Manufacturing depression

by Gary Greenberg

  Doctors who give their version of this spiel today probably have never heard of Frank Ayd. Nor have the millions of patients who have listened to it. But his impact, at least to judge from the swollen ranks of the depressed, has been immeasurable.

  Neither have most doctors or patients heard of Dan Fellowes, not even by his real name. But it was his idea to send Recognizing the Depressed Patient to fifty thousand general practitioners. And, as if playing Gideon with Ayd’s bible of depression wasn’t enough, he also got into the popular music business. “I found a musicologist, Leonard Feather, who compiled an album of blues songs,” he said. “It was the most beautiful expression of how life and the problems of life create depression.”

  Symposium in Blues, which was released in 1966, had all the trappings of a commercial entertainment—a major label (RCA), star performers (Louis Armstrong, Ethel Waters, Leadbelly), songs like “I Been Treated Wrong” and “I’m on My Last Go-Round,” and, in case the point wasn’t clear, Feather’s symptom-targeted liner notes (“songs that reflected anxiety and agitation, guilt, hunger or loss of appetite…”). The album wasn’t available in stores, however. Fellowes explained to Martin:

  We gave this to doctors. “Here’s a phonograph record for you.” They took it home…The music didn’t say “Elavil” at all. And there was no intent to give them a lecture. The liner notes talked about the fact that the songs came out during the Depression…It was talking about the psyche of the American public.

  The package insert for Elavil was slipped into the dust jacket of Symposium in Blues, right next to the record sleeve, and when a doctor opened the album, he was greeted with the Elavil logo, along with a list of the symptoms of depression. But the album was more than a way to put doctors in the right mood while they read Recognizing the Depressed Patient. It was also one of the earliest instances of viral marketing, the kind in which the advertiser lets loose a message in a dim corner of the culture, where it can circumvent whatever immunity we have to the pitchman’s manipulations, attach itself to reputable sources of knowledge, and replicate.

  What better host for this kind of virus than the doctor? What better culture than one that has not yet begun to suspect the motives of drug companies and doctors or to worry about their power? Patients surely didn’t know about the record or the book, let alone the phalanx of detailers that eventually included one sales rep for every eight doctors in America. They never saw the medical journals, chock full of ads using the latest Madison Avenue techniques to urge doctors to dispense the latest drugs, or attended their industry-funded seminars. A doctor didn’t mention to a patient, as he used his Merck pen to write a prescription for a Merck drug, that he’d learned about the disease at a conference at a luxury hotel in the tropics that was sponsored by Merck, or that the free sample he was handing to you had been given to him by the detailer who took him to lunch the other day. He himself may not have known that the journal article the detailer gave him touting the drug’s virtues was written by a researcher getting paid by its manufacturer, who had been forbidden to publish his results if they didn’t meet the company’s approval.

  Despite all these advantages, and despite the best efforts of men like Dan Fellowes, the minor tranquilizers were still beating the antidepressants by a mile. But even before Robert Spitzer and his committee reinvented depression, Elavil and Tofranil and all the rest caught a huge break. In 1968, not long after he arrived in office Richard Nixon made pharmacological Calvinism a matter of national policy. He declared war on drugs, or to be more accurate, he declared war on the users of some drugs—largely LSD and marijuana, which evidently threatened America’s moral fiber as surely as Communism had when Nixon was a senator and vice president.

  As the drug war machine cranked into action, minor tranquilizers got caught in the crossfire. Vice President Spiro Agnew warned publicly about the dangers of “mood drugs” in 1970, on which, he said, “over one-half million citizens are now dependent.” A prominent doctor, in a comparison that probably left drug makers reaching for the Valium, said that benzodiazepines were not much different from “other products which also affect the mind but which we do not label medicine…for example, alcohol, tobacco, coffee, tea, and marijuana.” The National Institute on Drug Abuse reported in 1976 that Valium was one of “the most frequently abused drugs in the United States.” In 1979, Edward Kennedy, chairman of the Senate Subcommittee on Health and Scientific Research, and a man who knew a bit about excess, opened a hearing on the benzodiazepines by declaring that “these drugs have produced a nightmare of dependence and addiction…thousands of Americans are hooked and do not know it.”

  Whether people were really hooked on Valium remained a matter of debate, but that was never the real problem in the first place. Psychotropic hedonism was. Addiction was just a code word for “drugs that you should be afraid of because they make you feel so good.” (Indeed, given the fact that Librium and Valium alone were what Fortune called “the greatest commercial success in the history of prescription drugs,” it is a safe bet that at least some of the people excoriating them knew this firsthand.) It was not difficult for drug warriors to convince the Drug Enforcement Administration to put minor tranquilizers on its list of controlled substances. Doctors prescribing them would now be under the jurisdiction of the DEA, with all the paperwork that entailed—and all the Dr. Feelgood implications. Patients would have to worry about the stigma of taking a drug that professional scolds were lumping in with marijuana and alcohol. To judge from the falling sales figures—not as sudden or severe as the decline of the “psychic energizers” in the early 1960s had been, but noticeable by the early 1980s—patients and doctors alike were thinking twice about the benzodiazepines, their market dominance a victim of America’s confusion about drug-induced pleasure.

  The drug warriors’ revival of pharmacological Calvinism hit its stride at about the same time that the American Psychiatric Association was resurrecting Kraepelin’s categorical disease entities, which in turn coincided with the FDA’s effort to implement Kefauver-Harris by approving only drugs that were effective with specific diseases. Antidepressants—drugs that didn’t get you high like Valium did, that treated the newly fashioned major depressive disorder, and that allegedly did so by attacking depression at its biochemical source—had hit the trifecta.

  Still, their sales continued to languish. In 1970, they had constituted only 2 percent of the prescriptions written for psychotropic drugs, and even when the new DSM came out and minor tranquilizers were falling into disfavor, they remained a minor player. One more element was needed, and, unknown to even the most savvy marketers, it was already on the way—and had been since 1968.

  That was the year that Arvid Carlsson, the Swedish scientist who had discovered the role of dopamine in Parkinson’s disease and then proved the principle of chemical neurotransmission to the world, began work in earnest on an idea he’d been hatching for a few years. While he believed that Joseph Schildkraut was correct in asserting that depression was a neurotransmitter problem, he thought that Schildkraut had made a mistake by singling out dopamine and norepinephrine as the culprits. Those may have been the chemicals most affected by the tricyclic antidepressants, but, Carlsson pointed out, all the members of that family also blocked serotonin reuptake. And a close look at the statistics showed that the strongly serotonergic drugs were the ones that improved mood—as opposed to the other items on the HAM-D—most dramatically. A drug that targeted serotonin, Carlsson thought, might be the key to a better antidepressant.

  Carlsson’s suggestion that psychopharmacology return to its pioneering discovery—and to the theory that serotonin was the key to mental health and illness—fell on deaf ears until he finally convinced the Swedish firm Astra Pharmaceuticals to finance his research. By 1971, he and his team had come up with zimelidine (Zelmid), the first drug designed specifically to inhibit serotonin reuptake. A year later, a team at Eli Lilly synthesized its own serotonin-specific drug: 3-(p-trifluoromethylphenoxy)-N-
methyl-3-phenylpropylamine.

  When the paper announcing Lilly’s drug was published, the company was still calling it LY-110140; it hadn’t bothered to name the compound yet because no one was quite sure of its commercial value. Serotonin, after all, is distributed throughout the body and plays a role in digestion, appetite, sleep, and blood pressure, among other functions, so 110140 might have had many uses. It might even have been used only for research purposes, to assay or manipulate or otherwise investigate serotonin.

  Much as Abbott had done with serotonin itself in the early 1950s, the company solicited opinions from leading researchers about the new drug’s possible uses. Arvid Carlsson, who was at one of the meetings, told me that when one scientist suggested that Lilly might try the drug out as an antidepressant, the company representative replied that this wasn’t in the company’s game plan.

  But in the meantime, Astra, spurred on by Carlsson, was taking zimelidine through clinical trials in Sweden. In 1981, the company received approval to market Zelmid as an antidepressant throughout Europe. The drug did well enough that Merck licensed it from Astra and applied to the FDA for approval in the United States. But in 1983, just as the process was getting under way, trouble arose in Europe: zimelidine syndrome, a flu-like condition, and, more ominously, an outbreak of Guillain-Barré syndrome, a sometimes fatal neurological disorder, among users. Astra withdrew the drug in September 1983, and Merck abandoned its application with the FDA soon after. According to Carlsson, neither company thought the population of depressed people would ever be big enough to justify the expensive research necessary to investigate the link between the drugs and the illnesses.

  But the brief success of Zelmid had caught Lilly’s eye. By 1984, the company was finally interested enough in 110140’s antidepressant qualities to give it a name—fluoxetine—and to conduct and publish research showing that the drug relieved depression as well as the tricyclics, and with fewer side effects. By the end of 1987, the FDA approved the compound, now known as Prozac, as an antidepressant. Zoloft, Pfizer’s entry into the new market, was approved in 1992, and SmithKline Beecham (successor to Smith Kline and French) introduced Paxil a year later.

  SmithKline also introduced the abbreviation SSRI to the marketplace, underlining the new drugs’ major claim to superiority: that they were selective, targeted precisely at serotonin. Marketers had already jumped on this distinction. The first Prozac ad in JAMA, in 1988, touted it as “the first highly specific, highly potent blocker of serotonin uptake,” and throughout the 1990s, the industry put selectivity at the heart of its campaign. Antidepressants, the journal ads proclaimed, were clean drugs, strong and effective drugs, high-tech, can’t-fail magic bullets that destroyed depression without making patients high, without addicting them—and indeed, without causing any collateral damage.

  With a come-on like this, it hardly mattered that none of this is exactly true. Not only do the drugs perform poorly in trials, but while they do bind to serotonin receptors at higher rates than they bind to other receptors, and at higher concentrations than the tricyclics do, they by no means bind only to serotonin sites. They are active all over the brain, so while they may not cause as many side effects as the tricyclics, they still cause so much discomfort that there is a cottage industry devoted to reducing nonadherence among SSRI takers. Patients, researchers have found, were reluctant to take psychiatric drugs in the first place, and when they start feeling jittery and agitated, or when they can’t sleep and have upsetting dreams when they do, or when they get constipated or nauseated, or when they hear about the reports linking antidepressants to suicide and violence, and above all else, when they find that they suddenly can’t reach orgasm or don’t want sex at all, they often just stop. Indeed, nearly 70 percent of people stop taking antidepressants within the first month.

  None of this—the lousy sexual performance, the lousy clinical performance, the lousy side-effect profile—is a secret anymore, if it ever was. A search of PubMed, the NIH database of published research, turns up 744 papers on the subject of sexual side effects of antidepressants alone. The data used by the FDA to approve the drugs, including the ones in which the drugs didn’t work, are in the public domain. Even the head of the agency’s division of neuro-pharmacological drugs wondered, when it came time to approve Celexa in 1998, if its clinical trial results showed any “clinical value,” only to be told by a colleague that because “similar findings for…other recently approved antidepressants have been considered sufficient,” they had no choice but to go ahead. The agency also knew that reports linking SSRIs to the increased risk of suicide and violent behavior had begun to surface within a year of Prozac’s emergence on the market. Still, by 2006, antidepressants had become the most commonly prescribed class of drugs in the United States, at an annual cost of $13.5 billion.

  This dramatic success depends on the old tricks—downplaying side effects and overstating efficacy in marketing campaigns directed at prescribers. But it also hinges, at least sometimes, on outright lies. Psychologist Glen Spielmans and his team analyzed a group of ads from leading psychiatric and general medical journals. They discovered that in more than one third of the cases, the sources cited in the ads failed to verify the claim they were supposed to support. And that’s when the companies bothered to mention a source. Fully half the time, they didn’t even do that—or they cited a source that couldn’t be obtained. When Spielmans asked Wyeth for the data cited in an Effexor ad, the company responded, “Unfortunately, our internal policies do not allow for distribution of unpublished data.” As Spielmans pointed out, this is ironic given the tag line of the ad: “See depression, see the data, see a difference.”

  Doctors haven’t been keen on investigating how gullible their colleagues are, so it is hard to know just exactly how much of a role these lies play in doctors’ prescribing habits. We do know that 80 percent of “high prescribers” rated journals as “an important source of information” and that two-thirds of doctors overall are exposed to drug ads on a weekly basis. And we know that an investment of one dollar in professional advertising yields a five-dollar return in sales. Doctors, in other words, seem to respond to ads directed at them by writing more prescriptions.

  But even the shrewdest ad will not work if the market doesn’t exist. Before a doctor writes a prescription, especially for a psychotropic drug, he has to believe that the patient is suffering from an illness that can be treated. Which is where the ad men, with a little bit of inadvertent help from the APA and the FDA and the drug czar, found their sweet spot. Because after the long haul through medical school, where doctors learned to think of diseases as targets and drugs as magic bullets, after practicing in a society that expects and even demands results in return for its deference to doctors, after reading in the Journal of the American Medical Association that depression is the second leading cause of disability in the world, but that only 10 percent of the 15 or 20 million citizens who will be depressed in any one year are getting treatment, and that this undertreatment is costing society about $43 billion per year (and that was in 1996, when a billion dollars was real money), after learning, in short, that depression is a public health emergency and that “safe, effective, and economical treatments are available,” treatments that work as precisely as antibiotics or insulin, that are being talked about by all their colleagues, and taken by their friends and their patients and maybe even themselves—after all that, it’s hard to understand why doctors wouldn’t render the diagnosis and reach for the prescription pad. And it’s easy to see why they were four times more likely in 2005 to prescribe drugs for the treatment of depression as they had been in 1987. Not because they were in cahoots with the drug companies, not for any other reason than that they wanted their patients to feel better and they really believed they finally had something to offer.

  Even without blocking the uptake of a single molecule of serotonin, the drugs begin to work their magic. Doctors—primed by the ads, the detailers, the enthusiastic (if industry-mani
pulated) articles by their peers—were ready to give their patients Frank Ayd’s pitch: that they had a disease, that it was no different from any other disease (except perhaps for how widespread it is), and that the cure was waiting for them at the pharmacy. They were ready, in other words, to change the setting in which antidepressant use takes place, to name their patients’ pain and create expectation for its cure, to mobilize, whether or not they meant to, the placebo effect.

  And it wasn’t only the doctors. The drug industry also had help from writers like Peter Kramer, from a news media more than willing to report breathlessly on the new wonder drug (within a couple of years of its introduction, Prozac was featured on the covers of both Newsweek and Time), from an FDA hamstrung by its own lax standards, and, perhaps above all else, from millions of satisfied customers.

  For four glorious years, Lilly had a monopoly on this emerging market, and the company made the most of it—the drug’s $1.5 billion in sales in 1992 accounted for one-quarter of Lilly’s revenue that year. SSRI competition finally emerged, first with Pfizer’s Zoloft, introduced in 1992, and then, the next year, with Paxil. Some of the competition was fierce. Pfizer, for instance, undercut Prozac’s price by 20 percent and deployed its detailers on 660,000 annual Zoloft visits, with marching orders to harp on Prozac’s side effects—especially anxiety, which Pfizer claimed was less pronounced in Zoloft patients. Ads for Zoloft in professional journals turned anxiety into a code word for suicide, reminding doctors of the lawsuits that Lilly had been fending off for years. The success of the whispering campaign quickly showed up on the bottom line: by 1995, Pfizer had captured nearly one-third of the overall antidepressant market, ringing up $900 million in sales. That same year, SmithKline raked in nearly a half-billion dollars from Paxil. Lilly was still on top of the heap at $1.47 billion (capturing nearly 41 percent of the 1995 market), but, as the Wall Street Journal reported in mid-1996, “the king of antidepressants [was] under attack.”