Manufacturing depression

by Gary Greenberg

  The news wasn’t really that bad, however. Even if Lilly was losing market share, its sales were still up by 18 percent—less than Zoloft’s 38 percent increase and Paxil’s 45, but still enough to account for nearly 30 percent of Lilly’s revenues. Success was breeding success, the market expanding by 33 percent every year. And when the Journal of the American Medical Association reported in early 1997 that still only 10 percent of the depressed were getting treatment, it seemed that the rising tide was capable of floating as many boats as the drug companies could launch.

  Supply was creating demand. It was enough to inspire exuberance in even the most rational executive. But a company selling a product whose patent expires in a few years doesn’t settle for divvying up the market, at least not if it can avoid it—and especially not if the National Depressive and Manic-Depressive Association declares a treatment gap that will only be closed when “patients [become] informed consumers and advocates.” Who better than the drug companies to inform them, and what better medium than advertising? Which is why in mid-1997, Eli Lilly hired the Leo Burnett Company, one of America’s leading advertising agencies, to launch an ad campaign for Prozac that would skip right over JAMA and the New England Journal of Medicine and head for Cosmopolitan and Reader’s Digest. After all, as Willie Sutton might have said, that’s where the consumers were.

  According to a Burnett vice president, “This is one of the most serious assignments we’ve ever had,” and its mission was clear: to inform readers that, as Mike Grossman, Burnett’s director of public relations, put it, depression “isn’t just feeling down. It’s a real illness with real causes.” Lilly spent $22 million in the last six months of 1997—nearly two-thirds of its entire advertising budget for the year—“assisting people in their depressed stupor,” as Grossman put it, “to raise their hand for help.”

  The first ad was a three-page spread: a drawing of a rain cloud over the caption “Depression Hurts,” a sun shining on the slogan “Prozac Can Help” (both images were crude, as if drawn by a third-grader), and, on the last page, the fine print about side effects. It turned out that you didn’t need to be in a depressed stupor at all, but merely under the weather, to have the “real illness,” that “doctors believe” may be caused by “an imbalance of serotonin in your body.” The copy under the cloud suggested, “You may have trouble sleeping. Feel unusually sad or irritable. Find it hard to concentrate. Lose your appetite. Lack energy. Or have trouble feeling pleasure.” And when people feel this way, the reader discovered, just before moving to the sunny side, “the medicine doctors now prescribe most often is Prozac.”

  One year and $47 million after launching the print campaign, Lilly and Burnett took Prozac to television with three sixty-second spots, including “Checklist.” Filmed in black and white, the ad is a montage of suffering and isolation—stills of a woman on a couch, another woman in bed, a man on a pier, another sitting in a restaurant, a woman at her desk, another showering, and finally just a sorrowful face, each person looking as empty and sad and hopeless as Lutherans in an Ingmar Bergman movie. Violins repeat a minor chord while a woman narrator, sounding not a little depressed herself, ties each image to a target symptom:

  Have you stopped doing the things you enjoy? Are you sleeping too much? Are you sleeping too little? Have you noticed a change in your appetite [at this point the man in the restaurant inexplicably vanishes, as if transfigured]? Is it hard to concentrate? Do you feel sad almost every day? Do you sometimes feel like life [pregnant pause] is not worth living? These can be signs of clinical depression, a real illness with real causes. But there is hope. You can get your life back. Treatment that’s worked for millions is available from your doctor.

  The commercial provides a toll-free number that you can call for a personal symptoms checklist that will make it “easier to talk with a doctor about how you’re feeling,” and ends with Lilly’s logo and the “Welcome Back” tagline.

  Other companies gave Lilly’s campaign the same flattery they had given Prozac: they imitated it. Zoloft and Paxil soon had ads and websites of their own, and by 2000 companies were spending $128.5 million to advertise their antidepressants to consumers. Laboring under FDA guidelines, which, among other restrictions, required them to disclose side effects and not make claims unsupported by research, the campaigns were barely distinguishable from one another when it came to the information presented. But advertising is much more about presentation than information, and the most creative presentation of the same old facts had to be Zoloft’s.

  In 2001, Pfizer introduced a cartoon character, a Bizarro version of the smiley face—a round blob that bounces along, sighing and moaning and even ignoring a butterfly as a narrator (male, more chipper than Prozac’s narrator) goes through the usual pitch. The blob ads even featured an animated version of a serotonin synapse; when the Zoloft logo appears at the top of the screen, the synapse becomes a much more lively place, and the narrator says, “When you know more about what’s wrong, you can help make it right.” He adds that only your doctor can know for sure if you have depression, but he suggests strongly that now that you’re educated, you go and ask.

  You have to admire the economy of ads like these. In a one-minute cartoon, they distill a century and a half of medical history into the simple message that if you are suffering, you may very well be sick, that your sickness is internal and biological, that it can be cured with a precision-targeted medicine, and above all else, that anyone can be depressed, that indeed the whole world can be insane. “Depression strikes one in eight,” says the Prozac TV ad. The number is 17 million in the print version, 20 million in the Zoloft ad, and 20 percent of women on Paxil’s website—a whole lot of fellow citizens, in other words, whose suffering was presumably no worse than yours, but who had finally come to understand that their unhappiness, no less than their infections and their high cholesterol, was just another disease and got on the Prozac bandwagon.

  Some critics worried that it just wasn’t fair to deploy the techniques of consumer advertising—which, as the British Medical Journal put it in an editorial decrying the practice, is “the science of arresting the human intelligence long enough to get money from it”—on vulnerable people. The FDA stepped in from time to time—for instance, in 2004, when Wyeth ran a radio spot for Effexor that started out:

  Hey you, listening to the radio…How’re you feeling these days? Okay? Not bad? Come on, is that where you want to be? When was the last time you did something you once looked forward to doing? You know, symptoms of depression could be holding you back.

  The FDA faxed a note to Wyeth, warning the company that “by failing to draw a clear distinction between major depressive disorder and normal periodic feelings of low interest or low energy, the advertisement broadens the indication for Effexor XR.” Which was no doubt what Wyeth had in mind. They just weren’t artful enough about it to avoid the FDA’s displeasure. Mostly, however, the drug companies stayed on the right side of the law—not too hard to do when all it took was the careful use of the subjunctive, as in “depression may be related to an imbalance of naturally occurring chemicals between nerve cells in the brain” (emphasis added). When a couple of researchers pointed out to the FDA that, according to Essential Psychopharmacology, a standard medical textbook, “there is no clear and convincing evidence that monoamine deficiency accounts for depression,” the FDA wrote back to say that this was an “interesting issue,” but that “these statements are used in an attempt to describe the putative mechanisms of neurotransmitter action(s) to the fraction of the public that functions at no higher than a 6th grade reading level.” The alleged stupidity of the citizenry, in other words, justified the drug companies’ lying to them.

  And every indication was that the ads, unlike the drugs, really worked. Ad industry research indicated that every dollar spent on consumer advertising yielded $1.37 in drug sales. Adam Block, an independent researcher at Harvard estimated in 2007 that more than a half million doctors’ office visits
were inspired every year by consumer advertising of antidepressants. Using epidemiological data, he estimated that only one in fifteen of those patients was likely to be depressed, but using statistics derived from other studies, he determined that more than half of them would get a prescription, which meant, he said, that only “six percent of the increase in antidepressant use due to [direct-to-consumer] advertising is by people who are clinically depressed.”

  Block concluded, however, that this wasn’t necessarily a bad thing. Even if the majority of the money spent on the drugs was for nondepressed people, he argued, the cost of untreated depression was so great that “treating everyone in the country with an SSRI would…provide a net benefit.” The drug industry hasn’t yet proposed this as official policy, but maybe they don’t have to. People aren’t just turning up at doctors’ offices with their personal symptoms checklists and asking their physicians to complete the examination. They’re coming in self-diagnosed—and already asking for the drug they’re sure will help. And doctors are more than happy to oblige.

  At least that’s what a group of researchers found out when they pulled a Rosenhan on 152 family doctors in 2003 and 2004.

  The team, led by Richard Kravitz, a University of California researcher, developed a method kinder and gentler than Rosenhan’s for sneaking in and seeing what doctors do when they don’t think anyone other than the patient is watching. They deployed standardized patients (SPs)—people, often actors, trained to present the symptoms of a particular disease and generally used to sharpen the diagnostic skills of medical students. (Doctors who agreed to be in the study knew that two SPs would visit them in the next year, but not what their complaint would be or what the researchers were studying.) Kravitz taught female SPs how to simulate one of two DSM diagnoses: major depressive disorder and adjustment disorder with depressed mood. The depressed patients complained of wrist pain and of the requisite five DSM symptoms: feeling “kind of down” for one month, worse in the past two weeks, of fatigue, sleep troubles, loss of appetite, and sensitivity to criticism. The adjustment disorder SPs had just been laid off from a job and were suffering from back pain, fatigue, “feeling stressed,” and sleep troubles.

  Kravitz sent his SPs on 298 visits to doctor’s offices, equally divided between the depressed and the adjustment-disordered patients. In addition to their symptoms, the SPs were armed with two scripts, both of which described something they’d seen on television. On about one-third of their visits, SPs didn’t deliver either script. In another third, they talked about an ad they saw for Paxil. “Some things about the ad really struck me,” they told the doctor. “I was wondering if you thought Paxil might help.” In the remaining third, they said that they’d seen a show about depression. “It really got me thinking. I was wondering if you thought a medicine might help me.” (Kravitz chose Paxil more or less at random; he received no drug company money for the study.)

  Fifteen of the forty-eight depressives who didn’t ask for drugs received prescriptions anyway. Twenty-seven of the fifty-one SPs requesting Paxil got an antidepressant, and fourteen of them got the brand they asked for. And thirty-eight of the fifty SPs who asked for “a medicine” got one. Those posing as adjustment disordered got similar results: 10 percent got drugs when they didn’t ask, nearly 40 percent when they did, and 55 percent when they requested Paxil (of whom two-thirds got their brand choice).

  These results, while less embarrassing to the profession than Rosenhan’s, still don’t put doctors in a favorable light. Only half the depressed patients got the minimal indicated treatment, and half of the patients who didn’t qualify for the treatment received it. Doctors failed to spot depression in 20 percent of the cases; they diagnosed it in nearly 40 percent of nondepressed people. And diagnosis rates increased significantly—from 65 to 88 percent in the depressed SPs and from 18 to 50 percent in the maladjusted—when a patient asked for drugs, a request that is not a known symptom of depression or of any other disease except substance abuse disorder.

  This increase was probably not due only to doctors covering their asses by justifying their prescriptions with a diagnosis. More likely, it occurred because talk of the cure put them in mind of the disease. The marketing effort, in other words, may create a collusion that neither doctor nor patient needs to be aware of. Indeed, it may be best if it works its magic entirely in the shadows.

  GlaxoSmithKline might wish that the ads were effective enough to dictate doctors’ brand preferences. But the company couldn’t miss the fact that although SPs rarely received Paxil if they didn’t ask for it, a simple request goosed sales noticeably. This finding is consistent with research indicating that while consumer advertising increases sales of a class of medications, it is old-fashioned detailing that determines the success of specific drugs. The consumer ads soften up the market; the detailers move in for the kill.

  And that’s the real triumph here, at least for the pharmaceutical industry—and for their ad agencies. For nearly fifty years, they’ve been on a campaign to convince Americans—doctors and consumers alike—that they suffer in enormous numbers from a disease called depression. This has not been some idle public health effort, but an attempt to link that disease to a particular cure, and it turns out that if you ask your doctor for the cure, your chances of getting the diagnosis go way up. In fact, in people who don’t have the symptoms, it nearly triples.

  Kravitz and his team didn’t do more than note the fact that talk of drugs increases diagnosis. They’re much more worried about the clinical appropriateness of the prescriptions, and about whether or not the ads lead to overtreatment. But then again, the official diagnosis may not even matter. Doctors may well be prescribing antidepressants to patients who ask for them for the same reason that they prescribe antibiotics to patients who, in their opinion, are suffering from a virus: because they are in the business of relieving suffering, and the patient is signaling that a pill will make her feel better, because, that is, they want to please their patients. And given their performance in clinical trials, what better drug to prescribe as a placebo than antidepressants?

  What does matter is that your doctor has ratified your request. If you go to the office with your Personal Symptom Checklist and you leave with a prescription for antidepressants, is it really important for him to run the numbers and tell you out loud that you have depression? Does it even matter if he thinks you do? Does a doctor have to say “bacterial infection” to make you think that this is what your antibiotics are for? After a half-century of being carpet-bombed by this message, it is virtually impossible to suffer prolonged sadness without considering the possibility that you have depression. Frank Ayd’s spiel is obsolete. Or, more precisely, it has become an essential part of the climate of opinion in which we experience our unhappiness.

  As Kravitz’s study inadvertently proves, you can teach people how to be depressed. He went to a lot of trouble to teach them well, but I think advertising and all the other channels through which the depression message is broadcast are also good teachers. And even if you have some doubts about whether the ad or the television program is really describing you, when your doctor hands you the pill, he’s confirming the diagnosis whether he means to or not.

  But what matters above all else about Kravitz’s study is that he has actually out-Rosenhanned Rosenhan. He’s pulled a prank he didn’t even mean to pull. Because in real life, none of those SPs was actually depressed, at least they weren’t when they were screened for the job. Yet 60 percent of them got a diagnosis, and nearly 45 percent of them got drugs. Try faking a case of diabetes. I don’t care how good an actor you are or how well informed. Unless you brought a real diabetic’s urine with you, or your doctor is criminally incompetent, you are not going to go home with a prescription for insulin.

  Okay, so maybe that’s not entirely fair. But it wasn’t my idea to compare depression to diabetes in the first place. That was the drug companies’ brainchild, as in “Depression doesn’t mean you have something wrong with
your character. It doesn’t mean you aren’t strong enough emotionally. It is a real medical condition, like diabetes or arthritis”—which is what you learn when you go to the Myths and Facts page on Pfizer’s zoloft.com website. Or prozac.com’s version: “Like other illnesses such as diabetes…depression is a real illness with real causes.”

  It’s easy to see why the depression doctors want to make that comparison. Diabetes provides a classic magic-bullet scenario: your pancreas stops producing insulin (or, in the case of type 2 diabetes, your cells lose their ability to absorb insulin), and the deficiency is treated with regular medication. No one would be ignorant or insensitive enough to suggest that your illness is related to your character or your emotional strength. No one would blame the victim or imply that a diabetic is weak for taking his medicine. A depressed person who thinks of himself this way, in other words, is a loyal patient for life.

  But doctors don’t have to convince their diabetic patients that they have a “real illness.” The symptoms generally speak for themselves. A diabetes doctor doesn’t have to worry about the clinical appropriateness of treatment. He doesn’t have to wait for a new definition of diabetes to be hashed out in committees of his brethren and then learn the new diagnostic criteria. He doesn’t have to worry about whether someone is going to show up at the office claiming to be diabetic, or perhaps hiding diabetes, and then embarrass him when he misses the diagnosis; all he has to do is to take a urine or blood sample. He doesn’t have to talk about chemical imbalances that he knows aren’t really the problem or contend with package inserts that say, in plain black and white, that the drug makers have no idea why their drug works.