By James Davies
Reprinted from Cracked: The Unhappy Truth About Psychiatry by James Davies. Copyright Pegasus Books © 2013
In May 1993, a mental disorder that had been in the DSM-III was repackaged, renamed, and given new life in DSM-IV. It was called Premenstrual Dysphoric Disorder and was listed in the DSM as a mental disorder. Up to 8 percent of women were said to suffer from the condition. And the condition was apparently nasty. Its main symptoms occurred two weeks before menstruation and included feelings of fatigue, anxiety, emotional instability, disinterest in daily activities, and difficulty in concentrating. In short, Premenstrual Dysphoric Disorder was first presented as if it were an exaggerated from of PMT – premenstrual tension.
By the early 2000s, the number of women in the United States being diagnosed with PMDD was rising exponentially. And there are some obvious reasons for this. The first was that in 1998 the FDA in the United States recognized the condition as an official mental disorder. This freed up doctors to start diagnosing PMDD when previously they had no disorder category to which they could match premenstrual experiences. Second, the pharmaceutical industry now had the green light to market the disorder and its cure. Eli Lilly was the first to step up to the plate by spending $30 million on advertising its chemical cure.
The following advert by the company is illustrative of the type of commercial with which it flooded the airwaves once FDA approval was granted. It shows a woman who has lost her keys growing increasingly frustrated. The voiceover breaks through:
Think it’s PMS? It could be PMDD—premenstrual dysphoric disorder. You know, those intense moods and physical symptoms the week before your period. Sound familiar? Call to get free information about PMDD and a treatment your doctor has to relieve its symptoms. Why put up with this another month?
Alongside such direct-to-consumer pharmaceutical adverting, Eli Lilly launched a marketing campaign targeting psychiatrists, gynecologists, and mental health providers, who were soon all receiving promotional materials: flyers, free samples, invitations to Lilly talks, and unsolicited visits from pharma reps. As the word spread in the medical community, diagnoses of PMDD quickly soared, and so did the prescriptions.
“I really am not myself at the moment. I feel so anxious and confused all the time. I just feel, well, different. It started about six months ago. I began to get these god-awful cramps before my period. It was like my guts were digesting acid. I also got these piercing headaches and I’d feel emotionally all over the place. I went to my local doctor and he sent me to a gynecologist. The gynecologist said he couldn’t find anything physically wrong with me and said I was probably suffering from a psychological disorder I’d never heard of before—Premenstrual Dysphoric Disorder—and he said he could help.”
Let me fill the gaps in Sarah’s knowledge.
The pharmaceutical giant Eli Lilly makes Sarafem. Its name is a rework of “seraphim,” a Hebrew word meaning “angel,” a word with obvious female overtones. Its packaging also conjures up stereotypical female associations. The pill is encased in a pretty pink-and-lavender shell, and is heralded by Lilly as a wonder cure for this distinctly female premenstrual disorder. So far so good.
Now here comes the interesting bit. What Eli Lilly initially concealed from the millions of women taking the pill is that the pill is actually Prozac. Chemically, Sarafem and Prozac are exactly the same. The only difference between them is that their names and packaging are different. Sarah, like thousands of other women up and down the country, was taking Prozac and didn’t know it.
There are a lot of possible interpretations for why Eli Lilly engaged in what you or I may be tempted to see as corporate deception. The first is that it obviously saved the company a great deal of money. It is cheaper to repackage existing pills than it is to develop new ones. In addition, Eli Lilly’s patent protections on Prozac were running out a year after Sarafem would be released, so marketing Prozac under the new trade name would effectively extend patent protections for many more years. Money matters.
But surface appearances also matter. Women, like Sarah, are more likely to feel comfortable consuming pink pills than they are blue pills, because of the associations attached to the different colors. As an article on drug marketing in the Boston Globe said at the time, “Drug designers propose colors for a particular medicine and help make sure there are no symbolic mistakes.” An example of a symbolic mistake would be making a pink Viagra, or a menstruation pill that is dark red. Symbolic mistakes occur when the color, shape, or name of a pill does not resonate with its particular consumer. There has to be a positive correspondence between consumer and product to maximize the sales effect—or so the rationale goes.
A final interpretation for Prozac’s rebranding was because Prozac is associated primarily with depression, and women labeled with PMDD are not depressed, just as the patient Sarah wasn’t. So a new name was required to effectively strip from Prozac associations that were potentially undesirable to this new consumer group. Laura Miller, a marketing associate for Eli Lilly, put it this way:
[Women] wanted a treatment with its own identity. Women do not look at their symptoms as a depression, and PMDD is not depression but a separate clinical entity. Prozac is one of the more famous pharmaceutical trademarks and is closely associated with depression.
By rebranding Prozac as Sarafem, Eli Lilly divided the one chemical into two separate pills for two different disorders. One pill continued to be marketed as an antidepressant. The other they marketed as a so-called premenstrual corrective. A new pill was born not because a new chemical had been found, but because a popular brand had been changed.
“Eli Lilly is not alone in rebranding pills,” Nathan Greenslit, a young and dynamic professor at MIT’s prestigious Science, Technology, and Society Program, told me. As we exchanged messages about Sarafem online from our respective offices in London and Boston, Greenslit, who has spent years studying the pharmaceutical industry, revealed that in 1997 the FDA approved GlaxoSmith-Kline’s antidepressant Wellbutrin as a smoking cessation pill. But, again, because Wellbutrin was too well known as an antidepressant to be publically accepted as an antismoking pill, Wellbutrin was rebranded as Zyban. GlaxoSmithKline marketed them as separate drugs that targeted separate and discrete disorders: Wellbutrin targeted depression, and Zyban nicotine addiction.
The website for Zyban puts it this way: “Zyban is a nicotine-free pill. Not a patch. Not a gum.” Under the heading “Zyban: Helping Smokers Quit Neurochemically,” the site claims that “while it is unclear exactly how Zyban works, it is thought to act on the part of your brain that is addicted to the ingredients in cigarettes.”
“As was the case with Sarafem and Prozac,” Greenslit wrote in an article, “the patient-directed information does not admit that Wellbutrin and Zyban are the same drug.” By giving the old drugs a makeover, they could treat complaints other than those they were designed to treat originally.
What is interesting about these interpretations as to why the rebranding of Prozac occurred is that each is so entirely plausible that it is tempting for you and me to look no further for explanations. The name was changed for reasons of marketing and money—simple as that. But what if it isn’t that simple? What if something else is going on here, something far trickier to understand, but something essential to understand to complete the picture? Hold that question in mind.
* * *
How can it be that a sugar pill with no active chemical properties can lighten your mood or decrease your anxiety? The time has come to resolve that mystery.
In the 1980s, the National Institute of Mental Health sponsored a fascinating experiment. It set out to evaluate the effectiveness of antidepressants and psychotherapy in the treatment of depression. The experiment was actually very simple. It asked each patient before treatment began the following question: “What is likely to happen as a result of your treatment?” After the researchers calculated the results, a startling conclusion emerged: the answer the patient gave to this question predicted his or her therapeutic outcome.
In other words, those who expected to feel better improved the most, while those who expected little or no improvement received the least benefit from treatment. Furthermore, this result was the same regardless of whether the patient had been treated with antidepressant medication, psychotherapy, or a placebo. In short, whatever the therapy patients were given, the patients with positive expectations improved most. The conclusion: expectations matter. But if expectations matter, is there anything that can be done to increase a patient’s expectation of recovery and, in turn, his or her therapeutic outcome? The simple answer is yes. To understand what can be done, imagine yourself in the following scenario:
You are sitting in a lecture room listening to a presentation. But you are finding it difficult to concentrate because you have a throbbing headache. Now imagine that the man sitting next to you notices that you are in pain and offers you a small, white pill, quietly telling you the pill will help you feel better. After some deliberation you decide to take the pill (perhaps unwisely) and see what happens. And after a while you notice, disappointingly, that nothing happens at all. So you turn to the man at the end and ask what he gave you. He replies he gave you a sugar pill.
Surely we now have an explanation for why the pill did not work—it was made of sugar. That would be the obvious answer. But the trouble with obvious answers is that they tend to lead us astray. For example, we know from placebo studies that the probable reason the pill did not work will have less to do with its being a sugar pill than with where it was given to you and by whom. You see, the man next to you was not a medical doctor, and the lecture room is not a hospital, and these contextual factors are significant enough to affect whether the pill will work.
For example, if you were to meet the same man in a hospital and believe him to be a doctor, a curious thing will happen if he gives you the same pill: its effectiveness may increase by up to 40 percent. It is the same man and the same sugar pill, but the effects are now dramatically different. And they are different because the contextual meanings have changed. You are in a hospital and you believe this person to be a doctor, and so you now expect the pill to work. And because you expect the pill to work, it is more likely to do so.
The point of the above illustration is to show that cultural meanings matter. They have the power to increase or decrease, almost subliminally, our expectations for recovery and therefore the extent to which we actually improve.
But how does this insight help us deepen our understanding of why Eli Lilly repackaged Prozac? How could this research be used to explain to Sarah why the pills she was taking were simply a glammed-up version of Prozac? Just recall for a moment that key to the repackaging process was altering the color of the pill. There’s our clue. Prozac is green and white, while Sarafem is pink and lavender. So maybe there is something about the color of a pill that impinges not only upon its attractiveness but also upon its effectiveness? Could the very color of a pill help it work?
An early experiment published in the Lancet explored this very question. What the group of researchers did was gather up 56 medical students and hand each of them a package containing red and blue pills. All the students were told was that one color represented a tranquilizer and the other a stimulant. After taking the pills, the students were then asked which set of pills were the stimulants and which the tranquilizers. The majority concluded that the blue pills were the tranquilizers and the red pills the stimulants. They reached this conclusion because when taking the blue pills they felt far less alert and much drowsier than when taking the red pills. The researchers then told the students that, chemically, the blue and red pills were exactly the same—both sets of pills were made entirely of sugar.
But how could dummy pills yield such different effects? The answer is once again found in recognizing the power of cultural meanings. For the medical students, the colors red and blue each held very different meanings: red means “hot,” “up,” and “danger” (meanings fitting stimulated behavior), while blue means “down,” “calm,” “cool,” and “quiet” (meanings fitting tranquilized behavior). The meanings attached to the different colored pills affected the drug’s perceived action and effectiveness.
Scientists like Dr. Alan Branthwaite are not surprised by the power of placebo effects. I spoke to Dr. Branthwaite on a Tuesday morning in early February 2012. His manner was scholarly and cautious, but as soon as I raised the topic of placebo effects he became animated. “These effects are staggering when you first encounter them, and they still stagger me today,” he said. “The human body is so fundamentally linked to the mind that if you can spark the right mental associations, then the body responds, sometimes in dramatic and unexpected ways.”
Branthwaite won the medical community’s acclaim when he published a paper in the British Medical Journal on placebo effects. What his study set out to discover was whether the presence or absence of a trusted brand name on a pill can affect recovery. “So we devised a neat experiment,” he said enthusiastically. “We gathered up about 835 women who regularly use painkillers for headaches and then randomly assigned them to one of four groups. And we gave each group a different pill. The first group had aspirin labeled with a popular brand name; the second group had the same aspirin but with no brand name; the third group had a sugar pill marked with the popular brand name; and the fourth group had an unmarked sugar pill. The question was which group would improve most?”
What Branthwaite and his team were expecting to find was a small increase in the effectiveness of the branded pills. “But what shocked us entirely,” he continued, “was the extent of that difference. The group given the branded placebos improved almost 20 percent more than those with the unbranded placebos, while branded aspirins also worked significantly better than unbranded ones. So here we had striking evidence that the presence of a trusted brand name can dramatically improve a pill’s efficacy, even if that pill is completely inert.”
Taking these two studies together shows how surface things that may seem inconsequential to you and me (the color or brand name of a pill) are crucially important to the healing process. Subliminally they play with our expectations for recovery, and expectations, in fear of repeating myself, can dramatically affect outcomes.
* * *
To try to get to the bottom of how these meaning effects work, I spoke to Daniel Moerman, distinguished professor of anthropology at the University of Michigan. Moerman has dedicated much of his professional life to investigating how cultural meanings affect bodily functioning, and is widely known in academia as a world leader in placebo research.
I suggested to Moerman how odd it is that the meanings we ascribe to a pill can sometimes be more powerful than its active substance, especially in the realm of psychopharmacology.
“Well, James, you’re an anthropologist, right? You know the power of meaning! Every culture has its symbols and objects of veneration and it is no different with us. Once, for us, we revered crosses and statues of the Virgin Mary, but now pills and stethoscopes capture our worship. So even an inert pill can affect us because it has shape and form and a context, and it has language attached. It comes in a blue box or a pink box, it’s taken in a pharmacy, doctor’s room, or hospital with all the panoply of a thousand years of medical tradition behind it to give it overwhelming symbolic weight.”
One of Moerman’s recent studies shows how the power of medical sanction should not be underestimated. Medical approval is crucial to all of us, even the most critical among us. What Moerman did was gather up 117 studies of ulcer drugs published between 1976 and 1994 to discover what drugs worked best. As the results came in, he realized something astonishing.
In 1975 a new ulcer drug called Cimetidine was released. It enjoyed excellent clinical success, eradicating, on average, 80 percent of ulcers. As time passed, however, the drug’s efficacy strangely became lower and lower, until today Cimetidine can only claim to eradicate 50 percent of ulcers. So what had happened to the pill? Moerman explains the deterioration by pointing out that five years after Cimetidine’s release, a new drug called Ranitidine arrived. This competing drug was considered superior to Cimetidine, and as the new drug’s popularity grew, the old drug’s effectiveness declined. The correlation was staggering.
Of course, there are different interpretations of this (questions of changes in methods, etc.), but the most compelling, and to Moerman the most plausible, is that Cimetidine lost its power because as new and supposedly superior drugs arrived, Cimetidine ceased to be thought of as a superior drug. Belief in Cimetidine had waned, and with it so did its clinical effectiveness. “So it’s clear,” said Moerman animatedly, “meaning matters!”
* * *
Let’s now return to our original question. We know that Sarah was unaware that the drug she was taking was Prozac. She thought that Sarafem was a specific pill developed for her specific “problem.” Sarah was wrong. What Sarah also did not know was that Prozac was rebranded as Sarafem more for financial than scientific reasons.
But the question still remains about whether Prozac was rebranded for other reasons too. Was it rebranded because the company knew that Prozac would provoke only a small placebo effect in women suffering from so-called PMDD? Did Lilly turn Prozac into a distinctly female pill to achieve a higher placebo effect in this new patient group?
In order to answer this question, we first have to answer a broader one: Does Big Pharma more generally make use of placebo studies to manipulate higher placebo effects in patients? I put this question to Moerman.
“In my experience, people in the pharmaceutical industry are either incredibly good actors or are remarkably dense. They are good actors if they make use of these studies but pretend not to, and they are dense if they don’t make use of them at all.”
So, to use Moerman’s words, is the pharmaceutical industry led by actors or dunces? I decided to cut to the chase and find out for myself. I contacted Eli Lilly’s headquarters in the UK and asked whether placebo studies are consciously used to create higher placebo effects in their consumers. The head of corporate affairs reluctantly responded after some prompting: “I forwarded your enquiry to colleagues in our Global HQ, as only they would be in a position to answer. Due to competing priorities, they are unable to provide a response.”
Competing priorities? What did that mean? I politely wrote back asking Eli Lilly for some clarification. They wrote back curtly: “There is no Lilly response. Your questions are very specific and the person or people who might be able to answer them have chosen to decline on this occasion.”
So now you understand the problem. Questions like mine are rarely answered by the industry because there are just some things companies don’t want you or me to know. Pharmaceutical companies are notoriously secretive. They have a history of not only concealing how pills are developed and marketed but of concealing negative trials that show their drugs in a bad light . So in the absence of corporate transparency, all we can rely on is indirect evidence to answer the actor/dunce question. Let’s look at some of that evidence now.
We know that companies regularly pay academics to discuss their work with company employees and executives. We also know these companies have conducted their own in-house studies and data mining, and have in recent years funded the work of leading placebo scientists like Ted Kaptchuk and Fabrizio Benedetti (even Branthwaite’s study of brand names I spoke of earlier was pharmaceutically funded). Furthermore, American journalist Steve Silberman recently exposed how a massive study of placebo effects (undertaken by the Foundation for the National Institute of Health) is being funded by companies including Merck, Eli Lilly, Pfizer, AstraZeneca, and GlaxoSmithKline. “In typically secretive industry fashion,” Silberman told me during our interview, “the existence of the project itself is being kept under wraps. FNIH staffers are willing to talk about it only anonymously, concerned about offending the companies paying for it.”
During our conversation, Silberman also mentioned that a few years ago when the Public Library of Science hosted a debate about whether the pharmaceutical industry should be allowed to continue advertising antidepressants to the general public, Randall S. Stafford, a senior consultant with Merck, GlaxoSmithKline, Bayer, and Procter & Gamble, argued that banning the ads might result in an abrupt reduction of drug effectiveness by reducing the placebo effect en masse.
“This was an astonishing and very rare admission,” Silberman said to me, “because the pharmaceutical industry won’t publically acknowledge that the placebo response is giving a boost to the drugs they make. But here we had a major consultant openly declaring that the adverts are all about generating the expectations that help increase placebo effects, and that if you cut back on the marketing, the pills’ effectiveness will dramatically decrease.”
Furthermore, again and again we find striking correspondences between what placebo research tells us and how actual pills are being developed. These correspondences strongly suggest that companies are taking account of placebo research when developing and marketing their pills. For instance, recall for a moment the study I quoted earlier that showed that blue sugar pills create sedating effects and red sugar pills create stimulating effects, even when both pills are made of sugar.
A team in the Netherlands has recently researched the actual colors of the pills that you and I take. And it turns out that nearly 80 percent of all sedative pills were green, purple, blue, or white (green, purple, and blue are sedating colors) while of all the antidepressants (uppers) only 5 percent fell into the green, purple, or blue category, 40 percent into the white category, and all the rest were colored in “stimulating colors.” Is it just coincidence that companies are manufacturing pills that largely match the placebo research?
Without being allowed to observe these companies at work directly (trust me, I and many colleagues have requested access countless times, but the doors remain closed), the question of whether companies are manipulating placebo research must continue to hang in the air. Of course the indirect evidence strongly suggests that placebo research is now regularly manipulated, that pills like Prozac are not just altered for reasons of marketing and money but because the features of Prozac, its color, its name, and its associations, would not successfully evoke the placebo response in women labeled with PMDD. Did these companies believe that women required a different pill, with different hopes and expectations attached, to get the result they wanted?
I know what I think, but given the absence of direct evidence, I’ll just have to leave you to reach your own conclusion.
* * *
“Sarah, I have a question for you,” John says carefully. “Did you know that the drug you are taking is actually Prozac?”
Sarah’s head tilts sharply to one side. “Excuse me?”
“What I mean is, did you know that Sarafem and Prozac are chemically exactly the same?”
Sarah sits back for a moment, looking at John skeptically. Then a sudden wave of anxiety flashes across her face. She sits forward sharply. “You’re kidding me, aren’t you?”
“I’m afraid I am not, Sarah.”
“But . . . but that’s wrong, isn’t it?” said Sarah, looking distressed. “Why would they do that?”
“Well, Sarah, that requires a complicated answer.”
“But isn’t Prozac for depression? I don’t have depression.”
“No, you don’t, or at least you didn’t. The important thing is that since taking the pills you feel different. Perhaps we could spend a little time now thinking about precisely how you feel different. So before we do anything, I want to explore the role that the pills you are taking have played in how you feel now.”
“What, you mean that Sarafem may be responsible for why I feel different?”
“I don’t know,” John responded. “It’s possible. Even so, it is important for us to find out.”
“Jesus, I didn’t know.”
“The sad truth, Sarah, is that most people don’t . . .”