Manipulated by Incentive

Manipulated by incentive

We are vulnerable to the influence of incentives—whether from money, prestige, or power. Most insidious is when an incentive is designed to appeal to our personal narrative about the kind of person we are. No one wants to be thought of as bad, so we often perform intellectual contortions to justify our pursuit of an incentive as good.

Thalidomide is a drug with a tragic history. Discovered and developed by a West German drug manufacturer in the 1950s and widely marketed as a sedative and antinauseant, it was responsible for thousands of infant deaths and deformities. The story of how this happened exposes a lot of the nuances of incentives.

Thalidomide was first produced by the Chemie Grunenthal company. It was “a drug without a disease,”5 and thus the first tests in animals and humans were actually to find something it could cure. The scientists there could not find a dosage high enough to kill rats and no side effects were observed on any animals tested. From this they concluded that the drug was completely nontoxic. They still didn’t know what humans could use it to cure, but these original animal testing results formed the foundation of their enduring belief that the drug couldn’t hurt anyone.6

The Grunenthal company then began handing out free samples to doctors. There was no clinical trial as we understand it now. Instead the initial users of the drug were the clinical trial. Thalidomide was put out into the world and “tested” on people without their knowledge or consent. The general population was the lab.7

As Trent D. Stephens explains in Dark Remedy: The Impact of Thalidomide and Its Revival as a Vital Medicine, the drug couldn’t kill a rat, so Grunenthal thought it had to be safe. The company developed a large marketing campaign that sold the drug as a sedative. They partnered with drug manufacturers in other countries and brought the drug to market all over the world. In most countries it was available without a prescription. Doctors were routinely given marketing confirming the drug’s safety—information that they passed on to their patients. Thalidomide was sold under various brand names, but with the assurance that it was like aspirin: mild, useful, and certainly not anything that could hurt you.

This, however, was not true. Slowly doctors in different countries began to notice physical impacts where the only thing the patients had in common was their use of thalidomide. First it was permanent nerve damage in the hands and feet, and then it was significant birth defects in the fetuses of otherwise healthy mothers.

The research began to come together showing that thalidomide caused deformities in gestation that were so severe that babies could often not survive after birth. If they did, it was with debilitating and painful abnormalities, most often the lack of properly formed limbs. Grunenthal doubled down on their marketing, casting suspicion on the medical articles discussing the dangers of thalidomide and attacking the credibility of the physicians who wrote them.8

The account describes years of resistance to acknowledging the harm and compensation claims settled out of court. It reports no criminal convictions of companies for selling a drug without verifying claims such as “safe for pregnant women.”9

One way to understand how this happened is to look at the story through the lens of incentives.

As you can imagine, there are financial incentives throughout the story of how thalidomide came to market. The heads of the team that discovered the drug were paid a percentage of the profits of its sales. And thalidomide was a big seller. There was a huge global market for sedatives, and in some countries it became the second-highest-selling drug.

There was also a significant financial incentive for drug companies to partner with Grunenthal. It meant that they could “avoid the cost of research, development, and testing.”10 And for Grunenthal itself, by not performing proper clinical trials on humans it saved a ton of money.

Thalidomide’s main competition as a sedative was a class of drugs called barbiturates. These had known undesirable side effects, and thus an incentive for doctors to recommend thalidomide was that it was purportedly safer than the alternative.

The drug companies regularly paid doctors for their endorsement, even going so far as to draft their lab articles.11 This created more incentives. The first financial incentive was payment, but then, once your name is linked with endorsement of a drug, you have an incentive in terms of maintaining your reputation by continuing to defend that drug.

In the later criminal trials and civil suits, a lot of drug company executives pointed out that their legal and moral responsibility was to their shareholders. The incentives to keep their shareholders happy in the form of large dividends were far more significant than fairly compensating victims.

One important part of this episode is that thalidomide was not approved for the proposed sale in the United States at the time. The Richardson-Merrell company had obtained the rights to distribute to the American market and so sought to obtain the approval of the Food and Drug Administration. The drug approval process in 1960 differed from the later system. The FDA’s requirement for evidence of safety before marketing came from legislation enacted in 1938 after another pharmaceutical tragedy; the general requirement to demonstrate efficacy came later.12 But their efforts were often focused on drugs after they hit the market, with pre-market testing and evaluation less regulated and vulnerable to manipulation.

Unfortunately for Richardson-Merrell, and fortunately for the rest of the country, Dr. Frances Kelsey* had just begun to work there. Having spent many years as a researcher, she was deeply concerned about the lack of evidence for the safety of the drug.13 Neither Richardson-Merrell nor Grunenthal could provide any documentation to back up their claims of safety for both mother and fetus during pregnancy. Dr. Kelsey determined that there was no way that they had done the research to know that thalidomide did not cross the placental barrier. She stalled approval as long as she could, sending back their application multiple times for being incomplete. She later explained that she had a very low incentive for approving the drug, as it was not a lifesaving one. The delay tactics worked, and she was able to prevent commercial approval of thalidomide while reports of its horrible effects were so widespread that countries began to take it off the shelves.14

« An incentive is a bullet, a key: an often-tiny object with astonishing power to change a situation. »

Steven D. Levitt and Stephen J. Dubner15

How do incentives work to make people seemingly lose their minds? To make normal people double down on their immoral decisions? In light of how numbers of drug company executives and doctors never admitted that thalidomide caused such horrible effects, it is worth explaining more about the power of incentives.

The initial financial component is easy to understand. Money is desirable, and therefore it is a common incentive that modifies behavior everywhere. But money alone isn’t often enough. Undoubtedly if you ask people whether they would harm thousands of children for money, most would say no. Therefore, in order to understand the real power of incentives you have to look at the psychological condition they create.

Humans don’t like cognitive dissonance—“the state of tension that occurs whenever a person holds two cognitions (ideas, attitudes, beliefs, opinions) that are psychologically inconsistent.”16 We engage in self-justifying rhetoric to reduce this dissonance. In the case of incentives for endorsements, the often subconscious thought process might go like this: “I am a good person. I supported this drug. A good person would not support a drug that is harmful. So the drug can’t be harmful because I am a good person.” We don’t see these connections laid out, and so what is produced is the opinion the drug is good without having a window into the mental gymnastics that got us there.

Carol Tavris and Elliot Aronson have written about the manifestation of cognitive dissonance due to incentives in the medical profession. “Physicians, like scientists, want to believe their integrity cannot be compromised. Yet every time physicians accept a few dollars or other incentives for performing certain tests and procedures, for channeling some of their patients into clinical trials, or for prescribing a new, expensive drug that is not better or safer than an older one, they are balancing their patients’ welfare against their own concerns.”17

Consider the story of Andrew Wakefield, the lead author on the paper published in The Lancet that claimed it had found a positive correlation between vaccines and autism. Later debunked, but not before causing a widespread drop in vaccinations that has resulted in needless deaths, the article was retracted. He did not sign the retraction. Why?

He had “a conflict of interest that he had failed to disclose to the journal: he was conducting research on behalf of lawyers representing parents of autistic children. Wakefield* had been paid more than eight hundred thousand dollars to determine whether there were grounds for pursuing legal action.”18 Tavris and Aronson demonstrate that “unlike truly independent scientists, however, he had no incentive to look for disconfirming evidence of a correlation between vaccines and autism, and every incentive to overlook other explanations.”19

This is why understanding the power of incentives is both critical and tricky. Accepting an incentive can make us invested in defending the decision that brought it, especially when admitting a mistake would threaten our reputation or self-image.

The greater danger to the public comes from the self-justifications of well-intentioned scientists and physicians who, because of their need to reduce dissonance, truly believe themselves to be above the influence of their corporate investors. Yet, like a plant turning toward the sun, they turn toward the interests of their sponsors without even being aware they are doing so.20

If we’re not aware of the incentives that direct our behavior, we can end up doing things we might prefer not to. It’s also important to be aware of how other people can incentivize us to do things that go against our wishes and values—much like those who sold thalidomide for the powerful incentive of money. By being aware of the incentives that may be directing our actions, we can recognize any unsavory influences and refocus by considering what we are really valuing. We may be driven by rewards like money, attention, and power, but we may actually value things like making a positive impact on the world, kindness, and honesty. We need not blindly follow incentives if they don’t align with our values.