A Prescription for Trouble
Its reports on new drugs put billions in pharmaceutical company profits and untold lives at stake. So it’s a very big deal when the New England Journal of Medicine gets something wrong.
For guests staying at the Cleveland InterContinental Hotel, late-night trips to the ice machine can be quite an adventure. The hotel is connected, via a short walkway, to the Cleveland Clinic, one of the country’s busiest hospitals; make a wrong turn, and the décor shifts abruptly from faux Chinese vases and dark wood paneling to bright fluorescent lights, swinging metal doors, and nurses’ scrubs. So when the InterCon played host this September to a conference on the blurring line between medicine and commerce, it seemed the organizers could not have chosen a more fitting venue.
With endocrinologists and anesthesiologists toiling almost within earshot, the 300 prominent doctors, lawyers, and academics who’d turned out for the event nibbled hors d’oeuvres and talked shop outside the hotel’s ballroom. Richard Thornburgh, the former U.S. attorney general, gave the lunchtime address. Nina Totenberg, legal affairs correspondent for NPR, moderated a session. The main attraction, though, was conspicuously absent. Word was spreading that Jeffrey Drazen, the powerful editor in chief of the New England Journal of Medicine—and the featured dinner speaker—was stranded in Boston, his flight out of Logan delayed by steady rain in northern Ohio.
There can be few people more qualified than Drazen to address the subject of conflicts of interest in medical research. And there can be few people for whom the subject is as acutely uncomfortable. (It is easy to imagine Drazen, back in the Logan departure lounge, doing a little rain dance.) Since November 2000, when the Journal published what it later acknowledged to be a flawed study on Vioxx, a new painkiller from the pharmaceutical giant Merck, pointed questions about how the study made it into the world’s most influential medical journal, and about the journal’s response to the controversy, have dogged Drazen wherever he goes.
The circumstances surrounding the Vioxx affair are complex, and some of them still in dispute. The now infamous study, which touted the new drug’s benefits—it caused fewer stomach problems than naproxen, an older painkiller marketed as Aleve—also significantly understated its health risks: Namely, three more Vioxx patients had heart attacks than were reported, and another 47 experienced serious cardiac problems. These particular details didn’t emerge until Merck submitted testing data to the Food and Drug Administration. In the meantime, 900,000 glossy reprints of the article—which Merck purchased from the Journal for as much as $836,000—were on their way into the hands of doctors around the country. Some 20 million Americans took the potentially dangerous drug before it was withdrawn from the market in 2004.
To make matters worse, two of the study’s 13 authors were Merck employees. The Journal is by no means the only periodical of its kind to have faced accusations of conflict of interest, or to have published flawed data. But the Vioxx study, in combining both indiscretions, looked less like oversight than something far more pernicious. Journalists, sensing a how-the-mighty-have-fallen scenario, have covered the incident with gusto. Trial lawyers have also smelled blood: More than 14,000 lawsuits have reportedly been filed against Merck over the Vioxx affair, further entangling the New England Journal in the mess.
Last December—five years after the fact—the Journal finally took a stab at damage control, publishing an unusual “Expression of Concern” claiming its editors had “recently obtained information” about the three unreported heart attacks that had resulted in a “misleading conclusion” about the risks of Vioxx in the published paper. Drazen said he had been spurred into action when the Journal became aware of an internal memo from Merck revealing the company had tabulated the heart risks well before the Vioxx paper was published. The “Expression of Concern,” Drazen said, was intended to “show the world that we will not tolerate this type of behavior!”
The Wall Street Journal, for one, was underwhelmed by Drazen’s newfound indignation. It argued that the response was little more than an attempt to insulate the magazine from executive editor Gregory Curfman’s admission, in a videotaped deposition, that the New England Journal had been guilty of “lax editing” on the Vioxx study. The article further implied that the journal’s belated retraction may have put lives at risk: “Had the journal acted before the recall, its authoritative voice almost certainly would have damped the Vioxx boom.” Merck, meanwhile, responded to the New England Journal’s claims with a vigorous denial, stating that it had completely followed the publication’s own rules, a rebuttal that received extensive coverage. This August the Journal of the Royal Society of Medicine, a British publication, concluded that “the New England Journal of Medicine, and medical journals in general, have been damaged.” Rather than mitigate the controversy, Drazen’s actions only invigorated it.
Not surprisingly, the ongoing Vioxx imbroglio has driven the Journal into a permanent defensive crouch. For weeks I’d been involved in an elaborate game of cat-and-mouse with Drazen’s PR reps, trying to set up an interview. So when Drazen finally arrived at the conference, I decided to just walk over and introduce myself. After a half-hearted handshake and an awkward silence, he launched into a discussion about his son, who lives in Oslo and is studying something called large-scale turbulence. It was his way of making small talk.
Drazen, who was a renowned asthma researcher before taking the job as the Journal’s top editor in 2000, never strays far from the details of science. As we chatted, though, it was his bow tie that I kept thinking about. It bore the New England Journal of Medicine logo, a maroon seal showing a quill pen crossed with the snake-entwined staff of Asclepius, the god of medicine. “It’s a limited edition,” Drazen said, not really looking at me. I kept staring. This was no ordinary bow tie, not one of those clip-ons worn by gawky teenagers at their proms. A perfect ribbon, it seemed as inviolable as the fondant frosting on a wedding cake. “Did you do that yourself?” I asked. He said it was as simple as tying his shoes. Apparently moved to prove it, he tugged on the ends of the tie, unraveling it. Then, without looking down, he quickly remade the bow. This version, alas, was less immaculate than the original.
To Drazen, critics who suggest that the New England Journal was asleep at the switch when it let the Vioxx study into its august pages have misrepresented the role that medical journals play. To be sure, they take measures to safeguard the quality of their material—when manuscripts come in, they are subjected to peer review, in which experts evaluate them for accuracy, originality, and scientific significance—but beyond that, the argument goes, there’s not much the journals can do to ensure they get all the details perfect. In a letter to the Wall Street Journal objecting to its reporting on the Vioxx controversy, Drazen pointed out that “editors have no subpoena powers.” The medical journals are just a bridge between researchers and doctors; the real responsibility for a study’s findings lies with its authors. And that, basically, was the message he was about to deliver to his colleagues in Cleveland as they started in on their filet mignon dinners.
Taking the dais, Drazen warmed up his audience by telling them he had enjoyed the spinach salad—a good way, he quipped, of getting more people into the ICU. It took him more than 30 minutes, as diners moved on to the tiramisu, to get to his own involvement in the Vioxx debacle, which had been limited to its aftermath. During his earlier career as an asthma researcher, he had partnered with as many as 21 pharmaceutical companies—ties that led some to question whether he could remain impartial as editor of a medical journal. To quell that criticism, Drazen had sold his drug company stock, ended his consulting arrangements, and agreed to refrain from working on articles involving any of those companies for two years. As it happened, one of the companies he’d had ties to was Merck. Before an issue is sent out the door, the Journal’s editors use different-colored pens to make their final changes. (Drazen uses purple—“the royal purple,” he says.) It had been Curfman, the executive editor, who’d signed off on the Vioxx study with his green stroke.
Facing the Cleveland crowd, Drazen invoked Kierkegaard, the Danish philosopher who wrote that life must always be lived forward, but appreciated backward. Then he compared himself to Grady Little, at the moment Little chose not to pull Pedro Martinez in the seventh game of the Sox’s 2003 pennant series with the Yankees—a roundabout way of saying that hindsight is 20/20. Yes, he said, he could have gone back to the Vioxx authors and asked them to explain the discrepancies between their paper and the data that later appeared on the FDA’s website. But the word “on the street” was that the FDA numbers were “late data”—industry-speak for results that come in after a study’s formal cutoff. In other words, he had given the authors the benefit of the doubt, assuming they’d filed incomplete information because they themselves simply didn’t have all the facts yet. “The study was done in 22 countries,” he said. “They probably had case reports in Swahili, and they had [to get] them translated and collated and gotten in.”
The Journal, Drazen now believes, “had been hoodwinked.” But he doesn’t see that as cause for an overhaul of his organization; instead, his approach is to depend on contributors to just play fair—a high-stakes wager on a system that many, even inside the industry, see as being in need of repair. “[Medical] journals have become an extension of the marketing arm of the pharmaceutical industry,” says Richard Smith, ex-editor of the British Medical Journal and author of The Trouble with Medical Journals. “And the industry generally gets the results it wants.”
The studies these journals publish tend to be funded by drug companies, which is sort of like Merrill Lynch paying for the content of Barron’s. And the authors of the studies are often bound by contract to get verification of the relevant data from the sponsoring company. This process inevitably creates potential for the company in question to offer a few helpful pointers on how the data might be interpreted. “That’s the rub right there,” says Dr. Eric Topol, a cardiologist and one of the earliest critics of Vioxx. “Because the company might want to position things—spin, if you will—in a more positive light for their drug.”
Indeed, a good review in the Journal is to a new drug what Roger Ebert’s upturned thumb is to a new movie—a stamp of quality that can translate into huge profits. “For them to have reprints from the New England Journal, that gives them the seal of approval that this is good stuff,” says Topol. “And that gets them into doctors’ offices.”
The Journal has earned its unmatched credibility over the course of almost two centuries. But in recent decades, that reputation has become increasingly fragile. Drazen isn’t the first editor to face threats to its integrity—both from outside and within its own walls.
Though most Bostonians have never laid eyes on a copy of the New England Journal of Medicine, the publication is part of our figurative landscape, one of the institutions that foster our city’s sense of intellectual superiority. Published on Thursdays, it goes out to more than 180,000 print subscribers, about 85 percent of whom are physicians, and about 13,000 more online. Even if you don’t read it, you know its name: No other research journal, anywhere, is cited as much. The Journal’s very appearance projects an aura of gravitas: Its cover consists of nothing more than an index of its featured articles, which include both research studies and “editorials,” in which contributors comment upon issues relating to medicine and research.
Founded in 1812, the New England Journal is the oldest continuously published periodical of its kind in the world. It’s owned by the Massachusetts Medical Society, which bought it for a dollar in 1921. Behind these quaint, dignified, so-very-Boston details, though, there is another story, one of an ongoing tension between the commercial interests of the magazine and its role as a stolid, unimpeachable medium of medical news and analysis. Such is its clout that the outcome of this struggle will have significant implications for medical journals worldwide.
The Journal’s rise to eminence began in earnest in 1967, with the editor Dr. Franz Ingelfinger, who had a knack for convincing top researchers to place their best studies with him. It was under Ingelfinger’s reign, too, that the Journal began to operate more like a traditional news outlet than a branch of the medical community. In 1969, he instituted what remains known as the Ingelfinger Rule, refusing to run studies that had been published or even touched upon elsewhere—a medical man by training, he was also a canny magazine editor who knew the value of a scoop. Circulation rose rapidly during the Ingelfinger years, as well as the Journal’s international prestige, but that growth came with risks. While heightening the impact of New England Journal articles, Ingelfinger’s strategy also increased its vulnerability when one of those articles proved tainted. (If, say, you run an exclusive study featuring an anti-acne medication that later causes people’s teeth to fall out, you’re the one left holding the teeth.)
A gastroenterologist, Ingelfinger diagnosed his own esophageal cancer and in the mid-1970s was forced to retire. He recommended as successor his close friend Dr. Arnold S. Relman, a professor at Boston University’s medical school and former chief of medicine at Boston City Hospital, who was given the job in 1977. Where Ingelfinger had succeeded in generating buzz, Relman’s legacy would be safeguarding the Journal’s integrity. In 1984, he made it the first medical journal to require authors to disclose their financial ties to companies they wrote about. Later, he went a step further, enjoining contributors from reviewing drugs produced by companies to which they had ties.
At the same time Relman was tightening the screws, the pharmaceutical industry began to step up its marketing efforts, with the Journal reaping some of the benefits. This sudden infusion of money would do more to change the culture and the content of the Journal than any amount of editorial rules and regulations.
When Relman began as editor, the Journal maintained its own bank account and kept its own books, and at the end of each year would submit a financial statement to the Massachusetts Medical Society, along with a portion of its income. “It was all kind of friendly and low-key,” he says. “We weren’t talking about a lot of money.” With profits topping off at $1 million in a good year, according to Relman, the Journal had yet to become the cash cow it is today (the Journal made $88 million last year, $20 million from advertising). But it was headed in that direction.
Relman floated the idea of using the extra cash to establish a trust fund for research and education. Instead, he says, the Medical Society decided to combine the Journal’s bank account with its own. “I objected to it as strenuously as I could,” he says, recalling his admonition to the trustees: “Don’t allow the Society to live off the Journal, because sooner or later, that will compromise its integrity.” His plea, he says, “didn’t cut any ice at all.” (According to communications director Frank Fortin, the Medical Society created investment accounts that were “in the spirit” of Relman’s proposal. Whatever funds aren’t plowed back into the Journal, he says, “go toward advancing medical knowledge, and advocating for patients and physicians.”)
After 14 years at the helm, Relman left the Journal in 1991. It was a tough job, he says, and he was tired. But there was also a sense that he’d become too keen to tackle the issue of money in medical research, a claim Fortin denies: “No one here shies away from controversy.” The search committee charged with finding the next editor tapped Jerome P. Kassirer, a professor at the Tufts School of Medicine. Kassirer kept Relman’s tough conflict-of-interest policies in place, and by 1996 the Journal still stood alone in barring authors from penning editorials about companies in which they have a financial interest.
Despite the efforts to keep its content uncompromised, the New England Journal, like all medical journals, was vulnerable to self-interested contributors. During the same year its strict standards put it in the vanguard of the industry, the Journal somehow allowed itself to be duped by the authors of an editorial that touted the diet drug Redux. A published study had shown that the drug posed a significant risk for hypertension, but the Journal piece concluded that any risks were outweighed by its benefits. It turned out that the authors had been paid consultants for companies that manufactured or marketed the drug.
Kassirer learned about this the hard way. Newspaper reporters, who had received advance copies of the article, called the Journal’s offices asking about the conflicts of interest three days before the August 29, 1996, publication date. The authors had received the magazine’s standard letter, which warned editorialists that they could not have “regular” consultancies or “ongoing” financial relationships with their subjects, and they had written back stating that they had no such ties. The New England Journal later published an editorial on the sorry episode, and tightened the phrasing of its policy to make it less open to interpretation.
Even as the Redux fiasco played out, though, the Medical Society began to push its prized publication to be less picky about the articles it published, and to extend its brand to other titles, a move many of its editors saw as a blatant moneymaking scheme. One Medical Society consultant pointed out that the Journal published just 10 percent of the manuscripts it received, and asked, “Why should we let this meat fly out the door?” Line extensions were proposed—the New England Journal of Cardiology, the New England Journal of Oncology—to be filled with articles deemed not worthy of the main journal itself. The editors had another name for these proposed publications: the New England Journal of Rejects.
The Journal’s editorial offices are on the sixth floor of the Francis A. Countway Library, at the Harvard Medical School near Brigham Circle. The setting provides editorial staff with a psychological barrier—and geographical distance—from the Society suits, who operate out of a lavish, modern three-story building in a corporate office park in suburban Waltham. Two ponds grace the manicured grounds. Inside, an atrium towers overhead. The Journal’s editors call it the “Taj Mahal.”
Since before Kassirer’s term as editor, the Medical Society had been pushing to relocate the Journal’s editorial offices to the compound. Kassirer resisted the move. “I thought it was a really bad idea to have members of the Medical Society able to walk into the editorial offices of the Journal,” he says. “The big advantage to the Countway Library is that you can’t park around it. You can’t park, and nobody ever showed up there.”
Kassirer and other editors saw the proposed move as another step toward the commercialization of their enterprise, an undermining of the divide between business and editorial. Kassirer recalls, for instance, how the Medical Society zeroed in on the revenue-generating potential of the Journal’s newly created Web page, back in 1996. “They were talking about a lot of advertising on the Web page and [posting] content that wasn’t the New England Journal’s,” he says. “I had no authority over these decisions, because the Medical Society owns the Journal, and I was just the editor in chief.”
Kassirer took the spat public in 1998, penning a report headlined “A Higher Calling.” He reprinted the Journal’s own mission statement, which speaks of advancing medical knowledge and developing professional and ethical standards, but nowhere mentions maximizing profits. “Does the Society want to become a business?” he wrote—a controversial question considering the Medical Society is a nonprofit organization. A year later, Kassirer was gone.
Marcia Angell, then the executive editor, succeeded Kassirer as editor in chief. Already, Angell was well acquainted with the Society’s quest to capitalize on the Journal brand. “Titles got more and more grandiose,” she says, “until it sounded like something out of Gilbert and Sullivan.” Angell, too, ran afoul of her bosses. At one point, she says, Jack Evjy, president of the Medical Society at the time, asked her, “Marcia, what do you have against money?” She kept her job for less than a year. “They hated her as much as they hated me,” Kassirer says.
In Angell’s farewell editorial on June 29, 2000, she wrote that “the Journal is no longer an independent entity, but part of a larger enterprise.” Today, she says, the same thing holds true. “There’s nothing like money to make people want more of it.”
Six years into Jeffrey Drazen’s tenure as editor of the New England Journal, the dreaded move to Waltham has yet to happen. From his office, Drazen can still look across the street to the Brigham and Women’s Hospital, where he used to be chief of pulmonary and critical care medicine.
One morning, I meet with Drazen in his office. Bald and bespectacled, he carries himself stiffly, and rarely makes eye contact. Sitting behind his desk, in his chestnut-brown cardigan sweater and bow tie—this one polka-dotted—the 60-year-old looks very much the academic. The most notable thing about his office, meanwhile, is the abundance of grandfather clocks. There are lots of them, here and at his home in Winchester. He restores them, he explains, in his basement. A tinkerer by nature, Drazen studied engineering and physics as an undergraduate at Tufts University. “I grew up wanting to do research,” he says.
At Harvard Medical School, his very first patient was a 14-year-old girl from Peabody who was suffering from asthma. At the time, there weren’t any good treatments available to control the disease. So Drazen set out to create one. Years later, as he and his colleagues completed a trial on a treatment for bronchoconstriction, competing researchers stole their thunder, publishing the results of a similar trial in the British medical journal the Lancet. Drazen had been in a race for scientific discovery, and he had lost. Eventually, he and his team contributed to the creation of three widely used asthma treatments—Accolate, Zyflo, and Singulair—but the episode left its mark. Certainly it gave Drazen a keen sense of the competitive, first-past-the-post nature of medical research, a sense he carries with him to this day.
Before he became editor, there were complaints that the Journal was too focused on general-interest health issues, that the research end of things got short shrift. Immediately, Drazen set out to change this. He would publish more clinical drug trials—trials like the Vioxx study—rather than surveys of people’s habits, the kind that gauge, say, the health effects of eating chocolate or drinking red wine. The Vioxx affair, however, suggested that Drazen may have tilted things a little too far in the other direction, allowing his sympathy for his fellow researchers to cloud his judgment. He says, for instance, that one reason he waited so long to disavow the Vioxx study was that he didn’t want to damage the careers of the study’s authors.
Drazen’s tenure at the New England Journal has also been marked by a vigorous dedication to making its content timely. In 2003, for instance, he saw to it that the Journal was at the forefront of reporting on the SARS virus, publishing detailed case studies of the disease within two and a half weeks of the World Health Organization’s public advisory. In this regard, even the Vioxx study provides Drazen with a measure of satisfaction. While the Journal may not have provided a complete account of the coronary risks of the drug, he points out, it did at least report on some of its risks before other journals. “It was the first!” he says.
Even as the Vioxx controversy drags on, Drazen insists he has maintained a good relationship with the people at the Society. “It was a risk, when I took the job, that I could have been put in a position where I was being controlled,” he says, “but what’s happened is exactly the opposite. I’ve been given the resources I need to do what I wanted to do, and the Medical Society has pretty much otherwise left us alone.”
But Drazen hasn’t exactly been spoiling for a fight, either. Like his predecessors, he believes the Journal’s most precious asset is its name. “It would be foolish to squander that,” he says, “to make a few bucks.” And yet, he adds, “I’m a believer that if you don’t have a margin, you can’t have a mission. We’re not a charity.” While the deal Drazen cut with the Society required his approval of any use of the Journal’s name on publications, he is by no means a stickler when it comes to extending the brand. “I was willing to try new stuff,” he says. When Drazen arrived, the Journal was still working with old-fashioned paper manuscripts. Today it has podcasts. He still meets regularly with Society members to discuss further improvements. “You couldn’t have a wall between you and the publisher about every single issue,” he says.
Most remarkably, perhaps, Drazen has relaxed the Journal’s long-standing conflict-of-interest policies. Today the publication merely requires that editorialists have no “significant” financial interests with the drug companies being written about, meaning $10,000 or more. “The problem is that it’s difficult to define zero,” he explains. “Do you count a pen? A mousepad? An educational grant?”
While some would argue that Drazen’s shift risks the Journal’s getting too cozy with drug companies, others would say it’s not cozy enough. Dr. Thomas Stossel—a professor at Harvard Medical School, senior physician at Brigham and Women’s, and the brother of 20/20 correspondent John Stossel—is one of these people. Stossel believes that the “Expression of Concern” Drazen published overstepped his duties. “In my opinion, there’s no smoking gun to show that Merck deliberately misled anyone,” he says, adding, “To the extent that it hurts Merck, it hurts us all. Merck has no obligation to make drugs, and if tomorrow they decide it’s safer and more profitable to make dog food, they’ll do it.”
Stossel’s criticism points to an inevitable occupational hazard for someone in Drazen’s position: No matter what you do, someone’s going to think it’s the wrong course. But the embattled editor shows no signs of cracking just yet. Recently, he told a reporter that he intends to stay at the Journal until 2012, when it will celebrate its 200th anniversary. Perhaps by then the memory of the Vioxx scandal will have faded, and come to be regarded as what President Bush might call “just a comma.”
When I ask exactly what has changed in the wake of the Vioxx mess, and what he’s doing to make sure it doesn’t happen again, Drazen tells his assistant to print “The Paragraph.” In an instant, the piece of paper sits on the table before me: “Our expectation is that your revised manuscript will include full, accurate, and up-to-date reporting of adverse events. In general, this should include a table containing descriptions of all serious adverse events and all common or important other adverse events. We believe it is important that the abstract contain a statement regarding adverse events.”
Drazen says his editors are being “a lot more cynical” with authors and asking a lot more questions about the harmful side effects of drugs, “almost to the point of driving them nuts.” And that’s it. In the thick of a scandal involving his journal’s being “hoodwinked” into publishing a dubious study, in the midst of a financial boom in the medical research industry that has intensified competition to a mind-boggling degree, the editor has chosen to defer to the honor system. “I am not a person who wants to make more rules,” he says. “I just want people to behave.”