Ten percent of Americans—children, teens and adults—take antidepressants. Whether it’s Prozac, Paxil, Lexapro, Effexor or Cymbalta, 30 million of us take a pill daily in hopes it will keep dark moods at bay. Antidepressants are the most prescribed family of drugs in America, an $11.9 billion market in the United States in 2007.
In January, Erick Turner, a professor of psychiatry at Oregon Health & Science University and a clinician at the Portland VA Medical Center, shook up the medical community, provoked the pharmaceutical establishment and, perhaps, disappointed millions of depressed Americans. He published a paper in the New England Journal of Medicine
that revealed antidepressants are not as effective as we’ve been led to believe. For years, he implied, pharmaceutical companies such as Pfizer (maker of Zoloft) and Forest Laboratories (Celexa and Lexapro) have vastly exaggerated the performance of their drugs.
Turner calls it the “dirty little secret” of the psychiatric world.
It was a disclosure that was felt ’round the medical world, and for the past four months Turner has fielded press inquiries by the dozens.
Turner’s “is the kind of a paper that makes you wonder why someone didn’t do it a long time ago,” says S. Nassir Ghaemi, an associate professor of psychiatry and public health at Emory University School of Medicine and a leading researcher on bipolar disorder.
“It’s damaging” to the reputation of antidepressants, says Peter Kramer, author of Listening to Prozac
and a psychiatry professor at Brown University School of Medicine.
The paper has given Turner not rock-star status, perhaps, but at least a level of notoriety in a profession where physicians typically labor in anonymity. Since the paper’s publication, Turner has become a go-to source for reporters writing about depression and antidepressants.
But more important for Turner is how his work has shaken up the industry and raised questions about the integrity of the pharmaceutical industry and the weakness of the federal regulatory agency that’s supposed to protect the public from drugs that are dangerous or ineffective. It has also had the effect of raising an uncomfortable question: If antidepressants don’t work that well, why are so many Americans taking them?
The great irony of his discovery is that, for 18 months, Turner pimped for antidepressants. He wouldn’t put it that way, of course, as many doctors do the same thing. But starting in 2004, Turner, who is the medical director of the Mood Disorders Program at the Portland VA and has worked at the Food and Drug Administration and the National Institute of Mental Health, became a speaker for Eli Lilly, one of the world’s largest pharmaceutical companies.
It’s not unusual in pharmaceutical marketing for doctors to be hired as “consultants”—or members of a company’s “speakers’ bureau” in industry parlance—and then hit the road doing “doctor talks.” These are typically lunches or dinners where area M.D.s are invited by a company to eat and drink while listening to a respected physician describe the benefits of a particular drug.
Lilly approached Turner around the time the FDA was set to approve a new Lilly antidepressant named Cymbalta. He was an especially good catch for the company because, in addition to his academic appointment at OHSU, Turner had spent seven years as a researcher at the prestigious National Institute of Mental Health and another three years as a clinical trials reviewer at the FDA.
“They’re using your reputation and political capital, as it were, as sort of a frontman for the drug,” says Turner, 54.
After training in Indianapolis in the summer of 2004, Lilly sent Turner out into the field in the Northwest, receiving anywhere from $500 to $750 per talk. He says he did about 12 talks for Lilly over the next 18 months.
He says his motivation wasn’t so much the money—he netted less than $10,000—as it was his desire to keep up his reputation as an expert on clinical trials.
“In the beginning, I think I got narcissistic gratification,” he says. “They fly you somewhere else in the country and pick you up in a limo, and you stay in a nice hotel you could never afford otherwise.”
Some critics believe these relationships between industry and doctors—long common in all branches of medicine—are little more than an example of the corrupting influence of Big Pharma and its ability to turn doctors into sales tools.
“The practice of drug companies paying doctors to be their spokespeople is extremely problematic, because it creates a financial incentive for the physicians to intentionally be deceptive in their lectures to other doctors,” says Daniel Carlat, an assistant clinical professor of psychiatry at Tufts University School of Medicine. Carlat himself was once on the dole for Wyeth Pharmaceuticals, makers of Effexor, until he found the arrangement too compromising and quit.
“It provides incentives to tweak the data to tell only part of the truth,” Carlat says.
And indeed, Turner found he could say only what Lilly allowed him to say. He could use only Lilly’s overhead slides of results from clinical trials of the drug. He couldn’t offer his own expertise as a researcher and former FDA reviewer to his fellow doctors.
“I began to feel straitjacketed,” he says.
It was during this time that Turner began to rebel.
Depression didn’t just appear with the advent of Prozac in 1987. The condition has been known since ancient times, and is infamously difficult to treat. In modern times, the advances of psychoanalysis and psychotherapy did little to fend off what Winston Churchill once called the “black dog.” In the 1960s, two classes of antidepressants became available. Neither was particularly effective, and they were not widely used.
By the mid-1980s, however, researchers had hit upon the serotonin hypothesis of depression—too little of the chemical serotonin in the human brain leads to depression. New-generation drugs believed to boost serotonin levels could treat depression. The drugs were known as selective serotonin reuptake inhibitors, or SSRIs.
The first new-generation antidepressant to be sold in the United States was Prozac in 1987, manufactured by Eli Lilly. Lilly claimed Prozac not only worked, it was virtually free of side effects. The drug rocketed to blockbuster status. In 1990, a Prozac capsule appeared on the cover of Newsweek and was hailed as a “breakthrough drug.” A few years later, Kramer’s Listening to Prozac
became a bestseller. The drug achieved iconic status in American culture in a way few drugs ever had before or have since. The message was clear: Take Prozac and beat the black dog to the ground.
Prozac was followed by Paxil and Zoloft, and a host of other new antidepressants such as Effexor and Wellbutrin, 12 new drugs in all. In 2007, 233 million prescriptions for antidepressants were written in the United States, according to data-services company IMS Health, accounting for $11.9 billion in sales. So widely used are these drugs that a recent study found traces of Prozac in creekbed mud in Portland and a 2004 British study found traces of the same drug in the British water supply.
But another problem showed up in clinical trials that were never made public. These new antidepressants—which critics argued had turned America into a culture of the quick fix and that proponents praised as lifesavers—didn’t perform very well. Turner knew this. And he decided in 2004, at the very same time he was taking money from Lilly, that it was time to start telling the truth.
“I guess you could say I bit the hand that fed,” says Turner of his revelation.
The FDA requires that clinical trials—studies done on human test subjects—be done before the approval of a drug. These trials determine several things: what dose of a drug is safe, what dose creates an effect, what side effects crop up for patients. But the biggest question is efficacy: How well does the drug work compared to a placebo? Pharmaceutical companies often run as many as 10 clinical trials enlisting thousands of patients and costing tens of millions of dollars to find out just how much efficacy their drug possesses.
You’d think doctors and the public would be allowed to see the results of all these trials, but that’s often not the case. Only two clinical trials with positive results—meaning the drug performs better than a placebo—are required for FDA approval. It doesn’t matter if eight other studies show negative results. Typically, the two studies are published in peer-reviewed journals. The fate of any other pre-approval studies is left to the whims of the drug’s maker. There is no requirement that a pharmaceutical company make public all of its clinical trials.
In November 2004, Turner wrote a paper that was published in PLoS Medicine
, an influential, peer-reviewed medical journal. In the article, Turner argued that all clinical trials submitted to the FDA should be published online by the agency, so doctors could know the breadth of a particular drug’s research base before prescribing it to a patient.
The paper made him an official critic of the pharmaceutical industry and its long-standing practice of hiding data from public view.
Hardly anyone in the medical-publishing world listened.
Feeling as if his hands were tied, Turner quit doing doctor talks for the pharmaceutical industry in 2005, convinced he needed to make his case more strongly by collecting reports of pre-approval clinical trials for antidepressants that had not been published.
He found some of them online, deep in the FDA’s website. Then he reached out to other researchers in the field, getting some records from a researcher at the University of Nevada at Las Vegas and other records from a researcher in Seattle, where Turner says, “I literally went down to a Kinko’s and photocopied them.”
When he was done, Turner had amassed findings from 74 clinical trials. Of the group, 51 percent were “positive”—in other words, the drug had performed better than a placebo. Forty-nine percent were negative or had mixed results. The bulk of these negative trials had never been published anywhere. It was almost as if those trials, and the thousands of patients who participated in them, had never existed.
Turner’s paper was published Jan. 17 in the New England Journal of Medicine
, titled “Selective Publication of Antidepressant Trials and Its Influence on Apparent Efficacy.”
“As a result of selective reporting, the published literature conveyed an effect size nearly one-third larger than the effect size derived from the FDA data,” Turner wrote. “Effect size” means the amount by which a drug outperforms a placebo, and from the study it’s clear the pharmaceutical industry had gotten away with overstating the power of antidepressants for two decades. Put another way, the industry had made use of the drugs look more beneficial.
“By altering the apparent risk-benefit ratio of drugs,” Turner wrote, “selective publication can lead doctors to make inappropriate prescribing decisions that may not be in the best interest of their patients and, thus, the public health.”
In the calm language of science, Turner had declared war on Big Pharma.
“My wife and I were talking about going away for the weekend in January,” he recalls. “I said, ‘That’s right after my paper comes out, and then I’ll be busy with all sorts of interviews by the press.’ My wife said, ‘Now that you’ve gotten your paper published, you can relax.’”
There’s been no relaxing since the paper’s publication. Turner, who likes to play tennis, says he’s hardly had any free time for the game.
“The fact that the negative trials can just be hidden away means that practicing doctors can get a very false notion” of efficacy data for a drug, says Carlat of Tufts. “That’s the real crisis here.”
So startling was this news to the media that Turner has fielded dozens of press inquiries from the likes of The Economist and the Fox News Channel in the seven weeks since the study was published. “[Doctors] end up asking, ‘How come these drugs seem to work so well in all these studies, and I’m not getting that response?’” is how The New York Times
quoted Turner explaining the implications of his study.
Among researchers, Turner’s study generated enthusiasm of the sober, academic variety.
“We’ve known for a long time that companies don’t always publish their findings,” says Bruce Psaty, a University of Washington cardiologist and expert on clinical trials, noting that these findings usually come to light only in references within FDA documents or in later court litigation over injuries allegedly caused by a particular drug. “It’s a terrific paper,” he says, especially for its ability to quantify “industry-friendly spin” created through the magic of selective publication. “It’s potentially a landmark paper.”
Carlat adds, “It sure told us something about how FDA research papers are treated by the pharmaceutical industry.”
George Keepers, chairman of the psychiatry department at OHSU, says Turner’s paper will play an important role in a simmering debate among U.S. psychiatrists about whether antidepressants or psychotherapies such as cognitive behavioral therapy should be first-line treatments for depression. “I don’t think you change the way you practice the next day” in the aftermath of a paper like Turner’s, Keepers says. “But it makes you more cautious about conclusions that have been drawn to this point” about antidepressants.
Dr. Richard Felix, a psychiatrist with St. Patrick Hospital in Missoula, agrees.
“We already know we don’t have the data we need from the FDA’s studies,” Felix says. “But I think it reinforces the fact that physicians have to look at all the data—including something like [Turner’s paper]—before prescribing medicine for their patients.”
Pharmaceutical companies primarily left an industry trade group, the Pharmaceutical Research and Manufacturers of America, to respond on their behalf.
“There should be no doubt that America’s pharmaceutical research companies are committed to making available clinical trial information to make sure that patients and their physicians have access to relevant data from ongoing clinical testing,” said the group’s senior vice president, Ken Johnson, in a prepared statement. “The process—which includes clinical trial registries and databases—should be transparent and accessible.”
Turner’s paper leaves little question that antidepressants don’t work nearly as well as conventional wisdom would have it. Of the 30 million daily users in this country, many millions would be no worse off if they took a placebo every day. Yet doctors continue to prescribe the drugs, people continue to take them, and Big Pharma rings up the sales.
One could argue, “What’s so wrong with that?” If people are helped by believing that something works, even if it is not medically true, doesn’t it still work?
Carlat, who called Turner’s work important, nevertheless argues that the question may not matter.
“The fact is, we’re not allowed to prescribe placebos in our practices, and we have patients coming to us and banging down our doors because they are miserable,” says Carlat. “We can’t offer them a sugar pill, but we can offer an antidepressant even if its effect is 80 percent sugar pill. A lot of our patients are getting better. That’s why we prescribe antidepressants even in the face of this recurrent data.”
However, there is evidence that taking antidepressants can be harmful to some people’s health. In recent years, the FDA has mandated warnings on all antidepressants due to elevated risks of suicide and suicidal thinking in some people under the age of 25 who take the drugs. In addition, antidepressants can cause very intense withdrawal problems for some patients, most notably with Effexor and Paxil. The drugs can also cause internal agitation—akathisia—in some patients, and, tragically, antidepressant use sometimes has been connected, however peripherally, to school shootings, most famously at Columbine High School in 1999, where one of the shooters, Eric Harris, had been taking Luvox.
There is also the not-insignificant cost of the drugs. No one will hazard a guess as to how much of America’s investment in antidepressants may be money down the drain, but it’s obvious that at least a portion is.
Just three weeks ago, Emory University’s Ghaemi wrote an editorial in the American Journal of Psychiatry
that openly questioned antidepressant use. He wrote about a long-term study of antidepressant use in bipolar disorder, and its finding that a placebo was more effective than the drugs, calling it “one more nail in the coffin of antidepressant use in bipolar disorder.”
As much as 20 percent of all antidepressant sales are to treat bipolar disorder, suggesting billions of dollars are perhaps being spent needlessly.
That said, there are some people for whom antidepressants are highly effective and who view them as lifesavers.
Turner himself still prescribes antidepressants but admits he is careful not to overplay their benefits to patients in order not to give them false hope after decades of hype around the drugs.
“There are some people who seem to get a great response and a lot who don’t seem to get much response at all,” says Turner. He says he probably won’t be asked by the industry to run clinical trials on antidepressants anytime soon.
“I don’t think they would want to risk it after this paper,” he says. “They would not want to touch me with a 10-foot pole, unless it had a sharp end on it.”
Editor’s Note: Philip Dawdy has been prescribed eight different antidepressants over the past 20 years. At times, the results were good but not lasting. At others, the side effects were devastating.
Dawdy, a former
Willamette Week and
Seattle Weekly reporter, is an award-winning freelance reporter in Seattle. His reporting on mental health issues has won awards from the National Mental Health Association (now called Mental Health America) as well as numerous local and regional journalism awards. He writes the popular mental health blog