This Week in Behavioral #8

A lot of research has been done on how to increase organ donation in the US -- how can we ask, what info can we give, should we pay. This week we look at an elegantly simple solution - just keep asking.

Opt-Out Default Increases Organ Donation

The most famous paper on organ donation rates has long been “Do Defaults Save Lives” by Eric Johnson and Daniel Goldstein. You might have seen this figure in any number of books or news reports:


On the left side of the figure (in gold) is the donation rates for countries where the default is you are NOT a donor, and you have to check a box (for example, when getting your driver’s license) to BECOME a donor. This is often called an “opt-in” policy. On the right (in blue) are the donation rates for countries where the default is that you ARE a donor, and you have to check a box to NOT be a donor. You don’t need fancy stats to tell you the bars on the right are substantially different than those on the left. An out-out default appears to lead to more organ donation that an opt-in one.

This concept of moving from opt-in to opt-out defaults has become a powerful tool not just in policy circles but also for <a href=" "target="_blank">companies looking to foster certain behavior. In fact, for some the opt-out option has become <a href=" "target="_blank">a dirty trick for advertisers and data miners - who generally have less virtuous objectives than the organ donation crowd - to obscure important privacy decisions.

So the question becomes, can we incentivize behavior without sacrificing true choice and transparency? Or more specifically to this question, could the US increase donation rates simply by changing the default, while importantly, still giving people complete power to choose?

What about just mandating a choice?

Prof Judd Kessler, a Forbes 30 under 30 scholar… a few years ago :), and Al Roth, a Nobel Laureate set out to collect data on this, and other policies. One problem with the chart above is that there are many differences between those countries. There may be other drivers behind the staggering correlation.

So Judd and Al brought students into the lab in Boston. They had them log into a website they created that then interacted with the DMV website. In this way, they were able to ask students whether or not they wanted to be a donor in any way they wanted, while making this an actual choice; they changed their organ donor status according to their choices.

In the study, the organ donation question either used an opt-in paradigm as above, or they forced you to make a choice (often called “mandated choice”). Mandated choice has become increasingly popular, to the point that several states including California have adopted it for how they ask citizens their organ donation preference.

What’d they find? Mandated choice, if anything, works worse than the opt-in paradigm.

Donation rates were lower when there was not a default. They also look at the data coming out of California and find the same thing:


When the rest of the country was seeing an uptick in organ donation rates, California’s starting going down, and starting going down precisely after the policy change.

But what if you just keep asking?

But it’s not all doom and gloom.

The researchers find two ways of increasing donation rates:

First, ask and ask again. People who said “yes” last time say “yes” again this time. However, some people who said “no” now say “yes”. Deciding to be an organ donor seems to be a more stable preference than choosing to not be a donor. If this is true, we should just keep asking. Ask everyone, every year.

Second, increase the information you provide at the decision point. (Warning this gets a little gross) You increase donation rates when you include a list of all the organs and tissues that you donate when you die -- two kidneys, liver, heart, pancreas, lungs, intestines, corneas, skin, heart valves, cartilage, bone, tendons, and ligaments. Apparently once you get over the heeby jeebies of reading that list, it reminds you of how many lives you can really help with your donation. And that increases donation rates substantially. I just don’t want to read that list every year...

It's a powerful example of both how experimental data should always be the golden standard, and how nuanced behavioral research can seemingly move mountains.


Lucas Coffman, Frank

Lucas is a professor of behavioral economics at Harvard University and the Chief Behavioral Officer at Frank

Frank - Luke Coffman

Read more posts by this author.

Subscribe to The Frank Blog

Get the latest posts delivered right to your inbox.

or subscribe via RSS with Feedly!