Ethics Journal Fixing Disclosure Process to Catch More Conflicts

In one issue of the American Journal of Bioethics, at least six authors had conflicts of interest that went undisclosed.

The journal says that five others never responded to requests for their potential conflicts of interest for articles in that same November 2010 issue.

How did that happen?

The American Journal of Bioethics’ co-editor-in-chief, David Magnus from Stanford University, told me in an email that the omissions from the November 2010 edition happened three different ways. He said the journal knows it missed some conflicts, but that it is working on a way to prevent that in the future. He explained:

The corrections fall into three categories. In one case, someone had signed our form indicating no conflict of interest, but had also apparently turned in a separate sheet with relevant info that was inadvertently ignored.

In addition, though our instructions to authors on the COI forms ask for the corresponding authors to obtain all disclosures from all authors, it is possible that a lead author failed to do that and merely disclosed their own. We have made some changes to our procedures to ensure that our instructions are actually followed.

Third, some authors filled out forms stating that they had no COI, when on reflection, and given an opportunity to correct the record, they provided information that our COI committee determined constituted a COI.

All information collected was turned over to our COI committee who determined what should be disclosed and a correction was issued.

Magnus only took over as co-editor-in-chief earlier this year, so he can’t be held entirely responsible for problems nearly two years ago. I asked Magnus how any of these articles could have been published if the conflict information was missing. I also asked if other journal articles where conflicts were missed were going to be corrected. (There seem to be at least a few of those, given that some of the same authors mentioned in the correction have published elsewhere in the journal with no mention of these conflicts.) I did not receive a reply.

The conflicts are not minor. As I wrote on Monday, one author, Myra Christopher, disclosed a laundry list of conflicts at her employer, the Center for Practical Bioethics, which was the former editorial home for the journal. That disclosure was supposed to appear with an article about pain medicine prescription contracts.

Here is what the journal had to say in its August edition about three other authors on that same paper:

Lennie Duensing, Executive Director of the American Academic of Pain Management, disclosed no personal conflicts but reported that the American Academy of Pain Management received educational grants from Genentech, Allergan, Endo, and Purdue Pharma. In addition, support from AAPM’s annual meeting came from Pfizer, Endo, and NeurogesX. Duensing noted that all educational grants were unrestricted and for CME-accredited programs only.

Ben Rich disclosed that he participated in the American Pain Society and American Academy of Pain Medicine clinical practice guideline panel for chronic noncancer pain and that he conducted research and writing leading to publication in a peer reviewed journal of an article on ethical and legal issues raised by tamper-resistant opioid analgesic formulations for K.O.L. LLC.

Aaron Gilson stated that he personally “served on Advisory Committee for Cephalon, participated in a train-the-trainer session for Covidian, and served on REMSAdvisory Committees for Covidian and Meda Pharmaceuticals, and received a research grant from King Pharmaceuticals for secondary data analyses related to PMPs.” Mr. Gilson also disclosed that his institution, The University of Wisconsin Carbone Cancer Center, received unrestricted educational grants from Purdue Pharma to support the work of the Pain & Policy Studies Group. He also indicated that “all of these relationships have since ended.”

You might recognize that last name. John Fauber at the Milwaukee Journal Sentinel has written about Gilson in the past when similar conflicts of interest went unmentioned in journal articles:

Fueled by a continuous infusion of money from the manufacturers of drugs such as OxyContin over more than a decade, the UW research group has been a quiet force in the effort to liberalize the way those drugs are prescribed and viewed in the United States.

In addition, records show cozy personal financial relationships between drug makers and two officials with the UW Pain Group, Aaron Gilson and David Joranson. Those include helping a drug company win Food and Drug Administration approval for a new narcotic painkiller and working as speakers or consultants.

I asked all three authors what happened to lead to this correction and whether they had filled out disclosure forms before the November 2010 publication. Only Gilson wrote me back:

I was not involved in the manuscript submission process for the referenced article, so I am at a loss to explain what happened.  I believe that Myra Christopher would be able to provide the information that you request.

This seems to fall in line with how Magnus said the disclosure process was supposed to work. Otherwise it would not have been able to publish the actual conflicts in the correction in the August issue. But when I asked Christopher whether she reported conflicts of interest to the journal for everyone on the paper or just for her, she wrote:

I did coordinate the writing of this publication but not the submission of disclosure statements.  I submitted my disclosure.

So it seems Christopher did not follow the procedures that Magnus mentioned above. This is a bit surprising, given that Christopher is a member of the journal’s editorial board. I also asked her the same question I asked Magnus:

If, in fact, some authors did not provide disclosure information prior to the November 2010 publication, why was the piece published?

Christopher did not respond to that question.

So far one third of this paper’s authors had undisclosed conflicts that the journal decided later were worth publishing. What about the other eight authors? The American Journal of Bioethics says that three others – Evan Anderson, Lisa Robin and Steve Passik – “could not be reached or did not respond.”

I asked all three about what happened. Only Passik replied. He wrote:

Thank you for your inquiry. I was part of a group that met to discuss ethical issues involved in opioid agreements which led to the paper for the AJOB back in 2010. In answer to your questions: I am not certain why this correction was issued now. While I don’t have a specific recollection of all of the prepublication paperwork, I most certainly did complete whatever was asked for at the time. I routinely file conflict of interest statements here at Vanderbilt and have never hesitated to do so here or back when I was at Sloan Kettering or for papers, presentations at meetings etc. With regard to not responding when the correction was being put together, I don’t recall ever receiving an inquiry to do so.

So why couldn’t the ethics journal reach Passik? He did change jobs to take a new position at Vanderbilt University, but that was more than a year ago.

If the ethics journal couldn’t find him at Vanderbilt, they may want to hire some new interns. I found him in 30 seconds, and he replied just a few hours after I wrote him.

And he’s not sheepish about disclosing conflicts. Pain Management Today recently wrote, for example:

Dr Passik has disclosed that he has received research/grant support from Cephalon and Ligand Pharmaceuticals and is a member of the speaker's bureau and a consultant for Cephalon, Ligand Pharmaceuticals, Eli Lilly and Company, Pricara, and King Pharmaceuticals, Inc.

So to summarize thus far. Only one author has a clear recollection of how the conflict of interest process worked for this journal article. Myra Christopher, a member of the ethics journal’s editorial board, says she submitted paperwork for herself. It turns out, that she was supposed to collect and turn in all the authors’ conflict of interest information, according to the journal’s co-editor-in-chief. One of her coauthors, Gilson, says that Christopher handled the entire submission process for everyone on the article, which comports with what the journal now says should have happened. Yet Gilson’s disclosure information somehow made its way to the journal and was initially ignored. And another co-author, Passik, says he can’t remember but says he would have filled out any disclosure information that was asked of him. He could not be reached by the journal, despite the fact that he is quite easy to find.

Believe it or not, there are still a few authors with disclosure issues that I have not covered yet. I’ll write about them next week.

Image by Dan Allison via Flickr