Cochrane Collaboration's CAM Team Responds to Challenge of Economic Bias on Evidence
Written by John Weeks
Wednesday, 11 July 2007
Cochrane Collaboration's Complementary Medicine Team Responds to Challenge of Economic Bias; Cundiff Adds a Note on Conflicts
Summary: The Integrator article on a challenge of economic bias at the Cochrane Collaboration which ran in Medscape General Medicine stimulated a defense of Cochrane's policies, meant to "minimize financial conflicts," from Eric Manheimer, MD and Brian Berman, MD, who head up Cochrane's Complementary and Alternative Medicine effort. The Integrator asked David Cundiff, MD, the author of the Medscape commentary for his thoughts on the Manheimer/Berman defense, and for details on his disturbing charges of undisclosed conflicts of interest among Cochrane reviewers. One wonders whether even the most heroic efforts to diminish conflicts are sufficient in a world so saturated with Big Pharma's cash infusions.
Send your comments to
for inclusion in a future Your Comments article.
Cochrane's Complementary and Alternative Medicine Field at the University of Maryland School of Medicinesent the Integrator a clarification of Cochrane's policies on conflicts of interest. The CAM group is led by field administrator Eric Manheimer, MS and field convenor Brian Berman, MD. The Integrator then sent these comments to Cundiff for his perspective. I also requested from Cundiff details on the undisclosed conflicts of reviewers who he believe altered the outcome of the review in which he participated.
1. Cochrane Responds: "Several Safeguards Minimize Financial Conflicts ..."
"Greetings from the Complementary Medicine Field of the Cochrane
Collaboration. We were wondering whether you would consider printing the
following piece in the next edition of The Integrator Blog:
"This article seems to give the impression that the Cochrane
Collaboration is corrupted by financial conflicts of interest. While we cannot
comment on the specifics of the particular case between David Cundiff, MD, and
the Cochrane Peripheral Vascular Diseases Review Group, we would like to inform
The Integrator Blog readers that the Cochrane Collaboration takes
financial conflicts of interest very seriously and that the Collaboration has
several safeguards in place to minimize financial conflicts and thereby ensure
the integrity (real and perceived) of Cochrane Reviews. Most specifically, in
contrast with most print journals, sponsorship of a Cochrane review by any
commercial source or sources is prohibited.
"Furthermore, it is also prohibited
for a commercial source or sources to sponsor Cochrane entities that produce
Cochrane reviews, that is, Collaborative Review Groups. Further details of the
Cochrane Collaboration’s policy on commercial sponsorship are available on this webpage.
"The Cochrane Collaboration takes
financial conflicts of interest very seriously and that the Collaboration has
several safeguards in place to minimize financial conflicts and thereby ensure
the integrity (real and perceived) of Cochrane Reviews."
- Manheimer/Berman
"While it is challenging to prepare Cochrane Reviews – currently numbered at
over 3,000 – while also restricting funding sources, the Cochrane Collaboration
Steering Committee has recognized the potential for financial conflicts of
interest to distort findings of reviews, and has made every effort to minimize
such conflicts among the 50 different Cochrane Review Groups around the world
that coordinate the preparation of Cochrane Reviews."
Eric Manheimer, MS Cochrane Complementary Medicine Field Administrator
Brian Berman, MD Cochrane Collaboration Complementary Medicine Field Convenor”
Center for Integrative Medicine University of Maryland
School of Medicine Kernan Hospital Mansion Baltimore, Maryland
2. Cundiff's Perspective, Plus Details on 4 of 7 Cochrane Reviewers with Conflicts of Interest
"In response to the note from the
Cochrane CAM people, I say the following:
"I have the highest regard for the
principles on which the Cochrane Collaboration is based. I have not
investigated any Cochrane Reviews, other than ones involving
anticoagulation drugs, for bias and financial conflict of interest by
authors or editors.
David Cundiff, MD, Cochrane reviewer
"Kay Dickersin, PhD, US Cochrane Center Director, and
the Peripheral Vascular Disease Editor and Peer Reviewers received my
commentary about financial conflicts of the peer reviewers on my review of
anticoagulants for treatment of clots in the legs and lungs in September 2006
and still have not publicly responded. Meanwhile, each day over 200 people die from bleeding complications of anticoagulants of which roughly 5-10% are taking vitamin k inhibitors or heparins for venous thrombembolism.
"I am in the process of completing
an article titled, 'A Systematic Review of Cochrane Anticoagulation
Treatment Systematic Reviews.' So far, in 56 Cochrane Reviews, I have
found 13 editors/peer reviewers and 30 authors with financial conflicts of
interest. There is a pattern of methodological errors and biased interpretations of the data favoring the use of anticoagulants.
"I have found 13 editors/peer reviewers and 30
authors with financial conflicts of interest.
"There is a pattern of
methodological errors and biased interpretations of the data favoring
the use of anticoagulants."
- David Cundiff, MD
I asked Cundiff for information about his charge that at least 4 Cochrane reviewers, who substantially altered his work, had undisclosed conflicts of interest. He reiterated that "under 'Potential Conflicts of Interests' the only thing disclosed was the loss of my medical license due to stopping warfarin in a case." Then he offered the following.
"Details about the financial
conflicts of interest of four peer-reviewers for my review are the following:
Peer ReviewerMartin H. Prins, MD
Competing interest: Without disclosure to Cochrane Editor Dr.
Gerry Folkes or me, Martin H. Prins, MD, an epidemiologist from the Netherlands,
participated on the steering committees and in obtaining funding from
Santofi-Synthelabo and NV Organon drug companies for trials of fondaparinux
(Aristra) in treatment of deep venous thrombosis and pulmonary emboli. (1,2) Based
on the results of these two studies, the FDA approved fondaparinux for the
treatment of deep venous thrombosis and pulmonary emboli in May 28, 2004.
Anticipating this approval, Sanofi-Synthelabo announced the sale of Arixtra and
Fraxiparine (a low molecular weight heparin) on April 13, 2004 to
GlaxoSmithKline for about $360 million.(3)
Cundiff adds: 'The validity of relying on non inferiority trials depended on the assumption of proven efficacy and safety of standard treatment of deep venous thrombosis and pulmonary emboli with vitamin K inhibitors (e.g. warfarin) and heparins. This Cochrane venous thromboembolism review set out to test that assumption.'
Peer Reviewer David Bergqvist, MD
Competing interest: Dr.
Bergqvist has participated on advisory boards and/or research steeringcommittees for AstraZeneca, Aventis, Boehringer Ingelheim, Pharmacia/Pfizer,
and Sanofi-Synthelabo (4)
Peer Reviewer William Hiatt, MD
Competing
interest: Dr Hiatt served as a consultants to and received speakers' bureau
honoraria from the Bristol Myers Squibb/Sanofi-Synthelabo Partnership. (5)
Bristol Myers
Squibb owns Coumadin. Sanofi-Synthelabo markets enoxaparin, a low molecular
weight heparin.
Peer Reviewer Thomas Wakefield, MD
Competing
interest: As one of the 'PIOPED'
investigators, he co-authored, 'Multidetector Computed Tomography for Acute Pulmonary Embolism,' funded
by the US National Institutes of Health. (6) If anticoagulants don’t
work for venous thromboembolism, this and related studies are moot. He has
been an author on a study of the 'American Venous Forum' that is
funded by Sanofi Aventis (enoxaparin [Lovenox®, Clexane-a low molecular weight
heparin] and nadroparin [Fraxiparine®-a low molecular weight heparin]). (7)
At the University of Michigan Medical Center, he has an endowed chair as Professor
of Surgery in the Section of Vascular Surgery and directs the noninvasive
vascular laboratory. The need for the tests that this laboratory perform depend
on the assumption that anticoagulant therapy is indicated for venous
thromboembolism. One of his research studies, 'Inhibition of Deep Vein
Thrombosis -Induced Inflammation and Thrombosis by Melagatron,' is funded by
Astra Zeneca. (8)
Thanks for your interest. Best wishes,
David Cundiff, MD
Notes:
1. Buller HR, Davidson BL, Decousus H, et al. Fondaparinux
or Enoxaparin for the Initial Treatment of Symptomatic Deep Venous Thrombosis:
A Randomized Trial. Ann Intern Med. June 1, 2004;140(11):867-873 http://www.annals.org/cgi/reprint/140/811/867.pdf.
2. The Matisse Investigators. Subcutaneous
Fondaparinux versus Intravenous Unfractionated Heparin in the Initial Treatment
of Pulmonary Embolism. 0.1056/NEJMoa035451. N Engl J Med. October 30,
2003;349(18):1695-1702.
3. Sanofi-Synthelabo to Sell to GlaxoSmithKline
Arixtra(R), Fraxiparine(R) and Notre Dame de Bondeville Plant.
PRNewswire-FirstCall. April 13, 2004. Accessed July 29, 2006.
5. Hirsch AT, Criqui MH, Hiatt WR, et al. Peripheral
Arterial Disease Detection, Awareness, and Treatment in Primary Care.
10.1001/jama.286.11.1317. JAMA. September 19, 2001;286(11):1317-1324.
6. Stein PD, Fowler SE, Goodman LR, et al.
Multidetector Computed Tomography for Acute Pulmonary Embolism.
10.1056/NEJMoa052367. N Engl J Med. June 1, 2006;354(22):2317-2327.
7. McLafferty RB, Lohr JM, Wakefield TW, et al.
Results of the national pilot screening program for venous disease by the
American Venous Forum. J Vasc Surg. 2007;45(1):142-148.
Comment: My interest in this story was stimulated by an awareness of the Cochrane work which I gained directly from Berman during a session at the Rayburn House Office Building in 1999 in which we were both panelists. Berman
Maybe prospective reviewers should be given lie detector tests. Maybe failure to disclose should be elevated to white collar crime and get a reviewer jail time.
framed his data on the thinness of Cochrane-level evidence for
complementary and alternative medicine by reference to the awfully poor state of
evidence for much of conventional medicine. I have thought favorably of
Cochrane since, partly due not to evidence, per se, but to my respect for
Berman as explorer, clinician, researcher and educator.
What the Cundiff commentary suggests is that there may not be any clothes in which Cochrane can drape itself that can keep out the wind of profound pharmaceutical bias. There is the blunt issue of non-disclosure. Maybe prospective reviewers should be given lie detector tests. Maybe failure to disclose should be elevated to white collar crime and get a reviewer jail time. Try pushing that through a legislature. Considering the lives and dollars at stake, the idea makes sense.
Secondly, even with full and appropriate disclosure, the symbiotic relationship between Big Pharma and physician experts creates a nearly impossible terrain for the kind of clarity of thought and analysis on which we the public depend. Consider these lines which the poet Walt Whitman wrote 150 years ago in the Song of Myself:
" ... And these things tend inward to me, and I tend outward to them and such as it is to be of these, more or less I am and of these one and all, I sing the song of myself."
The way Big Pharma has tended inward to the core of the self of
medicine for the last three generations, no wonder that it is difficult
to extract unimpinged perspective from the conflict riddled body of
medicine. Cochrane's extra barriers to conflict of interest are admirable but insufficient.
To best serve the public, perhaps Cochrane
should consider tossing perspective reviewers into a centrifuge and see what, if anything, the truth serum
separator tubes are able to produce.
To best serve the public, perhaps Cochrane
should consider tossing perspective reviewers into a centrifuge and see what, if anything, the truth serum
separator tubes are able to produce. Short of that, we will always have the likelihood of finding, in Cundiff's words "a pattern of methodological errors and biased interpretations of the data."
There is one other strategy, and that is the strategy which the Institute of Medicine took toward complementary and alternative medicine when the IOM engaged its study, published in 2005. They simply refused to have anyone advise who has any relationship to the practices under review. Abhorrent as this was to the CAM fields - a blatant double standard, Cochrane might consider a strategy to always have reviewers from other specialties or field who have never had a single relationship with Big Pharma. Or maybe they should just give Cundiff a shot at leading a self-study team to examine how pervasive the pattern of misinformation and oversights may be.
Send your comments to
for inclusion in a future Your Comments article.
ShareThis
