About these ads

Retraction Watch

Tracking retractions as a window into the scientific process

Coming clean: A major figure in cardiology publishes a lengthy conflict of interest correction in JAMA

with 13 comments

Authors’ financial disclosures can be a thorny issue for scientific journals.  There’s often confusion over just what should be listed as a conflict of interest, and when relationships are revealed after papers are published, lack of disclosure sometimes leads to corrections.

For example, the Journal of Cell Science recently published this:

CRSBP-1/LYVE-1 ligands disrupt lymphatic intercellular adhesion by inducing tyrosine phosphorylation and internalization of VE-cadherin

There was an error published in J. Cell Sci. 124, 1231-1244.

Jun San Huang and Shuan Shian Huang had an equity position in Auxagen Inc. during the time the research was carried out and should have declared that there is a conflict of interest.

Part of the research described in the article was supported by NIH grants awarded to St Louis University (AA 019233 to J.S.H.) and to Auxagen Inc. (HL 087463-01 and DK 078438 to J.S.H.).

The authors apologise for this omission.

The rules can sometimes trip up even the most prominent figures in medicine, it turns out. This week’s Journal of the American Medical Association (JAMA) includes this lengthy letter from Eugene Braunwald, whose 1000-plus publications and textbook tower over the field of cardiology:

It has been brought to my attention that there were differences in my financial disclosures in a number of articles recently published in JAMA, and this warrants explanation.

The original research report by Preiss et al1 dealt with statins, and in that article I reported that the TIMI Study Group in which I work has received research support through the Brigham and Women’s Hospital from Merck, Astra-Zeneca, and Pfizer, each of which manufactures and sells a statin.

In a Commentary on the role of digoxin,2 Dr Gheorghiade and I mention and comment on a wide variety of drugs used in the treatment of heart failure (beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, aldosterone-blocking agents). Because 1 or moreof these agents are manufactured by virtually every pharmaceutical company, I listed all with which I had any connectionat that time.

A more recent Commentary was quite general.3 It focused on a pathophysiological approach for assessing patients with heart failure. The drugs mentioned are generic—nitroglycerin, dobutamine, and esmolol—and I did not perceive a possible conflict of interest. In a recent Editorial on reductions in heart failure hospitalizations, 4 Dr Gheorghiade and I mention eplerenone and tolvaptan. The former is manufactured by Pfizer, from which we no longer receive any support. The latter is manufactured by Otsuka, with which I have had no contact. Again, I did not perceive any conflict of interest.

However, the relationships present during the 36 months prior to submission of the latter 2 articles include research support from Merck, AstraZeneca, Johnson & Johnson, Beckman Coulter, Eli Lilly, Roche Diagnostics, sanofiaventis, Daiichi Sankyo, GlaxoSmithKline, Bristol-Myers Squibb, and Pfizer as well as participation in symposia, advisory board meetings, and/or consultancies for the following companies: Eli Lilly, Merck, Genzyme, Amorcyte, CVRx (no compensation), The Medicines Company, CV Therapeutics, Daiichi Sankyo, MC Communications, Ortho McNeil, Ikaria, Menarini International, CardioRentis, and sanofi-aventis.

These were not listed because I did not consider them to be relevant to these 2 articles. I did not understand that all financial interests, regardless of their perceived relevance, had to be disclosed. I now understand the requirements for financial disclosure.

I hope that this clears up any possible misunderstanding.

The fact that Braunwald’s disclosures required corrections is perhaps a good indication that even people who’ve thought about this issue a lot can still get lost. Braunwald, after all, chaired a committee that recommended limiting doctors’ outside income at Harvard teaching hospitals. The JAMA editors write that the lack of disclosure wasn’t all due to Braunwald’s confusion:

We appreciate Dr Braunwald providing his professional and transparent explanation to clarify the differences in the reported financial disclosures among several of his recent publications in JAMA.We fully recognize that determining relevant financial relationships involves judgment and that reporting of financial disclosure information certainly is not an exact science. In fact, in one of the Commentaries1 that Dr Braunwald mentions, his coauthor, Dr Gheorghiade, had reported to JAMA that he had received consulting fees from Bayer, Novartis, Sigma Tau, Johnson&Johnson, Takeda, Otsuka, and Medtronic. However, these disclosures inadvertently were omitted from the published article. The editors apologize for this oversight and regret this error. We urge all authors to fully disclose all possible conflicts of interest as recommended by the International Committee of Medical Journal Editors. A correction regarding the clarification of conflict of interest disclosures for Drs Braunwald and Gheorghiade appears in this issue.

For reference, here is the meat of JAMA’s conflict of interest policy:

Conflicts of Interest and Financial Disclosures. A conflict of interest may exist when an author (or the author’s institution or employer) has financial or personal relationships or affiliations that could influence (or bias) the author’s decisions, work, or manuscript. All authors are required to complete and submit the ICMJE Form for Disclosure of Potential Conflicts of Interest. In this form, authors will disclose all potential conflicts of interest, including relevant financial interests, activities, relationships, and affiliations (other than those affiliations listed in the title page of the manuscript), including

  • Any potential conflicts of interest “involving the work under consideration for publication” (during the time involving the work, from initial conception and planning to present),
  • Any “relevant financial activities outside the submitted work” (over the 3 years prior to submission), and
  • Any “other relationships or activities that readers could perceive to have influenced, or that give the appearance of potentially influencing” what is written in the submitted work (based on all relationships that were present during the 3 years prior to submission).5

Of potential interest to Retraction Watch readers is that Braunwald was involved in what was likely the largest number of retractions by a single author before Joachim Boldt. John Darsee retracted some 82 papers, 30 of which were published while he was at Harvard working with Braunwald.

About these ads

Written by Ivan Oransky

February 7, 2012 at 4:00 pm

13 Responses

Subscribe to comments with RSS.

  1. You know, I can see how someone in his position, with so many different projects with funding from various sources, and changing levels of involvement and consultation with drug companies, can end up in a situation like this. I also see how he came to the conclusion of not having to state conflicts of interest for some of those articles. It really highlights how a conflict of interest is also a subjective assessment, but I guess with things as they are now the best option and take home message from this, is to just put everything out there, whether you think it’s relevant or not?

    I think his clarification is very up front and transparent – if anything, we need more scientists doing the same.

    Also, I had to laugh at the editors accidentally ‘dropping’ a conflict of interest statement off his colleagues article! Ahh, woops.

    Chris

    February 7, 2012 at 4:52 pm

    • This also indirectly raises another issue. Up until Darsee was exposed in the early 1980’s, Dr Braunwald published about 12 primary papers per year. Darsee presumably included Dr Braunwald’s name as an author to increase the chances that the papers were accepted for publication. From the late 1980’s until now, the number of primary papers authored by Dr Braunwald has grown from ~8 to around 20 per year. That he can maintain this output at an age of 83 raises the issue of whether including the name of an eminent researcher among the authors of a paper provides undue influence on the likelihood of its publication. Would double-blind peer review, where the reviewers aren’t told the authors’ names or institutions, help prevent this potential source of bias?

      Michael Briggs

      February 7, 2012 at 6:13 pm

      • Perhaps Dr. Braunwald is able to keep up such prolific publication by being listed as a ‘guest’ or ‘honorary’ author on some of these studies? This possibility may also help to explain his apparent confusion over who funded what. It’s very simple, really: you’re either taking money from drug and/or device companies, or you’re not. A September 2009 survey reported in the Journal of the American Medical Association confirmed that the prevalence of both ‘honorary’ and ‘ghost authors’ in medical journal articles was “still a concern”. For example:

        “An ‘honorary’ or ‘guest’ authorship is often something ‘bestowed as a tribute to a department chair or to the person who acquired funding for the study.’”

        This widespread custom of academia also, not surprisingly, can indeed help to ensure publication. But isn’t bestowing a “tribute” like guest authorship to a prominent academic a lot like university students fraudulently putting their own names on purchased internet essays? The similarity: neither actually has to write them. The difference: the students are punished for their fraud. Academics are honoured for it.

        As a person with 37+ years experience in the public relations field, I caution those who are “confused” about things like how to accurately report their financial conflict of interest disclosures to be very cautious from here on in. It’s not only wrongdoing that attracts suspicion – it’s the PERCEPTION of wrongdoing. Some academics still insist that they actually see absolutely nothing wrong with being on the take from drug and device makers, with signing their names as guest authors to submitted papers, with being involved (as Dr. Braumwald has been) in a record number of journal retractions. As we have increasingly seen, some may be clinging to this insistence at grave peril to their ultimate reputation and legacy.

        More on this at “Medical Ghostwriting and Guest Authorship: Twins Separated at Birth?” – http://ethicalnag.org/2009/09/22/guest-athors/

        Carolyn Thomas

        February 7, 2012 at 9:54 pm

      • I really like the idea of a double-blind review, but I’m pretty sure that it’s unworkable in most instances. Fields and subfields tend to be small, with a small(ish) number of active researchers working on related problems. That usually means that reviewers can often identify the corresponding author based on techniques, ideas, or writing style.

        sfs

        February 9, 2012 at 2:22 pm

  2. @Carolyn Thomas Excellent advice. (I would have thought that if anyone would be aware of the dangers of accepting honorary authorships, it would be Dr Braumwald.)

    Michael Briggs

    February 8, 2012 at 8:45 am

  3. Part of the issue here stems from the fact that reporting requirements for conflict of interest are so variable between institutions. Also the NIH rules (when applying for funding) are a moving target. Until fairly recent there was a general consensus at $10k as the cut off for a significant vs. non significant COI. Nowadays though, go to any scientific/medical conference and even simple things like a cup of coffee or a free pen require documentation. Many institutions adopt a zero-tolerance rule now.

    Also there’s a delineation to be made between money for research, and income for consulting. Sitting on the SAB of a pharma company, and getting paid $10k into your personal bank account for a couple days on-site advising their scientists, is different from receiving $10k into a discretionary account to pay for basic research done in your lab’ to test the company’s drugs. That same $10k to support the salary of a nurse in a clinical trial is very different again. It’s debatable which one constitutes a larger COI – arguably the personal consulting money should not influence your research, so the requirement to report should be less, but most people don’t see it that way.

    In the end, there’s a very fine line between COI, and perceived COI, and really the only solution is just to declare everything and let people make up their own minds. In Braunwald’s case, better late than never.

    Virgilstar

    February 8, 2012 at 9:53 am

  4. What about patents? I’ve seen publications and reviewed manuscripts where one or more of the “academic” authors has filed a patent application for the drug they are reporting on but do not claim any conflict of interest. I point these out to editors when I review papers but what is the correct course of action? The patent may not have commercial value at the time of publication but the publication itself then looks like an advertisement for pharma to license the patent.

    BrainsRus

    February 9, 2012 at 7:44 am

    • Unfortunately in such cases, the journal actually has a far bigger conflict of interest. Journals make a LOT of money selling paper reprints to pharma rep’s to hand out at meetings and to doctors. Thus, if the drug is “in the pipeline”, then it can end up generating significant revenue for the journal, so they’re unlikely to make things difficult for the authors by demanding full disclosure of patents/IP.

      Virgilstar

      February 9, 2012 at 1:52 pm

      • It makes me wonder if the journals should be somewhat liable for patient damages. Hrm… there are probably a lot of unintended, unpalatable consequences attached to that idea.

        On the other hand, open-access journals shouldn’t have these kind of conflicts.

        sfs

        February 9, 2012 at 2:17 pm

      • Absolutely correct. The prestigious NEJM sold 929,400 reprints of a single (industry-funded) Vioxx-friendly ‘research’ article they’d published – mostly sold directly to the now-discredited drug’s own manufacturer Merck. Merck’s sales reps then distributed these reprints to physicians on their daily call routes as part of the aggressive Vioxx sales pitch. Reprint orders from this one ghostwritten Vioxx article brought in more than $697,000 in cash for the NEJM. Medical journal editors who don’t want to bite the Big Pharma hand that feeds them apparently learn to hold their noses and accept this.

        And journals also accept up to $500 million worth of full-page drug ads placed by Big Pharma every year. A cozy relationship indeed.

        Carolyn Thomas

        February 10, 2012 at 10:46 am

  5. @Carolyn Thomas: While I agree with the spirit of your message, I would advise you to be very cautious in posting accusations that someone has engaged in ghostwriting activities, since this is now widely recognized to be tantamount to fraud in academic circles. I don’t know, since he’s not in my field, whether Dr. Braunwald has ghostwritten papers or not. However, if he has not, you are potentially exposing yourself to a libel lawsuit by posting this online. Again, I absolutely agree with the general content of your post.

    DefendSmallScience!

    February 10, 2012 at 3:09 pm

    • While industry-funded medical ghostwriting is now (finally!) considered fraudulent, the widely accepted practice of “guest” or “honorary” authorship is still alarmingly common in academia. I’m certainly not “accusing” Dr. Braunwald or other authors mentioned here of any ghostwriting activities, but merely questioning his “confusion” around the simple definition of financial conflict of interest, coming as it does from an academic associated with, as Ivan points out here, “the largest number of (journal) retractions by a single author before Joachim Boldt”.

      Carolyn Thomas

      February 19, 2012 at 10:16 am


We welcome comments. Please read our comments policy at http://retractionwatch.wordpress.com/the-retraction-watch-faq/ and leave your comment below.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 35,982 other followers

%d bloggers like this: