Wed May 30, 2012
What Air Traffic Can Teach Us About Kidney Transplants
Originally published on Thu June 7, 2012 4:11 pm
This is the second of two stories we're doing this week on organ transplants. See the first story, Who Decides Whether This 26-Year-Old Woman Gets A Lung Transplant?
Nikolaos Trichakis is a Harvard Business School professor who studies air traffic. He was watching the news one night when a segment came on about the waiting list for kidney transplants.
He'd never thought much about organs, but it immediately became clear to him that his insights into air traffic might help patients waiting for kidneys.
Trichakis and his colleagues try to figure out how to balance fairness and efficiency.
In a purely fair system air traffic system, planes take off in exactly the order that they are scheduled. But if what you care about is efficiency, you may let a plane full of passengers with layovers take off first, so you don't have lots of people miss their connections.
In an organ-allocation system based solely on fairness, organs may simply go to the person who has been waiting the longest. But if you care about efficiency, you might give the healthiest organ to the patient who is likely to live the longest after the transplant.
Trichakis and his colleagues decided to try to figure out how to balance fairness and efficiency in kidney transplants. They spent last summer building a sophisticated computer model that included thousands of variables and decades of data on organs and patients and medical outcomes.
At the end of the summer, they ran their model against the formula doctors currently use to allocate kidneys. Trichakis' model was just as fair as the current system-- and enormously more efficient.
If you used their model to match patients and kidneys for one year, and you totalled up the extra life expectancy patients would gain, you'd get 5,000 extra years of life, according to their results.
Trichakis and his colleagues are now working with the doctors who determine how organs are allocated.
ROBERT SIEGEL, HOST:
Yesterday, we brought you the story of doctors trying to do the best for their patients in need of organs. Today, a group of mathematicians trying to do the best for the organs.
Chana Joffe-Walt, of our Planet Money team, reports on the effort to get the most life out of every available organ.
CHANA JOFFE-WALT, BYLINE: It may be possible to get 5,000 more years out of the kidneys we currently have available for transplant. And that is because last summer some guy picked up a magazine. The story starts here with a problem that's been facing doctors for years.
DR. JACK LAKE: Donated organs are a precious resource and you would like to gain the maximum amount of benefit from those donated organs.
JOFFE-WALT: This is Dr. Jack Lake. He's the president of the United Network for Organ Sharing, UNOS. That's the group that manages the National Organ Waiting List. And he says, right now, the waiting lists are designed with fairness in mind. So, doctors are focused on equitably distributing the organs across racial groups, blood types, which is good, he says.
But fairness doesn't necessarily mean that you get the most out of the donated organs.
LAKE: I think one of the things that people don't want to see happen is to see very young kidneys going into very old recipients, who may not have that long to live.
JOFFE-WALT: Doctors do not look at age when ranking patients on waiting lists. That is considered unfair, so doctors were kind of stuck, which is why it was so exciting when Dr. Lake happen to get an email from his daughter, who happened to get an email from her boyfriend, who happened to read in Harvard Working Knowledge that a couple mathematicians were trying to solve exactly this problem.
LAKE: I was fascinated. And so, I wrote to Professor Nikos Trichatos(ph) - it's a Greek name. It's hard to pronounce.
NIKOS TRICHAKIS: My name is Nikos Trichakis.
JOFFE-WALT: Nikos is not a doctor. He is a professor at Harvard. He just happened to be watching the news one night and they were talking about kidney waiting lists. He'd never really thought about the organs before.
TRICHAKIS: Not really. But obviously it was quite related to the work we'd been doing on air traffic.
JOFFE-WALT: The work around air traffic, Nikos and two colleagues have been trying to design and air traffic system that is fair but also efficient. And he says it's actually just like organs. In a purely fair traffic system, planes will take off in exactly the order they're scheduled. The same way in a fair organ system, organs may simply go to the person who's been waiting the longest for an organ.
But if what you care about is efficiency, with planes you may let the plane full of passengers with layovers takeoff first. And with organs, you may give the healthiest organ to the healthiest patient, so the organ can last longer.
Nikos brought the organ puzzle to his two collaborators, Professors Vivek Farias and Demetrius Bertsimas at MIT.
TRICHAKIS: We didn't even debate. I mean, we said this is obvious, let's move on.
JOFFE-WALT: This is how these three men seem to work. They gather in dark offices with white boards where windows should be, and they had a collective theory, they told me, about how to approach this fairness versus efficiency puzzle. Then organs came along...
TRICHAKIS: What is a better place to test it when the balance is life and death?
JOFFE-WALT: Doctors rank patients on organ waiting lists based on a couple variables - every so often they tweak it a bit. Nikos, Vivek, and Demetrius spent a summer building a sophisticated computer model that looked at thousands of variables, millions of choices, combinations, took into account decades of data on organs and patients and medical outcomes. It took three hours just for the computer program to churn through all the data.
TRICHAKIS: I was somewhere in the basement of the computer room...
(SOUNDBITE OF LAUGHTER)
TRICHAKIS: ...around here.
JOFFE-WALT: At the end of last summer, Nikos ran their model against the doctor's formula. And the model was just as fair as the doctors. It was also enormously more efficient. If you use this model to match patients and kidneys for one year, and you totaled up all of the extra time patients would get would live under this new model - consider all the people waiting for kidneys - you get more than 5,000 extra years.
LAKE: I was incredibly excited.
TRICHAKIS: Thousands of life spans.
JOFFE-WALT: Doctors are now working with the three men. And Vivek Farias says the hope is they can make the kidney allocation system as fair and as efficient as possible.
Chana Joffe-Walt, NPR News.
SIEGEL: And one final note, Planet Money has a new iPhone app where you can find the team's signature storytelling, blogging and innovative economics coverage. Transcript provided by NPR, Copyright National Public Radio.