by Lisa Kearns, MS, MA, and Arthur Caplan, PhD
A few months ago we called for a new conflict of interest (COI) disclosure policy. Recent events at Memorial Sloan Kettering Cancer Center (MSK) underscore the urgency in addressing COI. We encourage researchers to create an “ELF,” an electronic long-form disclosure statement that lists financial relationships, as well as any political, advocacy, or religious points of view, pertinent investments, and the like — any information that could help readers assess potential author bias. ELFs would be maintained by faculty and posted online, as part of their professional bios. The ELF could be linked electronically to all journal articles, included as a URL with author bios at the end of print versions or slide presentations, submitted when trials are registered on ClinicalTrials.gov or other databases, and required as part of NIH grant submissions and by agencies forming advisory panels. The ELF is intended to allow researchers to take ownership of their disclosure statements: Rather than merely completing journals’ variable COI disclosure forms when submitting articles, authors would instead proactively make available a comprehensive list and explanation of their financial and other relationships to anyone who cares to see them.
On September 8, a New York Times/ProPublica investigation revealed that Dr. José Baselga, the chief medical officer at MSK and one of the world’s leading oncologists, had, among other infractions, failed to disclose financial relationships in dozens of journal articles. Baselga called the omissions “unintentional.” This is precisely the kind of situation in which an ELF would come into play. If Baselga had created and kept up-to-date a comprehensive COI disclosure statement, it would have listed all of his relationships, including the ones that he failed to disclose to journals. It would have been accessible on his MSK staff page, and links to it would have accompanied all his articles, grants and presentations.
Why would Baselga have been more diligent with an ELF than he was with journal submissions? That’s where the enforcement of the ELF comes into play. In the same Times/ProPublica article that broke the story, an MSK spokeswoman told the investigators that the institution “has a rigorous and comprehensive compliance program in place to promote honesty and objectivity in scientific research” and that the doctor had “properly informed the hospital of his outside industry work.” It was up to him, she said, to relay the information to journals.
If this is accurate — if all of Baselga’s industry ties were indeed on record somewhere at MSK — and he had completed a COI form that included all that information, then a simple comparison of the two lists would have flagged Baselga’s omissions on his disclosure form. Under the policy we’re proposing, someone at MSK would be responsible for reviewing all ELFs for accuracy and completeness. That monitor would have alerted Baselga about the “inconsistencies” (his word), and he could have amended his ELF; readers of those dozens of articles could have followed a link to his full disclosure statement and seen an accurate reflection of his financial relationships.
We acknowledge that reviewing ELFs is challenging. MSK lists more than a thousand attending staff, plus hundreds more research fellows, scholars, and associates, in its annual report. Reviewing disclosure statements is, and will be, post-ELF, a big chore. But MSK has a big reputation to protect. It is in the interest of an institution of its stature to invest in dedicated staff to review physicians’ and researchers’ COI disclosure statements. What’s the point of having that “rigorous and comprehensive compliance program” in place if you’re not going to ensure that it’s adhered to?
Two other points about the ELF: First, it relies on “good actors” — those who believe in research integrity and who value transparency and are interested in reducing COI. Peer-reviewing colleagues’ ELFs, or having deans and chairs sign off on them, can catch errors and omissions, but it cannot prevent someone from deliberately hiding a relationship. The policy would have worked for Baselga only because, according to the MSK spokeswoman, he had already been forthcoming in his financial relationships with the institution.
Second, the ELF is not yet standardized. MSK’s chief executive emphasized in the aftermath of the Baselga article the need for more “standardized” reporting processes. If by “reporting processes” he meant something like, “All authors and researchers should maintain their own comprehensive, transparent disclosure forms,” then we agree. It’s important to bear in mind that no two ELFs need be alike, and transparency should not rely on someone crafting the perfect disclosure template to elicit all relevant information. A minimal universal template is fine but explanations of COI will obviously vary. Even the most detailed disclosure template will fail to capture — that is, ask about — every potential conflict all of the time.
In the wake of the MSK fiasco, JAMA devoted its “New online” edition for October 26 to COI and disclosure (and included a letter from Baselga with an amended disclosure). In an editorial, Bauchner, Fontanarosa, and Flanagin point out a requirement of the JAMA and JAMA Networks’ author instructions that submitters could satisfy with a link to an ELF: “Authors are expected to provide detailed information about all relevant financial interests, activities, relationships, and affiliations … including, but not limited to, employment, affiliation, funding and grants received or pending, consultancies, honoraria or payment, speakers’ bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.”
Would an ELF really have kept Baselga on the straight and narrow and off the front page of the Times? Not likely, given that, according to the investigation, he also put a positive gloss on disappointing results from clinical trials sponsored by Roche without disclosing his financial relationship with the company and flouted the disclosure rules of the American Association for Cancer Research when he was running the group. We don’t believe that improved disclosure policies will solve the management of COI in publishing and research. But the ELF is not meant to solve COI, just to advance disclosure into the 21st century. For those sincerely interested in research integrity and managing COI, a proactive, author-driven disclosure policy can be a helpful tool.