StATS: Manipulation of peer-review publications by pharmaceutical companies (March 9, 2007)

This weblog entry provides some discussion points that I want to share in a seminar:

I first noted this topic in a weblog entry late last year,

but have not had time to follow up on it until now.

Two years ago, Richard Smith, a former editor of BMJ, published an essay in the Public Library of Science, with the provocative title:

Dr. Smith was not the only one with harsh words for the drug companies. Marcia Angel, a former editor of the New England Journal of Medicine, wrote a book with an equally provocative title:

Another former editor of the New England Journal of Medicine, Jerome Kassirer, published a book,

Richard Horton, editor of Lancet, also criticized pharmaceutical firms in an article:

Why are all these editors of prominent journals so hostile towards drug companies? They are angry because they have seen repeated examples of how drug companies have manipulated the peer-review process to produce research findings that help them market their drugs. Peer-reviewed research can market a drug much more powerfully than advertising can. Dr. Smith notes

A large trial published in a major journal has the journal’s stamp of approval (unlike the advertising), will be distributed around the world, and may well receive global media coverage, particularly if promoted simultaneously by press releases from both the journal and the expensive public-relations firm hired by the pharmaceutical company that sponsored the trial. For a drug company, a favourable trial is worth thousands of pages of advertising, which is why a company will sometimes spend upwards of a million dollars on reprints of the trial for worldwide distribution.

How can drug companies manipulate research to make their drug look good (or to make a competitor's drug look bad)? Dr. Smith has a good list:

Is there a common theme to this list? A lot of these abuses involve selective reporting: publish the good news and sweep the bad news under the rug.

Dr. Smith cites the Olivieri case (reference #11)--is anyone familiar with that case? The Canadian Association of University Teachers posted an investigation about this case. "The case of University of Toronto clinician, Dr. Nancy Olivieri, gained attention when her research at the Hospital for Sick Children led her to believe that a new drug treatment posed dangers to some patients. It is alleged that the hospital and the university failed to come to her defence when Apotex, co-sponsor of the research, objected to her publishing her findings. It is further alleged that hospital and university officials and representatives of Apotex variously subjected her to workplace and other harassment." (www.caut.ca/en/issues/academicfreedom/olivierireport.asp)

What other instances of suppression of research can you cite? There is some evidence that research about problems with COX-2 inhibitors (generic names: valdecoxib, celecoxib, and rofecoxib, brand names: Bextra, Celebrex, and Vioxx ) was withheld. The details are difficult to follow, but here is the Wikipedia commentary about rofecoxib:

Months after the preliminary version of VIGOR was published in the New England Journal of Medicine, the journal editors learned that certain data reported to the FDA was not included in the NEJM article. Several years later, when they were shown a Merck memo during the depositions for the first federal Vioxx trial, they realized that this data had been available to the authors months before publication. The editors wrote an editorial accusing the authors of deliberately withholding the data (Curfman et al, 2006a). They released the editorial to the media on December 8, 2005, before giving the authors a chance to respond. NEJM editor Gregory Curfman explained that the quick release was due to the imminent presentation of his deposition testimony, which he feared would be misinterpreted in the media. He had earlier denied any relationship between the timing of the editorial and the trial. Although his testimony was not actually used in the December trial, Curfman had testified well before the publication of the editorial.[3]

The editors charged that "more than four months before the article was published, at least two of its authors were aware of critical data on an array of adverse cardiovascular events that were not included in the VIGOR article." This additional data included three additional heart attacks, and raised the relative risk of Vioxx from 4.25-fold to 5-fold. All the additional heart attacks occurred in the group at low risk of heart attack (the "aspirin not indicated" group) and the editors noted that the omission "resulted in the misleading conclusion that there was a difference in the risk of myocardial infarction between the aspirin indicated and aspirin not indicated groups." The relative risk for myocardial infarctions among the aspirin not indicated patients increased from 2.25 to 3 (although it remained statitistically insignificant). The editors also noted a statistically significant (2-fold) increase in risk for serious thromboembolic events for this group, an outcome that Merck had not reported in the NEJM, though it had disclosed that information publicly in March 2000, eight months before publication. (Curfman et al., 2006b, Supplementary Material).

The authors of the study, including the non-Merck authors, responded by claiming that the three additional heart attacks had occurred after the prespecified cutoff date for data collection and thus were appropriately not included. (Utilizing the prespecified cutoff date also meant that an additional stroke in the naproxen population was not reported.) Furthermore, they said that the additional data did not qualitatively change any of the conclusions of the study, and the results of the full analyses were disclosed to the FDA and reflected on the Vioxx warning label. They further noted that all of the data in the "omitted" table was printed in the text of the article. The authors stood by the original article. (Bombardier et al., 2006).

NEJM stood by its editorial, noting that the cutoff date was never mentioned in the article, nor did the authors report that the cutoff for cardiovascular adverse events was before that for gastrointestinal adverse events. The different cutoffs increased the reported benefits of Vioxx (reduced stomach problems) relative to the risks (increased heart attacks). (Curfman et al., 2006b).

Some scientists have accused the NEJM editorial board of making unfounded accusations.[4], [5] Others have applauded the editorial. Renowned research cardiologist Eric Topol [6], a prominent Merck critic, accused Merck of "manipulation of data" and said "I think now the scientific misconduct trial is really fully backed up" [7]. Phil Fontanarosa, executive editor of the prestigious Journal of the American Medical Association, welcomed the editorial, saying "this is another in the long list of recent examples that have generated real concerns about trust and confidence in industry-sponsored studies" [8].

On May 15, 2006, the Wall Street Journal revealed that NEJM produced the editorial on the advice of a public relations firm that "the rebuke would divert attention to Merck and induce the media to ignore the New England Journal of Medicine's own role in aiding Vioxx sales."[9] The Journal accused NEJM of misleading reporters into believing that Merck had deleted data regarding the three additional heart attacks, rather than a blank table that contained no statistical information.[10]

Additional commentary about conflict of interest on a DSMB monitoring a Vioxx trial, see

There are also allegations of problems with celecoxib (Celebrex). A front page WSJ article had this commentary:

One well-publicized dispute over data interpretation came in 2000 when a JAMA article said Pfizer Inc.'s painkiller Celebrex minimized damage to the stomach compared with older drugs. It later emerged that the authors used only six months of data even though they had some data extending to 12 months. When all the results were included, Celebrex didn't look markedly better than its rivals. (Separately, a study last year suggested Celebrex might increase heart risk.) Today, the 2000 article is part of a shareholder suit alleging that Pfizer misled investors about its drug.

"We were burned very badly," says Catherine DeAngelis, JAMA's editor in chief. She says the journal got tougher as a result. Today, "the single thing we change most often is the conclusion," she says. "It comes in as, 'This product is the greatest thing,' and we say, 'Under these circumstances, in this population, this medication seems to control a, b and c.' "

In an email, a Pfizer spokeswoman said the authors of the Celebrex study "presented the data in the manner that they felt was the most meaningful." Many patients discontinued the treatment after six months, "making analysis of data beyond six months difficult," she said.

Worrisome Ailment in Medicine: Misleading Journal Articles. Editors Demand More Data
To Ensure Full Disclosure Of Drug Risks, Trial Gaps. Sarbanes-Oxley for Professors
. Anna Wilde Mathews, The Wall Street Journal, , May 10, 2005

Do drug companies have a responsibility to report all results, both good and bad? In 2004, Elliot Spitzer charged GlaxoSmithKline's  "with 'repeated and persistent fraud' in concealing the results of studies that suggested that paroxetine was ineffective in treating depression in adolescents." (O. Dyer. Bmj 2004: 328(7453); 1395. [Medline] [Full text] [PDF]).

Is there another theme to this list? Another theme is to make your drug look good in comparison to a bad control group, a control group that is treated suboptimally. This is bad science, but also violates what important ethical principle?

What are some protections against research manipulation?

Dr. Smith lists some well publicized reforms:

Editors can review protocols, insist on trials being registered, demand that the role of sponsors be made transparent, and decline to publish trials unless researchers control the decision to publish

but suggests that this is not enough.

Dr. Smith's first suggestion has been made before, but to my knowledge, it has not been implemented at most journals. Have you ever served as a peer-reviewer? Would it have helped for you to review the original research protocol alongside the research publication? What sort of abuses would this prevent?

This page was written by Steve Simon while working at Children's Mercy Hospital. Although I do not hold the copyright for this material, I am reproducing it here as a service, as it is no longer available on the Children's Mercy Hospital website. Need more information? I have a page with general help resources. You can also browse for pages similar to this one at Category: Conflict of interest.