Suppose that you were applying to a prestigious national fellowship. There are only 100 slots, and thousands of plausible candidates across the country. Would you prefer to see your application routed to a single national applicant pool, or would you prefer to see each application routed to fifty state admissions committees across the country, each one of which can pick two winners?
Picking winners in each state ensures a nationally dispersed group. Yet a single national process will ensure higher average quality among those chosen. California’s just-missed number-three candidate probably compares favorably with the second-place winners in Idaho and South Dakota. Even if one addresses this obvious problem by picking some set of regions of equal size, local variability still poses a basic challenge to a decentralized system. Maybe the mountain states happened to have three outstanding candidates when New Jersey had a poor year.
This isn’t the world’s biggest problem if one is choosing Rhodes Scholars or whatever. A super-elite group of worthy winners will be selected either way. The fact that a centralized process would produce a slightly more impressive group than a decentralized process is neither here nor there. Which selection process would you prefer? Well that depends on your competitive standing and where you live.
The stakes are higher when you are waiting for a heart, a liver, or a lung. Yet the basic tension is similar. The United Network for Organ Sharing (UNOS) plays a fundamental role in allocating solid organs for transplantation within the United States.
In the year 2000, the U.S. Department of Health and Human Services issued a long-awaited “Final Rule,” which intended to ensure that organs were allocated “based on medical criteria, not accidents of geography.” The Final Rule was written in reaction to a growing literature indicating that precisely such “accidents of geography” were playing a large role in allocating scarce organs. A beautiful analysis conducted by an expert committee at the Institute of Medicine examined 68,000 liver transplant waiting list records to find (among other things:) “Current allocation policies fail to provide organs to the neediest patients,” and “sharing organs across larger populations” would save a notable number of lives by allocating more livers to the highest-priority patients.
A decade later, some of the same researchers, on a team led by the University of Chicago’s Mark J. Russo, found similar problems in lung transplantation. The title of their paper speaks for itself: “The local allocation of donor lungs results in transplanting lower priority lung transplant candidates.”
The number of lives at stake is rather startling. Examining 2009 data, the authors found 185 patients who died on the transplant waiting list after having been passed over for a suitable organ in their region when preference was given to a lower-priority candidate who lived closer to the donor. I don’t know how many of these patients might have survived. I do know that many lives are being lost.