Undisclosed conflicts of interest usually lead to corrections – but for some journals, that’s not enough

When authors are faced with filling out a journal’s conflict of interest form, deciding what qualifies as a relevant conflict can be tricky. When such omissions come to light, only rarely do they result in retractions – and certainly not author bans. But there are exceptions.

In October, the journal Chest retracted a 2015 review article exploring how mechanical ventilation can be used most effectively to manage acute respiratory distress syndrome (ARDS) after finding that the authors failed “to disclose all relevant conflicts of interest.” What’s more, the journal initially planned to ban the two authors with undisclosed conflicts from submitting papers to the journal for three years, but ultimately decided against it.

The Committee on Publication Ethics says that retractions may be warranted in cases of undisclosed conflicts of interest, but in our experience, most notices that cite that reason mention other problems with the paper, as well. Not this case – here, the only thing that seemed wrong with the paper was the authors’ failure to mention their ties to a ventilator company. The authors requested a correction – the usual fix, one accepted by the other journals they contacted – but to Chest, that wasn’t enough.

Here’s the retraction notice for “Mechanical Ventilation as a Therapeutic Tool to Reduce ARDS Incidence”:

“Mechanical Ventilation as a Therapeutic Tool to Reduce ARDS Incidence”, published in the December 2015 issue of CHEST (2015;148(6):1396-1404), was retracted after the Journal determined that the authors had not conformed to the Journal’s Instructions to Authors to disclose all relevant conflicts of interest by failing to disclose major competing interests that are, in the judgment of the Journal, likely to influence interpretations or recommendations.

The 2015 paper has been cited once, according to Clarivate Analytics’ Web of Science, formerly part of Thomson Reuters.

Indeed, two of the authors—Nader M. Habashi and Gary Nieman—have received honoraria and travel remuneration from Dräger, a German-based company that manufacturers mechanical ventilation devices, and both also hold several patents for ventilation methods.

When we reached out to Habashi and Nieman for comment, Penny Andrews, a critical care nurse who works closely with the authors though was not part of this particular study, provided additional context on their behalf.

Here’s some of the backstory: Based on documents Andrews sent us, on April 20, Habashi and his coauthors received an email from Jean Rice, the manager of journal operations at Chest, regarding an inquiry the journal had received from “an unidentified reader,” Andrews said, who believed the paper contained an undisclosed conflict of interest.

The reader had expressed concern over Habashi and Nieman’s undeclared relationships with Dräger and identified possible bias towards the company within the article, which the reader said included information about features that were proprietary to the Dräger ventilator.

On April 25, first author Nieman responded to Rice and journal editor Richard Irwin via email, addressing each of the reader’s statements, noting they’d be willing to add information about their relationship to the company in the disclosure section of the paper.

On June 20, Irwin sent a letter to Habashi concluding that solution was unacceptable.

To recap: 1) you and Mr. Nieman have spoken at Drager-sponsored talks and did not disclose this. The Subcommittee found the explanation for not disclosing this information unpersuasive. 2) You [spoke] at ICON conferences (Intensive Care On-Lin Network). Because Drager is one of only two commercial partners for ICON, the Subcommittee found that this relationship should have been disclosed. 3) You hold a patent that discusses a Drager product. Again, there is a close enough connection that this should have been disclosed. 4) The article takes for granted settings that are commonly known as features of the Drager ventilator. Although the authors provided an explanation, the Subcommittee still felt found [sic] there was bias in the article.

The letter concludes:

Our Ethics Subcommittee has asked us to 1) to retract the article; 2) to ban you and Gary Nieman from submitting a publishing in CHEST for a period of three years (from July 2016 to June 2019), and 3) to contact the appropriate heads at each author’s institution…

Ultimately, though, after reviewing several appeals, the journal decided not to ban the authors. On July 25, Rice informed Habashi of the journal’s final decision to retract the article, but based on feedback and appeals, had decided not to move forward with the author ban.

Chest’s conflict of interest policy does ask authors to disclose anything that might resemble a conflict:

A conflict of interest is a financial or intellectual relationship or other set of circumstances that might affect, or reasonably be perceived by others to affect, an author’s judgment, conduct, or manuscript. When in doubt, disclose.

We reached out to Irwin to see if he could provide additional insight, but he declined to comment further:

Everything we are willing to say about this retraction is in the retraction notice.

Andrews said that she and the team sent emails to six other journals in which they had recently published, explaining they had recently been made aware of a potential oversight in undisclosed conflicts of interest, and wanted to know if a clarification was warranted.

One journal said there was no need to issue a correction, and five decided to publish a correction – again, the usual course of action. JAMA, for instance, recently published this “Failure to Disclose Conflicts of Interest” letter from Habashi, Andrews, Nieman, and two other authors on September 7, which details potentially relevant conflicts for three JAMA articles published in 2013, 2014, and 2016.

Now, the central worry for Habashi and his coauthors is how a retraction might affect their careers, said Andrews:

You do not want this stain on your record, you don’t want anyone to believe you were intentionally or maliciously trying to hide something.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

One thought on “Undisclosed conflicts of interest usually lead to corrections – but for some journals, that’s not enough”

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.