When the Advocates of Full Disclosure Go too Far
“I believe that, in order to protect our constitutional rights, everyone should refuse to permit inquiry into his political beliefs.” -- Published testimony of Linus Pauling to California Senate Investigating Committee on Education in 1950
In the 1950s, during the McCarthy communist witch hunts, the U.S. Public Health Service revoked grants it had awarded Linus Pauling, the intellectual father of molecular biology, because of his opposition to demands that scientists disclose their political associations.
Half a century later, the International Committee of Medical Journal Editors (ICMJE) has promulgated a new Uniform Disclosure Form for Potential Conflicts of Interest that should cause Pauling to turn over in his grave. In addition to requiring information about financial interests, Section 5 of the form asks authors of journal articles to “report any personal, professional, political, institutional, religious, or other associations that a reasonable reader would want to know about in relation to the submitted work.”
Journals that have adopted the new disclosure policy include the New England Journal of Medicine, the Annals of Internal Medicine, The Lancet, the Journal of the American Medical Association, and the British Medical Journal.
Beyond the obvious and profound assault on privacy and liberty, the notion that personal beliefs, as signaled by political or religious affiliations, are relevant when assessing peer-reviewed research, is an assault on science itself.
The scientific enterprise is built on the idea that theories rise and fall on objective criteria. Experiments can be independently reproduced — or they can’t. A hypothesis is testable — or it isn’t.
This is why, in the 21st century, science has emerged as a truly global enterprise. Collaborations flourish without regard to national borders, creeds or ideologies. Indeed, in an ideal world, peer reviewers of articles about research would be blinded to an author’s identity, so judgments could be made solely on the merits of the science.
But let’s imagine the ICMJE world.
BioCentury asked Christine Laine, editor of the Annals of Internal Medicine and a member of the ICMJE secretariat, why it would be legitimate or necessary to require disclosure of a religious affiliation by the author of an article describing clinical or basic research.
Laine replied that if an author of a study revealing safety problems with a contraceptive was an official in a right-to-life organization, a “reasonable reader” would want to know that information to evaluate the research. If the author were a devout Roman Catholic, disclosure of her or his religious views wouldn’t be mandatory, “but it might be that the person would want to disclose it,” she said.
Leadership roles in advocacy organizations, even if they are uncompensated, must be disclosed under the new standards, Laine said.
By those standards, a “reasonable” reader would want to know that a scientist who has published on evolution, genetic testing, embryonic stem cells and similar issues was a born again Christian and the founder of a foundation dedicated to reconciling religion and science. This “reasonable” reader thus would expect articles by NIH Director Francis Collins would be embellished with a footnote indicating that he is the founder of the BioLogos Foundation, which “emphasizes the compatibility of Christian faith with scientific discoveries about the origins of the universe and life.”
By the same token, it would be “reasonable” to expect that any paper written by molecular diagnostic pioneer Randy Scott, executive chairman of Genomic Health Inc., would also be asterisked because he is chairman of BioLogos.
Collins hasn’t been shy about disclosing his religious associations, but a young scientist just starting a career in genetics would be foolish, indeed, to advertise similar associations.
It is hard to believe that peer reviewers and editors, the gatekeepers to publication in prestigious journals, won’t be influenced by knowledge of an author’s association with an organization they find repugnant.
That researcher also would have to be quite brave to keep her associations private because it is unclear what needs to be disclosed under the new rules. ICMJE has intentionally kept the standard for reporting non-financial associations ill-defined, Laine said, “to cover all the bases.” A narrower definition, she said, would allow authors to find loopholes.
Ironically, an example of a situation where the author’s potential conflict of interest would have been of interest to readers — but wasn’t mentioned by the journal — appeared in the New England Journal of Medicine on Oct. 14, the day after NEJM announced the new ICJME disclosure policy.
NEJM published a commentary about intellectual property policies related to biosimilars, and informed its readers that “no potential conflict of interest to this article was reported.”
The journal apparently didn’t feel it was necessary to note that the lead author, Alfred Engelberg, is the former counsel to the Generic Pharmaceutical Industry Association, or that he represented generic companies in suits against brand drug companies.
Reasonable readers might also want to know that Engelberg is an advisor to Sen. Chuck Schumer (D-N.Y.), the sponsor of legislation that seeks to implement the policies recommended in Engelberg’s commentary.
NEJM spokesperson Jennifer Zeis told BioCentury the non-financial disclosure policy wasn’t in place when the Engelberg commentary was submitted. This is a thin umbrella to hide under, as there is a clear financial interest in this industry dispute. And NEJM arguably should have known that transparency should have been considered in the case of commentaries and editorials, which by definition comprise opinions that are not subject to recapitulation and validation via the scientific method.
Every journal sets its own punishments for failure to make appropriate disclosures. According to Laine, the Annals of Internal Medicine publishes a correction in the journal that is permanently attached to PubMed citations, and it often sends letters about the incident to the department chair and dean of an academic author’s university.
While they are so busy guarding the gates of scientific integrity, maybe the ICMJE editors have not gone far enough.
If the political and religious affiliations of study authors are somehow relevant to the scientific enterprise, why shouldn’t those who conduct peer reviews have to disclose their religious and political beliefs as well?
And does anyone who has spent time in academia really believe that the political or religious beliefs of either study authors or peer reviewers are really more important than who denied whom tenure, who trumped whom for an NIH grant, who got the chairmanship of the medical faculty, or whether a study author was a post-doc for the study reviewer?
Let’s disclose all that as well. In fact, let’s see if we can make the entire scientific enterprise grind to a halt by simply declaring everyone conflicted and lacking in integrity.
Advocates for disclosure of non-financial associations may come up with elaborate hypothetical scenarios to justify the requirement, but these can’t come close to balancing the very real harm that the practice will do to individual scientists and to the scientific enterprise.
Linus Pauling provided more than scientific leadership — he promoted civic responsibility. Like him, every study author will need to decide how he or she will stand up for the bedrock values of democratic societies.
comments powered by Disqus
Andrew D. Todd - 11/24/2009
The problem about medical science is that it depends on human subject research. The names of the subjects are of course not published, so a skeptic cannot go and examine them. Independently verifying the experiment would involve doing whatever was done, to additional people. Humans are not laboratory rats. There are certain areas where the best available information was obtained by Nazi doctors in concentration camps, transgressing where no one had transgressed before, and where no one has transgressed since. Of course, in 1945, some, a few, of the Nazi doctors were properly hanged, by way of admonition. The principle of experimental replication falls foul of the Hippocratic Oath, and beyond that, of the criminal law. Medical experiments cannot be casually reproduced. Repetition of an experiment is usually only justified when a patient is so sick as to justify emergency measures. In the nature of things, a medical researcher publishing in a medical journal is more like an expert witness in court than like a true academic researcher. The researcher is in effect saying "trust me." Publishing false reports of medical research in a medical journal does come very close to Justice Oliver Wendell Holmes' definition of "shouting 'fire' in a crowded theater."
Further, one has to deal with the fact that large numbers of medical researchers are effectively employed by pharmaceutical companies. There have been scandals of all kinds, involving the practices which inevitably emerge when the researchers' vested interest is to find that the latest new drug is absolutely wonderful. I would concede that the leading medical journals' attempts to enforce disclosure are not the best way to go about cleaning out the Augean Stable. In the long run, a doctor who is economically situated as a businessman will behave as a businessman. If you want him to behave like a public servant, you have to pay him as a public servant. A starting point would be to convert the medical schools from the present system of tuition-plus-loans to a system of all-expenses-paid-plus-service-obligation, similar to that employed by the military service academies.
Note that all of this is specific to medical research. Martin Fleischmann and Stanley Pons made public fools of themselves with their claim of "Cold Fusion," back in 1989, but they did no very great harm. Other laboratories expeditiously built similar apparatus, giving their students a chance to learn how to build apparatus, and discovered that Cold Fusion did not replicate. No one died as a result.
Parenthetically, Fleischmann and Pons' apparatus looks a lot like a kind of electric battery, and the legitimate researchers in their field were therefore battery scientists. Battery scientists of the time-- Stanford Ovshinsky comes to mind-- tended to work with large consumer industries, that is, industries which hoped to buy a better battery to put in their products.
I would add that the considerations which apply to publication of medical research results do not apply to the discussion of things like patent policy, which is essentially based on public information. Anyone who has a mind can go to the Patent Office's website, and read the patents in an area he knows something about, and judge for himself how flimsy they are. Of course, what inevitably comes up in discussions of patents is that the Patent Office has a strong family resemblance to those mail-order colleges which issue degrees all the way up to the Ph.D. from disused motels in western Wyoming. The typical patent examiner is minimally qualified, and, generally, extremely young, with no memory of how things were done twenty years ago. He is expected to examine a patent application in ten hours or so, including office overhead and bureaucratic waste of time-- not enough time to do any kind of background reading in the subject matter. Yet, he is given, by law, the authority which naturally adheres to a full professor at Harvard Medical School, or MIT, or some comparable institution. Under the circumstances, it is hardly surprising that a patent is often little more than a rubber stamp applied to the unsubstantiated and self-interested claims of the patent applicant.