Thursday, March 15, 2012

Logical Fallacies in Defense of the European Society of Cardiology's Lenient Approach to Conflicts of Interest

The European Society of Cardiology just published its recommendations for interactions between medical societies and industry.(1)   The report emphasized the supposed benefits of industry relationships and funding, and suggested few restrictions on current practices.  Like other defenses of leniency towards conflicts of interest, its arguments seemed to rest on a series of logical fallacies.  

Wonderful "Innovations ... Through Productive Collaboration" - Appeal to Belief and the Fallacy of Division

The report made several arguments about the benefits of professional - industrial interaction.  These included:
In recent decades, cardiology has been a fast-moving medical speciality. Many advances have come from basic and clinical research conducted by universities and by pharmaceutical and medical device companies. Innovations have been realized in part through productive collaborations between clinicians, academia, and industry. Such links are essential and need to be encouraged and supported by appropriate investment if medical progress is to be sustained.

The notion that interactions between physicians and industry is a source, perhaps the most important source of wonderful "innovations" is often used to justify the sorts of interactions, involving payments by industry to academia and physicians, that are now prevalent. In this report, this first appears as an assertion without any evidence to justify it.

Later, the report noted:
The Association of American Medical Colleges has stated that there are benefits from effective partnerships between industry and academic medical centres.

This statement came with a citation, to a 2008 report entitled "Industry Funding of Medical Education."(2) This AAMC report, in turn, contained many similar assertions, e.g.,
An effective and principled partnership between academic medical centers and various health industries is critical in order to realize fully the benefits of biomedical research and ensure continued advances in the prevention, diagnosis, and treatment of disease.
However, the AAMC report also failed to cite any evidence in support of these assertions.

There are two major problems with these assertions, which form the foundations of both reports. The first is they do not seem to have a clear evidence base.

While the notion that current research efforts marked by substantial interaction between physicians, academia and industry are producing wonderful innovations is appealing, what evidence there is suggests that clinically important innovation is rare. For example, I quote a 2009 review of global drug discovery by Light(3) on whether new drugs represent advances over older treatments:
the best evidence of clinical quality comes from systematic efforts to assess therapeutic advantage and adverse effects compared with existing drugs. A detailed analysis of therapeutic quality in new drugs over the past twenty years found that 14 percent of all new chemical entities are either therapeutic breakthroughs or substantially superior to existing medications. Likewise, a comprehensive review of all new drugs approved between 1989 and 2000 in the United States concluded that 14.8 percent were new chemical entities that provided significant clinical improvement, and a Canadian review board concluded that 10.7 percent of new chemical entities in 2000–2004 did so.
Thus while many new drugs are introduced, only a few introduced recently were chemically unique or had effects different than older treatments.  I would argue that most of these will prove not to be truly clinically innovative, in that good clinical research will not show that they produce substantial improvements in clinical outcomes without major adverse effects for a good number of patients with not uncommon problems.  (It is not hard to think of really important innovations developed before the era of major industry-academic-physician interactions: antisepsis for surgery, anesthesia for surgery, antibiotics, hormone replacement therapy with insulin, thyroid hormone, etc, smallpox vaccination, polio vaccination, etc, etc)  However, it is actually very hard to think of more than a handful of new drugs or devices discovered in the last 20 years which were that hugely innovative.  (The closest might be multi anti-viral drugs used to treat HIV, and Gleevec for chronic myelogenous leukemia.)  If there has been very little really important innovation in the last 20 plus year era of enhanced industry-academic-physician interactions, these collaborations could not have produced tremendous innovation. 

So the report's arguments rests on an assertion that is not clearly justified, and which appears to be at best a huge exaggeration. Resting an argument on a belief that is not further supported by evidence amounts to a obvious logical fallacy.  It is an appeal to belief.

Furthermore, while both reports emphasize physician-industry collaboration or interaction, there are many ways interaction or collaboration can occur without involving substantial payments by the latter to the former. It is possible for two people or organizations to work together without one paying the other.  Yet both reports use the broad assertions about interaction and collaboration to justify interactions and collaborations in which industry makes substantial payments to physicians or academic institutions. This appears to be an example of the fallacy of division, that is, an unjustified assertion that "what is true of a whole must also be true of its constituents."

CME Would Wither if Financing Were Reduced, and Industry is the Only Possible Source of Such Money - A Slippery Slope and a False Dilemma

The ESC report made the argument that without industry funding, CME would wither. For example,
Should Europe choose to follow the strategy proposed in the USA, severing links between industry and medical societies, CME could be severely compromised. Relying completely on public funding is not a viable option for Europe at the moment. The removal of industry support for medical associations would be followed by increased fees and reduced attendance at congresses especially by clinical trainees and young fellows. It is the view of the ESC that in the absence of alternative funding, or until alternative funding is identified, maintaining links with industry is appropriate....

The argument is that CME would fail if financial support for it would be reduced, and that industry is the only possible source of financial support at the current level. The notion that CME must inevitably fail if financing of it were reduced is a slippery slope fallacy, that is, a statement that a inevitably causes b when such inevitability is not proven. Why could not adequate CME be done at a lower cost, even if the result were less luxurious? Furthermore, the notion that industry is the only source of funding is a false dilemma fallacy. In fact, there are other possibly sources of funding. Given that physicians are among the world's best paid professionals, why could they not pay for their own CME?

"Conflicts of Interest are Unavoidable and ... They Cannot Be Abolished" - False Dilemma

The third major argument on which the report bases its recommendations is encapsulated above, and was buttressed by:
The risk of bias in medical education is not restricted to activities that are supported by industry. It can affect any type of scientific communication, even an educational meeting organized independently by a university or medical association.

Underlying these assertions seem to be extremely broad definitions of bias and conflicts of interest, coupled with unwillingness to see a distinction between trivial and serious varieties of each. This again appears to be a false dilemma. The distinctions should be between no conflicts, trivial conflicts, and serious conflicts, not between no conflicts and any conflicts, even if trivial.  This also could be called an example of how "the perfect is the enemy of the good."  Even if perfectly preventing all conflicts of interest is impossible, does this imply that preventing serious conflicts of interest is not worthwhile.   


We have noted that logical fallacies are increasingly deployed to defend the status quo in health care, and particularly to defend the interests of those who are profiting the most from the current dysfunctional system.  We have noted that several defenses of the conflicts of interest generated by financial relationships between physicians and medical academics on one hand and commercial health care firms on the other, were based on logical fallacies.  (See examples here, herehere, and here.)  I have yet to see a coherent, logical, fact-based argument that the benefits for patients' and the public's health of physicians and medical academics working part-time as consultants, advisers, speakers, and directors of health care corporations outweigh the obvious risks of biasing medical decision making, education and research in favor of vested interests.

So we add to our ongoing series how, based on a series of logical fallacies, the European Society of Cardiology provided a series of recommendations that allowed nearly any kind of relationship among CME speakers and selection committees and industry, as long as the relationships were disclosed.  The only relationships banned were those "which would represent a significant conflict of interest" for the Chairperson of a Congress Programme Committee.  Similarly, the only stipulations for the society's cardiology journals were that interests affecting editors and editorial board members must be declared, and board members and reviewers should decline reviews of manuscripts "relating to topics, drugs, or devices, in which they have significant commercial of academic interests."  The rules for guideline committees were somewhat more rigorous, but "receipt of consultancy fees or fees for lecturing would not debar an individual from being a member of a committee but must be fully disclosed."

It is discouraging that the web of conflicts of interest that currently enmeshes much of academic medicine and many medical professionals is so heavily defended.  It is more discouraging that its defenders include so many prominent academics and practicing physicians.  It is more discouraging that so many well trained people resort to logical fallacies to make their arguments, and do so in prestigious scholarly journals.

Our continuing series about how logical fallacies are used to support the status quo and the powers that be in health care suggests, if nothing else, that health care professional education ought to include courses in logic.

Finally,  in 2011, I noted, "I have also yet to see an argument in favor of conflicts of interest made by anyone who does not have such conflicts."At least, however, up to that point I had not noted any such arguments made by people who had much power to enforce their views, as opposed to the ability to just express them.  Last month, however, I discussed how the leader of one of the most acclaimed US medical schools made an argument in support of conflicts of interest based on logical fallacies,  Now we have just seen such arguments made by the leaders of European cardiology.  The new paper suggested that disclosure is the best way to manage conflicts of interest.  True to this belief, the paper included a disclosure section that took up an entire journal page.

It seems likely that number and magnitude of ongoing commercial interests so disclosed may have influenced the content of the position paper.  Yet while it may be unsurprising, it is most disappointing that conflicts of interest are now being uncritically and illogically publicly defended by people in positions to exert so much influence on health care.

The noted cognitive psychologists George Loewenstein, Sunita Sah, and Daylian Cain just asserted in JAMA(4):
Conflicts of interest, including fee-for-service arrangements, are at the heart of the astronomical increases in health care costs in the United States, and transparency is not substitute for more substantive reform.
True health care reform requires such substantive reform of the financial arrangements among corporations that sell health care services or products and health care professionals, others who make decisions about patients' or the public's health, and academic health care institutions. To decide how to accomplish such reform, we need a better discussion informed by logic and evidence, sans logical fallacies. Those who lead health care ought to be able to participate in this discussion under these conditions.

1. ESC Board. Relations between professional medical associations and the health-care industry, concerning scientific communication and continuing medical education. Eur Heart J 2012; 33: 666-674. Link here.
2. Association of Americian Medical Colleges. Industry Funding of Medical Education. Washington, DC: AAMC, 2008. Link here.
3. Light DW. Global drug discovery: Europe is ahead. Health Aff 2009; 28: w969-w977. Link here.
4. Loewenstein G, Sah S, Cain DM. The unintended consequences of conflict of interest disclosure. JAMA 2012; 307: 669-670. Link here.

1 comment:

Afraid said...

Nothing new or unexpected here. And I don't think we the people really have any trust in institutions to tell the truth anymore. The web of COI is too tangled.

Individuals can tell the truth, institutions have no heart, no ability to deciper right from wrong. Institutional survival is the natural selection factor most at work at the inhuman level of evolution.

So when a human gets control of an institution and governance is upended, this sort of thing would automatically happen. The institution becomes only as moral as the individual in charge.