computational techniques for automatically determining dynamically which edges (possible donations) to test for compatibility and other viability; 3) developing market designs and computational techniques for integrating the deceased-donor waiting list and kidney exchange; and 4) developing market designs and computational techniques for generalizations where a patient can have multiple donors and vice versa. The techniques developed here will apply to other barter exchanges in a broad range of other domains as well.