Everyone, face left. Screenshot from The Atlantic

by George Taniwaki

Which side of your face do you show when someone takes a photograph of you? Most people naturally turn right which shows the left side of their face.

The Atlantic Feb 2014 features a video by the science writer Sam Kean that discusses this observation and a theory about why it occurs. It may add insight into the right-brain, left-brain debate. Or it could be completely unrelated.

I usually turn left, showing the right side of my face, though that isn’t a natural behavior for me. I learned it as a child. My natural inclination is to face directly at the camera. But in grade school, on picture day, a photographer told me not to look straight at the camera and to turn my head. When I do that, I naturally want to look left, which shows the right side of my head. This is true even though I part my hair on the left.

When asked to turn my head to the right and show the left side of my face, I feel like I am showing off.


by George Taniwaki

Back when I was in high school, every summer my friends and I would go to the amusement park. One of the attractions we visited was the hall of mirrors. Ten points worth of tickets allowed you to go through the hall once. The goal is to go in through the entrance and find the exit before you starved to death. ( I’ve been told that there are numerous bodies of lost patrons still stuck inside the hall, but I digress.)

The hall of mirrors consists of a room with a series of tracks on the floor and ceiling laid out in an equilateral triangular array, called an isohedral tiling pattern (Fig 1).


Figure 1. Isohedral tiling. Image from Wikipedia.org

Each triangular cell is about 3-foot to a side. The tracks on some of the sides have a floor-to-ceiling partition inserted in them. The partitions form walls creating a maze with a single entrance, a single exit, and exactly one path between them. Solving a maze of this type is a great mathematics problem.

The maze is called a hall of mirrors because the partitions are not just solid opaque panels. Instead, they are all of one of two types, either mirrored on both sides or clear plastic.

I had never been in the hall of mirrors. While waiting in line, one of my friends urged us to go twice. The first time would be figuring the maze out. The second time we would race through. He then bet me he could get through it faster than me both the first and second times.

I eagerly accepted because I knew a trick for solving simple mazes called the wall follower algorithm. In the wall follower algorithm, you place one hand (say your left hand) on the wall as you enter the maze and never take it off. As you move through the maze, if you reach an intersection, keep your left hand on the wall, meaning you take the leftmost turn. If you reach a dead-end, keep your left hand on the wall, meaning you will return to the intersection and take the next path. Eventually, you will reach the exit.

If you remember the series of correct paths you took, the next time you enter the maze, you will not need to keep you hand on the wall. You just need to remember the turns you took at each intersection. For example, left, right, center, right, right, left, right.

Seeing how eager I was, all my other friends also made the same bet and I accepted. After giving our tickets to the operator, my friends ran into the maze. I was surprised that they would dash off without caution. I was determined to show them that I could beat them and solve the maze faster by simply walking through the maze using my logical skills.

What I didn’t know were three facts. First, because of the mirrors and clear walls, as you stepped into each triangular cell, you couldn’t be sure which direction would lead you to an adjacent open cell and which led you into a wall. Using the wall follower algorithm by placing your hand on the wall definitely helped, but it was slow going moving around.

Second, the maze was crowded. There were lots of other people who were moving around, sometimes in the opposite direction as me, and it was difficult to navigate around them. To do so, I often had to take my hand off the wall and as I was getting jostled, I couldn’t be sure I placed my hand back on the same wall. Similarly, it was difficult for me to remember if I had returned to the same point as before. This meant I couldn’t memorize which turn I should make at each intersection.

After a few minutes of trying to solve the maze, I noticed that all of my friends were already outside the maze watching me. I began to panic and became disoriented. Eventually, feeling sorry for me, they began shouting and pointing the directions for me to take. Finally, with their help, I reached the exit of the maze and stepped out to be with them. Except, I wasn’t quite at the exit yet and bam, I walked right into a clear wall mashing my eyeglasses into my face. So much for my superior maze solving skills.

Upon exiting the maze, I learned the third fact. My friends had all been to the amusement park previously that summer and had memorized the path through the hall of mirrors.

As we agreed, we went through the hall a second time. I did a lot better, but still made several wrong turns and became disoriented a couple of times. I was the last one out of the maze again. I had to pay off on two losing bets with each of my friends that day.

However, I did learn an important lessons about gambling (and investing in the market too). First, don’t place a bet on a game of skill simply because you know something your opponent doesn’t (like the algorithm to solve a maze). Only place a bet if you have actual experience winning the game you are betting on (like having run through the maze before). Second, if someone challenges you to a contest (who can run through the maze fastest), they probably already have the skills needed to win and you should avoid the bet.


Solving a maze using pencil and paper is another interesting problem. And is one that should not induce panic attacks about getting lost. One way to study a maze is to first identify the walls. A maze with a single entrance and single exit must have at least two separate walls as shown on the left of Figure 2.

In the case of a maze with exactly two walls, you can solve it using the wall follower algorithm described earlier. But a faster solution exists. Notice that any path where the wall on both sides is the same color ends in a dead-end. By following the path that has one wall of each color on each side you will quickly find the solution. Notice that this technique is faster only if the walls are already color coded.


Figure 2. Three simple mazes with two walls (left), three walls adjacent to each other (center) and three walls where one is enclosed (right) Image by George Taniwaki

A maze may have more than two walls. If there is only one entrance and exit, there will still be only two exterior walls. Any additional walls will be totally enclosed within the exterior walls. If an interior wall is at any point adjacent to two walls that are part of a solution, then a path following this wall will also add a solution. A wall that is adjacent to only a single wall (is totally enclosed within a wall) will not add another solution. An interior wall that is adjacent to one or fewer walls that is part of a solution will not add another solution either.

In Figure 2 above, the center maze has three walls and has two solutions. You can turn either left or right at the blue wall. The right maze has three walls but only a single solution since the blue wall is totally surrounded by the red wall.

For a more complex example, consider a maze (Fig 3a) that is included in a recent advertisement for Dropbox, a cloud file sharing service.


Figure 3a. Dropbox print ad. Image from Dropbox

It is hard to see the solution to this maze just by inspection. But if we color code all the walls we will discover there are four separate walls (Fig 3b, to save time I only added color at the turns and intersections). The two interior walls (in blue near the top of the puzzle) are both completely contained within a single wall (in red), so they do not add any new solutions, so there is only one solution. The solution is the path that stays between the two exterior walls (green and red). The solution is easy to recognize when the walls are color coded (assuming you do not have red/green color deficient vision). Try it and see how easy it is.


Figure 3b. Dropbox print ad with color coded walls and enclosed paths highlighted. The solution is the path that keeps walls of different colors on each side.

Notice that there are two errors in the maze. There are two paths (filled in yellow) that are completely enclosed, meaning they are not connected to the entrance or exit and so cannot be reached. Despite the errors, this is a nice maze. A good maze has the following attributes:

  1. The path for the solution is quite long and traverses all four quadrants of the grid, meaning that finding the solution path is not obvious if the walls are not color coded
  2. There are many branches off the solution path, meaning that there are many potential places to make an error
  3. Many of the dead-end branches are long and also have branches, meaning that discovering whether a path is a dead-end branch or part of the solution takes a long time

Nearly every state in the U.S. maintains a registry of people willing to become deceased organ donors. The intent of an individual to be a donor is stored as a Boolean value (meaning only yes or no responses are allowed) within the driver’s license database. Nearly all states use what is called an opt-in registration process. That is, the states start with the assumption that drivers do not want to participate in the registry (default=no) and require them to declare their desire (called explicit consent) to be a member of the registry either in-person, via a website, or in writing.

One of the frequent proposals to increase the number of deceased organ donors is to switch the registration of donors from an opt-in system to an opt-out system. In an opt-out system, all drivers are presumed to want to participate (default=yes) and people who do not wish to participate must state their desire not to be listed.

Let’s look at the logical and ethical issues this change would present.

Not just a framing problem

Several well-known behavioral economists have stated that switching from opt-in to opt-out is simply a framing problem. For instance, see chapter 11 of Richard Thaler and Cass Sunstein’s book Nudge and a TED 2008 talk by Dan Ariely using data from papers by his colleagues Eric Johnson et al., in Transpl. Dec 2004 and Science Nov 2003 (subscription required).

The basic argument is that deciding whether to donate organs upon death is cognitively complex and emotionally difficult. When asked to choose between difficult options, most people will just take the default option. In the case of an opt-in donor registration, this means they will not be on the organ donor registry. By switching to an opt-out process, the default becomes being a donor. Thus, any person who refuses to make an active decision will automatically become a registered organ donor (this is called presumed consent). This will increase the number of people in the donor registry without causing undue hardship since drivers can easily state a preference when obtaining a driver’s license.

However, these authors overlook two important practical factors. First, switching from opt-in to opt-out doesn’t just reframe the decision the driver must make between two options. It will actually recategorize some drivers.

Second, it changes the certainty of the decision of those included in the organ registry, which affects the interaction between the organ recovery coordinators at the organ procurement organization (OPO) and the family member of a deceased patient.

There are more than two states for drivers regarding their decision to donate

Note that the status of a driver’s intent to be an organ donor is not just a simple two-state Boolean value (yes, no). There are actually at least three separate states related to the intension to be an organ donor. First, upon the driver’s death, if no other family members would be affected, would she like to be an organ donor (yes, no, undecided). Second, has she expressed her decision to the DMV and have it recorded (yes, no). Finally, would she like her family to be able to override her decision (yes, no, undecided). The table below shows the various combinations of these variables.


Driver would like to be organ donor
Driver tells DMV of decision
Driver would permit family to override decision


1a Yes Yes No Strong desire
1b Yes Yes Yes or Undecided Weak desire
2a No Yes Yes or Undecided Weak reject
2b No Yes No Strong reject
3a Yes No Yes, No, or Undecided Unrecorded desire
3b No No Yes, No, or Undecided Unrecorded reject
4 Undecided Yes or No Yes* Undecided

*No or Undecided options make no sense in this context

Opt-in incorrectly excludes some drivers from the donor registry

Now let’s sort these people into two groups, one that we will call the organ donor registry and the other not on the registry.

Under the opt-in process, only drivers in categories 1a and 1b are listed on the organ registry. These drivers have given explicit consent to being on the registry. Drivers in categories 2a, 2b, 3a, 3b, and 4 are excluded from the registry. Thus, we can be quite certain that everyone on the registry wants to be a donor. (There is always a small possibility that the driver accidentally selected the wrong box, changed their mind between the time they obtained their driver’s license and the time of death, or a computer error occurred.)

In most states the drivers not on the organ registry are treated as if they have not decided (i.e., as if they were in the fourth category). When drivers not on the registry die under conditions where the organs can be recovered, the families are asked to decide on behalf of the deceased.

Under an opt-in process, drivers in category 2a are miscategorized. They don’t want to be donors and didn’t want their family to override that decision, but the family is still allowed to decide. The drivers in categories 3a and 3b are miscategorized as well. The ones who don’t want to be donors (3b) are also forced to allow their families to decide. The ones who want to be donors (3a) are now left to let their families decide.

Opt-out incorrectly includes some drivers in the donor registry

Under an opt-out process, drivers in categories 1a, 1b, 3a, 3b, and 4 are grouped together and placed on the organ registry. If the donor registry is binding and the family is not allowed to stop the donation, then the process is called presumed consent. (Note that many authors use opt-out and presumed consent interchangeably. However, they are distinct ideas. Opt-in is a mechanical process of deciding which driver names are added to the registry. Presumed consent is a legal condition that avoids the need to ask the family for permission to recover the organs.)

Drivers in category 3a who wanted to be registered are now correctly placed on the registry. But any drivers in category 3b who don’t want to be on the registry are now assumed to want to be donors, a completely incorrect categorization. Similarly, all drivers in the fourth category who were undecided are now members of the definite donor group and the family no longer has a say.

Only drivers in category 2a and 2b are excluded from the registry. We can be quite certain these people do not want to be donors. But some (category 2a) were willing to let the family decide. Now they are combined with the group of drivers who explicitly do not want to donate.

The distribution of categories into the registry under the opt-in and opt-out process and how they are treated are shown in the table below.

Categories added to donor registry
Categories not added to donor registry


Opt-in process 1a, 1b both treated as if in category 1a (explicit consent) 2a, 2b, 3a, 3b, 4 all treated as if in category 4 (family choice) Drivers in registry are nearly certain to want to be donors. Actual desire of drivers not on registry is ambiguous
Opt-out process 1a, 1b, 3a, 3b,4 all treated as if in category 1a (presumed consent) or 1b (family choice) 2a, 2b both treated as if in category 2b (explicit reject) Drivers not in registry are nearly certain to not want to be donors. Actual desire of drivers on registry is ambiguous


Ethical implications of misclassification

If there are no drivers in categories 3a, 3b, and 4, then switching from opt-in to opt-out will have no impact on the size of the donor registry. However, if there are any drivers in these categories, then some will be incorrectly categorized regardless of whether opt-in or opt-out is used. This miscategorization will lead to some ethical problems.

Under opt-in, there may exist cases where the drivers has made a decision to donate (category 3a) or not (categories 2a or 3b) but family members overrules it. These errors are hard to avoid because they are caused by the lack of agreement between the drivers and other family members.

However, under opt-out combined with presumed consent, there may exist cases where neither the driver (category 3b) nor the family want to donate, but cannot stop it. Similarly, the driver may want to let the family choose whether to donate (category 4) and the family does not want to donate but cannot stop it.

It appears that from an ethical perspective, opt-in is less likely to create a situation where the respect for individual’s right to make decisions about how the body should be treated is denied. For further discussion of the ethical issues see  J. Med. Ethics Jun 2011, and J. Med. Ethics Oct 2011 (subscription required).

Next we will look at the impact switching from opt-in to opt-out will have on the interaction between the organ recovery coordinator and the family. See Part 2 here.

[Update: This blog post was significantly modified to clarify the “decision framing” issue.]

This is the final blog entry on solving the problem of getting hospitals to cooperate and allow a national kidney exchange to match all their live donors. You can see part 1 here and part 2 here.


There are several ways to encourage hospitals to submit all their pairs to the exchange rather than withhold some. One mentioned in a paper by Itai Ashlagi and Alvin Roth in Nat. Bur. Econ. Res. Jan 2011 is to use a lottery that rewards hospitals with more matches if they enter their easy-to-match pairs into the exchange. Hospitals that enter more O blood type donors will be rewarded by getting more matches for their O blood type patients. (As noted earlier, currently most transplant centers do not place their matched patient-donor pairs in exchanges. This must change for exchanges to reach their full potential.)

The solution suggested by Mr. Ashlagi and Mr. Roth reduces the number of total transplants compared to full cooperation with no incentive, but produces more transplants than under the current practice of noncooperation.

Some other possible solutions could involve incentives that don’t reduce the total number of transplants. For instance, the exchange could publicize the count of paired matches made for each hospital internally vs through the exchange (basically shaming the hospitals that don’t fully participate). Even if the transplant centers do not reveal the total number of pairs to the exchange, this number is publicly available. Each transplant hospital is required to report the total number of swaps it performs to the United Network for Organ Sharing (UNOS). The exchange knows how many swaps it facilitated. The difference between the UNOS’s count of live unrelated donors and the exchange’s count of pairs entered will be the number of transplants the hospital conducted internally.

Another solution that doesn’t affect the total number of transplants is to reward cooperating hospitals with first choice of donors if there are multiple matches.

Finally, exchanges can do a better job of handling preferences and constraints requested by the participating hospitals. Rather than having a collection of regional exchanges in order to meet the needs of a set of hospitals, a single national exchange can include preferences for maximum shipping time/distance, maximum donor age, minimum and maximum donor kidney size, maximum HLA mismatches, etc. for each hospital and even each patient. It can give preferences to juvenile patients, patients with high cPRA, patients who have been waiting more than six months in the exchange, etc. The exchange can use these constraints to find the favored matches without sacrificing the total number of transplants.


One would expect the kidney exchange market to evolve into a natural monopoly with one exchange gaining all the participants by offering the highest likelihood of a match in the shortest possible time.

However, we are not seeing that at this time because of difficulty in getting exchanges and participating hospitals to work cooperatively and quickly. These problems can be resolved and I expect kidney exchanges to grow until nearly all live donor transplants are mediated through them.

I hope that regional kidney exchanges do not form and instead the problems in the national exchanges are solved. The formation of regional exchanges would split up the pool of potential matches. Finding the easy matches locally and the pushing the hard matches to another pool will lead to suboptimal number of transplants. This is a serious issue because it means some patients will die while waiting for a transplant.

[Disclosure: I do volunteer work for the National Kidney Registry, one of the several exchanges that are the subject of this three-part blog post.]

This is a continuation of yesterday’s blog post on why national kidney exchanges are not reaching their full potential.

In yesterday’s post, we described how a single national kidney exchange would be efficient. By having a large pool of candidates, it will lead to both more matches and faster matches. But we observe some hospitals do not join an exchange. And even hospitals that do join an exchange still perform some or most of their matches in-house. Below are some reasons. Part 3 will outline some solutions.

Hospitals believe they will get more transplants doing swaps in-house

Everybody wants to do what is best for the patients. However, that is hard to know what that is in practice. Hospitals want to do what is best for their own patients, the ones they know and care for. It is difficult for doctors at a single hospital to judge what is collectively best for all the patients in the U.S. One of the problems facing a kidney exchange is that maximizing the number of transplants in the pool may not maximize the number of transplants within a hospital that is a member of the exchange.

Let’s say there are two transplant hospitals A and B. Hospital A has 3 pairs in its pool and can match all 3 of them. Hospital B has 4 pairs and can match 2, for a total of 5 transplants as shown below. Black lines show matches used while orange lines show matches that are not used.


Five transplants when hospitals don’t cooperate. Graphic based on Nat. Bureau Econ. Res.

Now let’s combine the pairs from the two hospitals in an exchange. If we do so, we find we can get a total of 6 transplants as shown in the figure below.


Six transplants when hospitals cooperate. Graphic based on Nat. Bureau Econ. Res.

Hospital B goes from 2 transplants to 4. But notice that Hospital A drops from 3 transplants to 2. The patient in Pair A1 no longer gets a kidney and Hospital A performs one fewer profitable transplant. Hospital B and patients in pair B1 and B2 benefit at the expense of Hospital A and the patient in Pair A1. Thus, Hospital A has an incentive to withhold its pairs from the exchange and perform the swaps in-house.

If every hospital performs all the easy matches in-house, then the exchange will contain fewer pairs. This will make finding matches harder. Even worse, the exchange will only contain hard-to-match pairs, making it even less likely that patients in the exchange will find a match. Hard to match pairs will be patients with O blood type (for more see this Mar 2010 blog post) and patients with high levels of antibodies to human leukocyte antigens (for more see this Feb 2011 blog post.)

Note that most hospitals may not even realize that they are withholding pairs from the exchange. If a patient and donor match (which is likely if the donor is blood type O), the hospital will just proceed with the transplant without even considering entering them into an exchange. By transplanting their easy-to-match O donor pairs directly, they leave the national pools with a surplus of O patients and a shortage of O donors.

Hospitals believe there is less delay doing swaps in-house

Another reason hospitals may prefer to handle swaps in-house is the perceived high administrative cost and delay caused by placing patients in an exchange.

For example, the United Network for Organ Sharing (UNOS is the national organization responsible for the distribution of deceased donor kidneys) has started a pilot program to create a national living donor kidney exchange. It taken over two years to develop a consensus of how to operate the program. Finally, in November of last year it conducted its first match run which found 3 sets of matches. Only one of them was accepted and resulted in 2 transplants. Since then it has not had a single match offer accepted and no further transplants have occurred. (See Jan 2011 blog post for details.)

Hospitals are quickly learning that a majority of offers made by the national exchanges do not lead to a transplant. With a lack of success in a national exchange, hospitals would be negligent to not try to help their patients by conducting matches within their own patient pools or form small regional pools.

Here’s an explanation why I think match offers may not lead to a transplant. Imagine a swap that involves three sets of patients-donor pairs. For each of the three transplants, the surgeon has to approve of the donor. If any one is rejected, then the entire swap fails. Then all transplant pairs require a cross-match test for compatibility. Again, if any one fails or cannot be completed within the required time limit, then the swap fails. Finally, all six surgeries must be scheduled. If any of surgeries cannot be scheduled within the required time windows, then the swap will fail. If each of the 3 step for each of the 3 transplant has a 7% chance of failure, then the cumulative chance of success for a 3-way swap is only about 50 percent (1 – 0.07)^9 = 0.52.

All of the steps above are easier to coordinate if they are conducted within a single hospital. An important key to success for a national exchange is to remind every transplant center how important it is to get the approvals and tests completed in a timely manner and to drive these transplants to completion.

Hospitals believe there are lower medical risks doing swaps in-house

Finally, some hospitals fear that participating in an exchange will expose them to higher risk donors. Each hospital does a very thorough examination of donors prior to accepting them into the transplant program. Accepting a donor that they did not evaluate exposes them to two risks, one real and the other perceived.

Let’s cover the perceived risk first. A surgeon at the transplant hospital probably believes the evaluation of donors done at her hospital is excellent and trusts all members of the transplant team. However, in an exchange, the donor comes from another hospital. The surgeon may not personally know the evaluation team at the donor hospital. She may not be familiar with the evaluation criteria used at that hospital. In fact she may believe that the testing done there may not be  is not as rigorous as its own.

I believe that this concern will be alleviated over time as hospitals become more comfortable with the concept of cross-hospital exchanges. There are only 268 transplant centers in the U.S. and most of them use a very similar criteria when evaluating donors. Even if the hospitals use different criteria for acceptance, the equipment they use are very similar so the test results themselves should be comparable across hospitals.

The real risk is that a patient and her surgeon may be subjected to is that the donor hospital may not be as careful in evaluating a donor in an exchange, knowing that it will not be responsible for the outcome of the transplant. This type of risk is known as moral hazard. It is one of the factors that led to the recent financial crisis. Banks reduced the effort made to ensure mortgages were properly evaluated when they knew they would not be responsible for losses caused by any future loan defaults. This is a real risk and has to be managed. One solution is to make sure a certain percentage of matches made in the national exchange include pairs in the same hospital. This should encourage hospitals to do a good job of evaluating donors, since they won’t know which transplants will remain in-house.

In addition to donor evaluation risk, accepting a kidney from another hospital also entails transportation risk. Performing a transplant completely within a single hospital means that the kidney travels a few feet between the donor and the recipient.

The trauma a kidney undergoes is divided between warm ischemia time (the time it takes from when blood stops flowing to the organ to the time it is packed in ice) and cold ischemia time (the time it takes to transport the chilled organ from donor’s operating room to the recipient’s operating room and reattach it). The warm ischemia time causes the most damage. It will be a few minutes and it won’t differ whether a kidney is recovered within the same hospital as the patient or in a different one. The cold ischemia time for an in-house exchange can be as short as 10 to 15 minutes. However, if the donor operation takes place in New York while the transplant operation is in Los Angeles, the cold ischemia time may be as long as ten hours if there are flight delays.

Some transplant centers will not accept live kidneys that have been transported by air. I believe this is an unnecessary restriction. All transplant centers accept deceased donor kidneys recovered from outside their hospital. These kidneys are often delivered by commercial or charter aircraft, sometimes with cold ischemia times of over 20 hours. (For more on shipping kidneys, see this Dec 2010 blog post and an upcoming blog post.)

The third and final blog post provides ideas to solve these issues.

One of the latest innovations in helping patients with end-stage renal disease (ESRD) find live donors is the kidney swap. A kidney swap starts with a kidney patient who knows a person willing to donate but is not tissue compatible (let’s call them pair 0). If that patient finds another pair in a similar situation (let’s call them pair 1), then there is a chance the two pairs may be able to swap donors to produce a compatible match. If this happens both patients get a transplant and both donors can provide the gift of life, though not to their originally intended recipient (see image below).


An example kidney swap. Graphic by George Taniwaki

It is generally too difficult for patients to find partners for a kidney swap on their own (for a rare counterexample see Globe and Mail Feb 2011). Thus, these swaps are usually facilitated by the hospital where the patients are registered on the transplant waiting list. A transplant nephrologist at that hospital will periodically scan the list of patients with unmatched donors (done by computer nowadays) and see if any pairs potentially match. If any do, a transplant surgeon will review the matches to approve/reject the surgeries, a blood lab will run cross-match tests to ensure the patients’ immune systems will not react to the potential donors’ organs, and a transplant coordinator will schedule the surgeries.

These kidney swaps are a growing source of kidney transplants and now account for over 300 transplants a year in the U.S. (for more details on the rise in kidney swaps, see this Jun 2010 blog post).

However, unless the hospital had a large pool of unmatched pairs (say over 20 pairs), it would be unlikely to find matches for its sensitized patients who require more closely matched kidneys. To find matches for these patients, it needs a bigger pool to choose from. Thus, smaller hospitals began to band together to form kidney exchanges. There are now several such exchanges in the U.S., each vying to become the largest in order to maximize the chance of finding a match and thus minimizing the wait time for participating patients.

Natural monopoly

If size of pool was the only factor that led to more transplants getting done faster, then the exchange with the largest pool, even if it was only slightly larger than the others, would provide more matches and do them faster. Hospitals that were members of other exchanges (or doing swaps in-house), would see the success at this exchange and would switch to it, making the pool even larger, leading to even more matches faster. Eventually, all hospitals would join this one exchange to take advantage of the gains in performance and all the other ones (including the in-house exchanges) would be driven out of the market.

However, we don’t see this. For instance, an article in SFGate  Apr 2011 highlights a 5-way swap that California Pacific Medical Center just completed, its largest single-swap ever. Yet California Pacific also is a member of the largest kidney exchange in the U.S.


A five-way swap. Graphic from SFGate

In a similar vein, I recently attended a seminar held by one of the transplant centers in Seattle. The director of kidney transplant program stated that the center had performed several swaps in the past year, all in-house. They were also working with all the other transplant centers in the northwestern U.S. (4 in Seattle, 1 in Spokane, and 3 in Portland) to form a regional exchange. This would be in addition to the in-house exchanges and national exchanges that all these transplant centers participate in.

These stories highlight the growing interest in kidney exchanges among transplant centers. But they also point to a failure by the national kidney exchanges to meet the needs of their member hospitals.

I believe there are three reasons we see this reluctance by transplant centers to rely on large national kidney exchanges for all their swaps. They are the fear in missing some  transplants, fear of loss of efficiency and control, and fear of medical risks. I will describe each of these in the next blog entry.

Economics has been called the dismal science. The term implies that left on their own, individuals will adopt, or retain, behaviors that may be good for them, but will lead to suboptimal social outcomes.

One of the most fascinating areas of economic research today is called behavioral economics. It is the study of how people make decisions and is used to explain why people become addicted to drugs, commit property crimes, or gamble against poor odds. It’s not just used for explaining bad behaviors. Behavioral economists also use their theories to explain why people marry, landscape their yard, or perform altruistic acts.

This research into the effect of behavior on economic decision-making actually has important policy implications. For instance, in 2006 Congress changed the rules for defined-contribution retirement plans known as 401(k)s. Prior to the change, most plans were opt-in, meaning employees had to choose to participate in the plan. The change in rules allowed companies to automatically enroll all new employees in the retirement program (into a safe investment as defined by the U.S. Dept. of Labor) unless they chose to opt-out.

Traditional economic theory says that the desire to invest in a retirement plan should be independent of whether the employee has to choose to opt-in or opt-out of the plan at the time of hire. And yet behavioral studies show people tend to prefer to do what everyone else is doing, or what an authority figure says is best. Data collected from firms that switched from opt-in to opt-out show that participation rates rise, from 40 to 70% when employees had to opt-in, to over 90% after adopting opt-out.

Richard Thaler, a leader in the behavioral economics movement, calls it “libertarian paternalism.” That is, the goal is not to restrict choices, but to offer them in a way that leads to better outcomes. Another way to increase retirement savings that Mr. Thaler promoted, and convinced Congress to accept, is called Save More Tomorrow. He is also one of the coauthors of Nudge, a great book on how to improve your decision-making skills. [Disclosure: I am a graduate of Univ Chicago’s Booth School of Business, where Mr. Thaler teaches.]


Nudge: Improving Decisions About Health, Wealth, and Happiness. Image from Amazon

Last week, my wife pointed me to an article in the New York Times that shows how behavioral economics research is now being used to “improve” the effectiveness of political campaigns. One of the most influential groups in the 2008 presidential campaign by Barack Obama is called the Consortium of Behavioral Scientists. One of the members is Mr. Thaler. You may recall that Mr. Obama taught constitutional law at the Univ. of Chicago and was already quite familiar with Mr. Thaler’s work.