EN
Back to the archive

The encyclopedia · R&D & Science · Technical decision · 2007-2009

Kidney donors kept the chain alive by donating after their recipient.

In 2007 an altruistic donor started a ten-transplant kidney chain where donors gave after their own recipient, not all at once.

Alliance for Paired Donation

the move

In a paired kidney exchange, an incompatible donor-recipient pair swaps with another pair, and everyone usually operates at the same time so no one can renege. That simultaneity limits how many pairs can join, because getting enough operating rooms and people together at once is hard.

The Alliance for Paired Donation instead began a chain in July 2007 from a single altruistic donor, and trusted each paired donor to donate only after their own recipient had been transplanted. Five surgeries were simultaneous; in the other five, bridge donors continued the chain up to five months later.

The result was a ten-transplant chain across six centers in five states, coordinated by two registries over eight months, with no donor reneging. Economists including Alvin Roth shaped the matching; computer scientists built the optimizer that found the chain.

why it works

  • Simultaneity required many operating rooms and made large chains impractical
  • A donor trusts an ideal, so they will still donate after their recipient is safe
  • One altruistic donor can seed a whole chain of transplants
  • Better matches also yield higher-quality kidneys that last longer
the payoffLet donors give after their recipient, not simultaneouslyinspired

what transfers

Remove the simultaneity constraint and a matching market unlocks far more trades; the trick is designing trust so a later commitment is still honoured.

what came after

By the time the chain was reported in 2009, the Alliance for Paired Donation had started six NEAD chains producing 19 transplants, with 13 bridge donors and none reneging. The approach helped national registries grow chains into their signature way of serving hard-to-match patients.

references

spotted an error? The archive wants to know.

same kind of clever