The idea sounded fishy to Dr. Adriane Fugh-Berman. She was not about to put her name on a ghostwritten article for a medical journal. But she was curious, so she played along for a while.
An associate professor at Georgetown University Medical Center, Fugh-Berman was contacted in 2004 by a medical communications firm working for drug maker AstraZeneca with a proposition: Would she like an author credit on a forthcoming article to be submitted to a journal? A few weeks later, Fugh-Berman said, she received a manuscript of nearly 2,500 words, complete with an abstract, footnotes and a table. An accompanying note asked her to return the draft with any changes within a week.
The paper was about the risks of warfarin, a generic anti-clotting drug, for people also taking herbal supplements. AstraZeneca was developing a rival drug that would supposedly be safer for supplement users. A positive article could give the new drug a promotional lift. Fugh-Berman was considered an expert on drug-supplement interactions, so her byline would carry some weight. But long concerned about overprescribing in medicine, she turned out to be the wrong person to ask.
Fugh-Berman declined the author credit, and that might have been the end of it. But a few months later, editors of a major journal asked Fugh-Berman to peer review an article they were considering for publication. She quickly recognized it as nearly the same as the draft offered to her, though with a different author’s name.
Alerted by Fugh-Berman, editors of the Journal of General Internal Medicine rejected the article. They then published her account of the episode, along with an editorial condemning what they called “an egregious case of unethical behavior” that aimed to inject ”bias and untruth into the scientific dialogue in order to enhance corporate profits.”
AstraZeneca and the medical writing contractor, Rx Communications, were not immediately fingered as the culprits. On advice from its lawyers, the journal identified the drug maker as “ABC Drugs” and the medical writing firm as “XYZ Communications.” As Dr. William M. Tierney, then the co-editor, recently told FairWarning: “They had a lot more money to pay lawyers than we do … I didn’t need to publicly shame them” to make the point.
But Fugh-Berman was determined to out the companies. In a piece for The Guardian she did.
Struggling with the fallout, AstraZeneca and Rx Communicationsdenied engaging in ghostwriting, calling the whole thing a clumsy mistake. Rx, they said, had accidentally sent Fugh-Berman a manuscript developed by another academic author.
For the big drug maker, the timing couldn’t have been worse. A short time before, an AstraZeneca executive appearing before a British parliamentary committee asserted that the firm never engaged in ghostwriting.
For Fugh-Berman, now a Georgetown professor, the experience deepened her interest in ghostwriting and the corruption of medical literature. In 2007, she founded a project called PharmedOUT , which aims to educate health care professionals about marketing practices of drug and medical device makers.
In an age of distortion, propaganda and fake news, medical literature might seem to be a safe space for honest scientific inquiry, with no room for bias or spin. It isn’t so. Court proceedings, investigations and whistleblower accounts have revealed a long history of drug companies manipulating the literature to promote their drugs or disparage rival products–with the aim of getting doctors to prescribe more of their meds.
Of course, they have more direct ways to goose sales. Drug companies in 2016 spent an estimated $6 billion on direct-to-consumer ads in the U.S., the only country other than New Zealand that allows them, according to arecent analysis in JAMA. That same year, the study found, the industry spent $20.3 billion on marketing to doctors and other health care professionals, including by providing free samples, free meals and speaking fees.
Less well known is the tactic of developing favorable medical articles to influence prescribing decisions. Over the years, records and interviews show, such articles have often involved ghostwriting and its sketchy sidekick, ”guest” or ”honorary” authorship. Together, they have exaggerated the role of the independent researchers or ”thought leaders” who are credited as authors, while minimizing the role of those who did the heavy lifting–drug makers and paid medical writers.
It’s one way companies have sought to skirt a ban on promoting their drugs ”off-label,” or for uses not approved by the Food and Drug Administration. Doctors can still write prescriptions for off-label uses, and favorable articles bylined by independent experts make it more likely they will.
“Ghostwritten articles can have a significant impact on … physician prescribing practices,” according to a 2010 Senate staff report prepared for Sen. Charles Grassley (R-Iowa). ”When prominent physicians and scientists lend their names to an article, it raises the credibility of the findings and conclusions presented.”
In its purest (or impurest) form, ghostwriting is someone doing all the work but getting no credit, while someone else who does nothing gets all the credit. Often, the situation has been more nuanced, yet with drug makers determining the focus of articles, analyzing the data and–working with contract medical writers–creating detailed outlines or first drafts before the named authors weighed in.
Two different things
In this setting, “writer” and ”author” are not synonyms but actually mean two different things–as reflected in a 2001 email from a medical communications firm to a drug industry client. “I don’t know that any decision has been made about who is going to write the manuscript (not to be confused with who is going to be the author(s) of the manuscript … ),” the email said.
Court records reveal cases of drug companies paying honoraria to invited authors, but there are other rewards for taking part. To get authorship credit without breaking a sweat can help with ”publish or perish” pressures. And research funding or consulting opportunities may follow.
For their part, journal editors generally have had little ability or strong incentive to scrutinize the provenance of articles, observers say. Drug companies are major advertisers in some journals, and as one study found, articles developed with industry funding were more likely to result in lucrative reprint orders.
Since the essence of ghostwriting is secrecy, no one knows how many ghosted articles exist, but there must be a great number. When JAMA surveyed authors of more than 800 articles published in 1996, it found that 29 percent had ghost authors or honorary authorship, or both. Another survey published in 2008 found the percentage had dipped, but only to 21 percent.
Yet the issue has sparked little public controversy. ”I think industry influence on medical journals is not well-known by either consumers or health care providers,” Fugh-Berman said.
A spokeswoman for the Pharmaceutical Research and Manufacturers of America, the leading industry trade group, told FairWarning it had no one who could speak to the ghostwriting issue.
Drug maker GlaxoSmithKline declined an interview request, but said in a statement that its policy “prohibits ‘ghost writing’ of journal manuscripts and abstracts.” According to the statement, Glaxo meets authorship requirements of the International Committee of Medical Journal Editors, and “external medical writers are either named as authors or included in the acknowledgement section of manuscripts.”
Pfizer Inc., the world’s largest drug maker, also refused an interview request, but said in an email that it, too, adheres to disclosure guidelines of the international editors committee.
Some in the industry say that critics have demonized what is actually a collaborative process in which academic experts are engaged from the start, and rely on medical writers, who typically have advanced degrees and strong writing skills, for editorial support only.
The issue is ”mostly bogus,” said James Beck, an attorney and analyst at the Reed Smith law firm in Philadelphia, which has defended drug and device companies in product liability cases. ”You see the resumes of many of these top doctors, and they have hundreds of articles,” Beck told FairWarning. “They can’t possibly write them all from scratch–and they don’t.” Getting editorial assistance ”is a standard scientific practice,” Beck said. It does “not change in the slightest what the underlying data is.”
In a recent blog post, Beck called out plaintiff’s lawyers who decry ghostwriting as hypocrites, saying they ”routinely massage (if not outright create)” the court filings submitted by their expert witnesses. But “let a drug/device company provide authorship assistance to a busy doctor or a scientist, and the same plaintiffs’ lawyers … start screaming and yelling that something terrible is happening.”
But critics say the industry-sponsored articles can lead doctors to prescribe drugs that are unsafe or more expensive than products that are at least as effective. “There’s truthiness in some of these articles, but it’s slanted in a way that distorts inappropriately the message,” said Dr. Jay Siwek, editor emeritus of the journal American Family Physician. The result, he told FairWarning, is “that physicians are misled, and the desired effect of writing more prescriptions for a drug … is achieved.”
The impact is long-lasting, Siwek asserted in a 2016 letter to the editor of The BMJ (formerly the British Medical Journal). “Once distorted information makes it into the medical literature, the damage is done: other authors unwittingly incorporate it into other articles, spreading the distortions.”
Repentant ex-ghosters have made similar points.
Alastair Matheson, a medical writer who ghosted articles from the mid-1990s until 2012, described the work as “‘spinning science to sell drugs,” and said he finally quit because ”I considered it to be unethical.” Matheson has written about ”medicine’s culture of misattribution” that aims to ”aggrandize the role of academics and downplay that of companies.”
What’s not to like?
Linda Logdberg, a ghostwriter for about a dozen years, said there were times when articles were submitted for publication “without the ‘physician-author’ seeing the final draft.”
In an essay in PLoS Medicine, Logdberg acknowledged that the work held powerful attractions. “The money was good. Really good.” she wrote. “And perhaps most important in the longer run—it was fun. Traveling, eating in high-end restaurants, wearing fashionable clothes, and rushing to meet important deadlines—what’s not to like?”
But Logdberg, who has a doctorate in anatomy, quit to become a biology teacher after a falling out with a medical writing firm. The dispute involved an article about the advantages of prescribing the Novartis drug Ritalin to treat children with attention deficit-hyperactivity disorder. “I was unwilling to turn this ugly duckling of a ‘me-too’ drug into a marketable swan,” Logdberg wrote. Unforgivably, she also discussed her ghostwriting activities with The New York Times.
Medical writer Stephen R. Braun burned his ghostwriting bridges with a 2013 essay about his spin-doctoring of articles on testosterone replacement therapies. Yet for Braun, a particularly memorable assignment did not involve ghost writing per se.
”The Truth About Impotence” was the title of a May 1998 broadcast of NOVA, the acclaimed public TV science series. The timing was striking. Viagra, the erectile dysfunction pill that would become a blockbuster for Pfizer, had just come on the market. Braun was listed in the credits as the show’s co-writer, along with the rest of the production staff. It’s uncertain how many viewers also saw the inconspicuous tagline: ”This program was funded by an unrestricted educational grant from Pfizer Inc.”
The program mainly described the impotence remedies, such as surgery, that predated Pfizer’s magic pill. But ”boy oh boy, did everyone on that project know that Pfizer was funding it,” Braun told FairWarning. “And again, we included all the other treatments but it ended up being an hour-long PBS show ad for Viagra,” Braun said. “Pfizer got a pretty good piece of marketing.”
Nova spokeswoman Jennifer Welsh said in an email that the broadcast was produced by the New England Research Institutes, the recipient of the Pfizer grant. “The mention of Viagra is a factual statement that it is the first pill to be approved by the FDA,” she wrote. ”NOVA programs provide factual science, they do not advocate products or positions.” Institutes officials did not return emails or phone calls.
It was a casual conversation that, after nearly 15 years, still burns bright in Leemon McHenry’s memory. While surfing off Point Dume in Malibu and waiting for a wave, a surfing buddy, Skip Murgatroyd, mentioned that he had to leave the water to attend a lecture at the University of California, Los Angeles.
McHenry knew Murgatroyd was a lawyer, but knew little about his practice. It turned out he worked for a Los Angeles firm that specialized in suing drug makers over alleged injuries from side effects of prescription drugs. The lecture Murgatroyd was attending was about bias and distortion in medical literature, so even surfing had to wait.
McHenry had a doctorate in philosophy and was teaching at California State University, Northridge. He had training in logic and critical thinking but none in the law. What happened next redirected his career, and his life. Murgatroyd invited him to review documents from one of his cases, including medical articles about the drug targeted in the lawsuit. McHenry recalled that when Murgatroyd later asked his reaction, he responded that, based on his reading of the medical literature, “you don’t have a leg to stand on.”
“All of that is ghostwritten,” Murgatroyd replied.
“This can’t be true,” was McHenry’s first thought. “This has got to be some lawyer’s spin.”
At the time, ghostwriting evidence had begun to trickle out in court proceedings and medical publications. As McHenry investigated, his skepticism melted away. “I thought it was incredible that the medical literature could be corrupted like that,” he said in a recent interview. He had shared a common belief about the purity of scientific and medical literature–that it was ”the most authoritative, trustworthy source of information that we have … To have that shattered is really a monumental kind of epiphany that one has.”
McHenry went to work as a research consultant for Murgatroyd’s law firm, Baum Hedlund Aristei and Goldman. The job requires sifting through mountains of emails, reports and other internal documents obtained in legal discovery, seeking evidence that companies had exaggerated the safety and underplayed the risks of their drugs. Some internal documents have revealed the publication strategies of drug makers and their use of ghostwriting and guest authorship to seed the literature with favorable articles.
More recently, McHenry’s sleuthing crossed into the field of toxicology, via lawsuits by victims of non-Hodgkin lymphoma, a type of cancer, who claim they contracted the disease from exposure to Monsanto’s Roundup weed killer. Last year, McHenry published a paper analyzing internal Monsanto documents about the development of defensive studies that found Roundup was safe. He asserted that the favorable studies were produced, in part, through ghostwriting by Monsanto officials and contractors.
Among the documents was a 2015 Monsanto email suggesting that ”we ghost-write” portions of the studies to hold down costs. The named authors ”would just edit and sign their names so to speak,” the email said. In a statement to FairWarning, Monsanto’s parent company Bayer said the Roundup studies ”underwent extensive independent peer review before they were published,” and that Monsanto’s role in the studies was limited to ”formatting assistance, a regulatory history overview, and stylistic comments on articles.”
McHenry, who is trim and fit at 63, works from the law firm’s 18th floor office in a tony area of West Los Angeles. If there is a dress code at Baum Hedlund, McHenry has been granted an exemption. On a day he met with a reporter, he wore jeans that might once have been blue, and a shirt so faded that the stripes were barely visible.
Along with prospecting for incriminating documents, McHenry has authored or co-authored a string of peer reviewed papers on ghostwriting and manipulation of medical literature. He said fear of being sued by drug manufacturers has led some journals to reject his papers–forcing him to publish them in journals in Poland, The Netherlands and India. (He acknowledged that working for a law firm that makes a living suing drug companies may also make publishers wary.) McHenry and his sometimes co-author, Australian psychiatrist and professor Jon Jureidini, are nearing publication of a book entitled, “The Illusion of Evidence-Based Medicine: Examining the Scientific Foundation of Clinical Research.”
In recent years, hundreds of internal documents about drug industry publishing strategies have surfaced in product liability suits and Justice Department prosecutions of drug makers. Some of these records have been published online, including in the Drug Industry Document Archive at the University of California, San Francisco library. For example:
—A 1999 document prepared for Pfizer by a medical writing firm outlined an ambitious publications strategy to promote the antidepressant drug Zoloft. Entitled “World Wide Publications Status Update,” it listed more than 80 articles published, or in the pipeline, including four in which the author was identified as “TBD.”
—A 2004 marketing plan for Lexapro, an anti-depressant made by Forest Laboratories, cited ghostwriting as one element. “Bylined articles will allow us to fold Lexapro messages into articles on depression, anxiety and co-morbidity, developed by (or ghostwritten for) thought leaders,” the document said. “We will identify a Lexapro thought leader to place 2-3 byline articles in trade journals, consumer publications and on the internet.”
The plan also called for training some 2,000 psychiatrists and primary care doctors–including ”national and local thought leaders”–to serve as faculty for a speakers bureau program to promote the drug.
In 2009, the Justice Department charged Forest, now part of Allergan, with fraudulent marketing practices, citing, among other things, an allegedly misleading article drafted by a contract medical writer, and kickbacks paid to doctors. Forest agreed to plead guilty and pay more than $313 million to resolve criminal and civil charges of lllegally promoting Lexapro and two other drugs.
Orders from the top
—Wyeth Pharmaceuticals, now part of Pfizer, hired the medical communications firm DesignWrite to turn out favorable articles about hormone replacement products used to treat menopause symptoms. A 2002 email to staff members of Wyeth and DesignWrite ordered up more supportive articles to counter bad publicity about cardiovascular risks. ”Senior management has charged the Publications Committee with increasing the number of positive … publications,” the email said. “They have asked us to publish at least 1 study per month. This increase in publication is necessary to challenge the…negative press”.
DesignWrite later boasted in a proposal to Wyeth that it had produced ”over 50 peer reviewed publications” about hormone replacement treatments, along with ”more than 50 scientific abstracts and posters, journal supplements, internal white papers, slide kits and symposia.”
—Merck came under attack for excluding some heart attack cases from a paper on a clinical trial of its blockbuster pain reliever Vioxx. Lead author Jeffrey R. Lisse, a University of Arizona rheumatologist, later explained that Merck had called the shots. “Merck designed the trial, paid for the trial, ran the trial,” Lisse told The New York Times. ”Merck came to me after the study was completed and said, ‘We want your help to work on the paper.’ The initial paper was written at Merck, and then it was sent to me for editing … Basically, I went with the cardiovascular data that was presented to me.”
Facing some 27,000 lawsuits blaming Vioxx for heart attacks and strokes, Merck took the drug off the market in 2004. In 2007, without admitting liability, it agreed to settle most of the lawsuits for $4.85 billion.
–Warner-Lambert, now part of Pfizer, used guest authors to promote off-label uses of the anti-seizure drug Neurontin, according to an analysis of company documents published in Annals of Internal Medicine. Separately, a commentary in the New England Journal of Medicine stated: “Although guest authorship and commercial bias in research are a well-recognized threat to scientific integrity, the documentation of comprehensive manipulation of research and publication related to Neurontin is remarkable.” Warner-Lambert pleaded guilty in 2004 to charges of illegal off-label promotion and agreed to pay $430 million in criminal and civil penalties.
—Perhaps the most notorious case of alleged ghostwriting involved the anti-depressant Paxil. SmithKline Beecham, now GlaxoSmithKline, had hoped for approval to market Paxil as safe and effective for adolescent depression, but abandoned the idea after clinical trials were deeply disappointing.
Records show the company and its medical writing contractor, a now defunct firm called Scientific Therapeutics, circled the wagons. “It would be commercially unacceptable to include a statement that efficacy had not been demonstrated,” an internal document said, “as this would undermine the profile of [Paxil].”
One planned paper on a Paxil clinical trial was shelved because ”the side effect data analysis was terribly unfavorable to our favorite antidepressant. And we hate when that happens,” quipped a Dec. 14, 2000 emailfrom Scientific Therapeutics. “There are some data that no amount of spin will fix,” another email said.
At odds with the data
But soon after, a paper about another clinical trial, called Study 329, triggered a wave of off-label prescribing of Paxil for adolescent depression. The article, published in the Journal of the American Academy of Child and Adolescent Psychiatry, concluded that Paxil “is generally well tolerated and effective for major depression in adolescents.” It would later be attacked as an extreme example of spin and distortion. It ”was largely ghostwritten,” according to a 2015 article in The BMJ, and ”claimed efficacy and safety … that was at odds with the data.”
The manuscript was drafted by Sally K. Laden of Scientific Therapeutics, who had also penned the email about ”our favorite antidepressant.” The lead author, Dr. Martin B. Keller, thanked Laden in a letter. “You did a superb job with this,” he wrote, adding that he and two co-authors were sending some ”rather minor changes.” Published in July, 2001, the article listed 22 authors, but not Laden. A footnote stated: “Editorial assistance was provided by Sally K. Laden, M.S. ”
Years later, when Laden’s deposition was taken in litigation over alleged suicide risks from Paxil, she acknowledged drafting the manuscript, but denied ghostwriting had occurred. Certain of the named authors were ”extremely involved” in development of the paper, she said, and she had “acted at their direction.” Efforts to reach Laden were unsuccessful.
Concerned about adolescent suicide risks, the FDA ordered a black box warning on Paxil and other anti-depressants in 2004. In 2011, the Justice Department charged Glaxo with fraudulently marketing Paxil and other drugs. The government complaint cited the article on Study 329, saying it ”distorted the study results and gave the false impression that the study’s findings were primarily positive, when they were, in fact, primarily negative.” The following year, Glaxo agreed to pay a record $3 billion to settle criminal and civil charges.
Recriminations about Study 329 linger to the present. Responding to a damning critique in 2015, Dr. Keller and eight of his co-authors issued a letter defending the paper and denying it was ghostwritten. ”There was absolutely nothing about the process involved in the drafting, revision, or completion of our paper that constitutes ‘ghostwriting,’ ” they wrote. “This study was initiated by academic investigators, undertaken as an academic / industry partnership, and the resulting report was authored mainly by the academic investigators with industry collaboration.”
GSK also denied in a letter to one of the authors of the 2015 critique that the article was ”fraud, fraudulent or misleading.” Noting that it had undergone peer review, the company said the paper ”accurately reflects the honestly-held views of the clinical investigator authors.”
It’s hard to know
Since ghostwriting is meant to be invisible and is revealed, if ever, years after the fact, it’s unknown how common it is now.
“I don’t know of anyone or any study that can really quantify how big of a problem ghost authorship has been,” or ”how big of a problem it remains,” said Dr. Darren Taichman, executive editor of the journal Annals of Internal Medicine. “It’s very rare that editors are in a position to truly know who did what on a research paper.”
Taichman is also secretary of the International Committee of Medical Journal Editors, a working group drawn from 13 medical journals in the U.S. and abroad. Those publications and some others subscribe to a set of voluntary guidelines on such issues as disclosure of financial sponsorship by drug companies and authors’ conflicts of interest.
According to the guidelines, authorship credit requires making a substantial contribution to the conception, drafting and final approval of an article. In 2013, Taichman said, the committee added a requirement that authors accept full responsibility for the content of the paper. This was done, Taichman said, because too often when questions arose the authors would say, ” ‘I wasn’t responsible for that aspect of the paper. You need to ask someone else.”
The goal is to breed trust “that authors will stand behind what they’re putting their name to,” Taichman said. But despite the guidelines, he said, ”if someone wishes to hide his or her contribution to the writing of a paper, it’s going to be pretty difficult for an editor to know.”
Some industry critics say ghostwriting has evolved into a more ambiguous, and potentially insidious, pattern they call ”ghost management.” There is more pressure for credited authors to be substantially involved, and industry sponsorship and assistance from medical writers are uniformly acknowledged in footnotes. But some say drug companies remain largely in control of the message by selecting topics and producing outlines or first drafts.
“What’s changed is that companies have become more careful about not entirely ghostwriting an article,” Fugh-Berman said. “But they’ll make sure that their marketing messages are in it.”
Said Siwek, the former editor of American Family Physician, “At some point, I think they [drug companies] realized that they could get the same effect and bypass this pesky issue of looking bad and using ghostwriters by just making it transparent.”
Paradoxically, Siwek says, greater transparency has not necessarily improved the situation. Rather than instill more caution in doctors, the disclosures tend to make them more confident about the rest of the information. Transparency “does nothing to decontaminate the content; the spin is still there,” Siwek said in a 2016 letter to The BMJ.
“The real story in a nutshell is the contamination of the medical literature by flagrantly biased information,” Siwek said, aimed at “selling products and increasing profits rather than improving the health of the American people.”