‘Unfair and unsubstantiated’: Journal retracts paper suggesting smoking is linked to lower COVID-19 risk

ecigarettereviewed.com via Wikimedia

A paper suggesting that smokers were significantly less likely than nonsmokers to contract Covid-19 has been retracted because the authors failed to disclose financial ties to … the tobacco industry. 

The article, which appeared as a preprint and then as an “early view” in the European Respiratory Journal last July, came from a group at the University of Piraeus, in Greece, and the University of Utah. The first author was Theodoros Giannouchos, currently a post-doc at the University of Utah, and the senior author was Konstantinos Farsalinos, a fairly prominent name in the world of vaping research. 

Some vaping advocates have pointed to a protective effect of nicotine against Covid-19. According to the preprint of the now-retracted paper:

Current smokers were 23% less likely to be diagnosed with COVID-19 compared to non-smokers. Of all COVID-19 patients, 34.8% were hospitalized and 13.0% experienced an adverse outcome. Male gender, older age, having one or more comorbidities, and chronic renal disease, diabetes, obesity, COPD, immunosuppression and hypertension were associated with hospitalization and adverse outcome. Current smoking was not associated with adverse outcome.

In the Competing Interests Statement section of that document, the authors attested that:  

The authors have declared no competing interest.

That assertion turns out to have been false. 

Per the retraction notice

The above mentioned article has been retracted at the request of the editors and the publisher. The article was originally published “early view” in the European Respiratory Journal on 30 July 2020. Subsequent to this, and prior to publication of the version of record in an issue of the European Respiratory Journal, it was brought to the editors’ attention that two of the authors had failed to disclose potential conflicts of interest at the time of the manuscript’s submission: That is, one of the authors (José M. Mier) at the time had a current and ongoing role in providing consultancy to the tobacco industry on tobacco harm reduction; and another (Konstantinos Poulas) at the time was a principal investigator for the Greek NGO NOSMOKE, which has its base at Patras Science Park, a science and innovation hub that has received funding from the Foundation for a Smoke Free World (an organisation funded by the tobacco industry). After careful review of the manuscript content alongside the new disclosures brought to light, and in consultation with the leadership of the European Respiratory Society (the publisher of the journal), the editors and society were in firm agreement that, if these conflicts of interest had been disclosed at the point of manuscript submission, the editors would not have considered the article for publication, for the following main reasons. 1) The manuscript presents some new data on, and provides a section of discussion of, the effect of tobacco consumption on patient susceptibility to COVID-19, and cites other studies that claim SARS-CoV-2 infection is less prevalent in smokers or tobacco users. 2) The European Respiratory Society, as a leading medical organisation in the respiratory field whose mission is to promote lung health and alleviate suffering from respiratory disease, has bylaws in place that do not permit individuals with ongoing relationships with the tobacco industry to participate in its activities: As well as society membership, and participation in congresses and meetings, publication in the society’s journals is also considered an activity covered by these rules. 3) Because of these bylaws, at the point of manuscript submission, the submitting author is specifically asked to assert that no potential conflict of interest involving the tobacco industry exists regarding the submitted manuscript. In the case of this article, the submitting author asserted that no such conflict of interest existed, and for this reason the article duly entered the journal’s peer review process.

The editors note that the Committee on Publication Ethics does not uniformly recommend retraction for failure to disclose conflicts of interest — although it doesn’t smile upon such lapses, either:

However, on balance the editors felt the decision was justified based on the nature of the undisclosed relationship, in the context of the sensitive subject matter presented, and on the need to align the published journal content with the bylaws of the publishing society. The editors also acknowledge that at no point was there a question of any scientific misconduct on the part of any of the authors, aside from the failure of two contributing authors to disclose their conflicts of interest relating to the tobacco industry. As these issues were brought to the editors’ attention before the final version of record for the article was published, this notice of retraction (published 4 March 2021) replaces the early view version of the published article, which has now been removed. The editors thank the authors for their cooperation in the matter. The authors did not agree with the decision to retract the article.

Farsalinos tells Retraction Watch:

Despite being the corresponding author, I was contacted by the editors of the journal only after the decision for retraction was made. They mentioned that the decision was made because two authors had conflicts of interests that were not declared. Each author was responsible for declaring his own conflicts of interest; therefore I was not responsible for these declarations. I responded that the discussed conflicts were irrelevant to the study’s main aims and objectives. Additionally, I  proposed to publicly release the full dataset and the statistical script so that all findings could be independently verified. The editors declined. I requested my proposal to be mentioned in the retraction letter, but that was also rejected by the editors. I disagree with the retraction and I consider it unfair and unsubstantiated. I fully support the accuracy of all analyses and content, and we will submit the same manuscript to another journal. 

Like Retraction Watch? You can make a one-time tax-deductible contribution or a monthly tax-deductible donation to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].

5 thoughts on “‘Unfair and unsubstantiated’: Journal retracts paper suggesting smoking is linked to lower COVID-19 risk”

  1. “The European Respiratory Society… has bylaws in place that do not permit individuals with ongoing relationships with the tobacco industry to participate in its activities: As well as society membership, and participation in congresses and meetings, publication in the society’s journals is also considered an activity covered by these rules.”

    In contrast, the Annals of Internal Medicine allows the American Diabetes Association’s Chief Scientific, Medical, and Mission Officer to be the senior author on a paper which, without any new data, encourages stricter glycemic control than do the American College of Physician guidelines. Stricter control directly implies greater sales of diabetes drugs. What is more, the Officer lists no conflict of interest, although ADA receives millions of dollars from the companies that are selling these expensive drugs.
    Ref: O’Sullivan C, Tejani AM, Finucane T. Guidelines Versus Guidelines. Ann Intern Med. 2018 Dec 18;169(12):895-896. doi: 10.7326/L18-0559. PMID: 30557432.

    Hats off to the European Respiratory Society and its journal.

    1. If all societies had similar guidelines, almost nothing would ever get published! Not that I am an apologist for tobacco companies, but medical societies dedicated to any particular condition almost always get funding, directly or indirectly, and contributions through drug companies that produce treatments for the condition in question. Diabetes is just one example, what about heart disease or obesity or rheumatism?

      1. Given what we’ve seen of the harm obesity research has done, including increasing obesity by promoting dieting, it seems like it would be an excellent step.

  2. About one year ago I was writing (in Greek, translated now from my archive):

    ===

    Let’s see a Greek e-cigarette product:

    https://www.nobacco.gr/en/premium-vg-sacred-armorb

    There we read (bottom left):

    “The Aquarius Project was established through the pioneer alliance of NOBACCO and LMBI (Laboratory of Molecular Biology and Immunology), Pharmaceutical Dept., Patras University.”

    Let us now turn to the scientific publication referred to in the first article, which is as follows:

    https://link.springer.com/arti…/10.1007%2Fs11739-018-02023-x

    In the scientific article we read in the addresses of the authors:
    Department of Pharmacy, University of Patras, Rio, 26500, Greece
    (Department of Pharmacy is the same as the Pharmaceutical Dept)
    Is there a Conflict of Interest (COI)? COI, of course, is often found in scientific publications, and it can be, as long as it is clearly stated. That is why scientific journals always have a place where issues of moral order must be stated.

    So I read in the scientific article:

    Ethics declarations
    Conflict of interest
    Authors report no conflicts of interest for the past 36 months.

    So I leave open the possibility that the University of Patras and the Department of Pharmacy will not be involved from January 2017 onwards with the AQUARIUS PROJECT / Nobacco (previously involved, see eg https://www.vice.com/…/ ae… / hlektroniko-tsigaro-erevna-ellada).

    So we have two scenarios:

    1. The University of Patras through its faculty members, is not involved from January 2017 onwards with AQUARIUS PROJECT / Nobacco, and then it should simply require Nobacco to withdraw or clarify its product relationship statement with the Department Pharmaceutical.

    2. The University of Patras, through its faculty members, continues to have an interest in the AQUARIUS PROJECT / Nobacco, and then simply has to ask the faculty members involved to correct their publication and report a conflict of interest (and possibly correct other publications where they have not declared COI).

    ==

    Poulas and co-workers have been publishing on the use of vaping products promoting them as healthy alternatives compared to smoking, they have been involving the name of the University of Patras to quality checking and advertising vaping products, and have been advocating the use of these products in public in several occassions.

    The University and colleagues have failed to take action for years.

    Hat tip to the European Respiratory Society and its journal indeed.

    There should be additional investigations and retractions.

  3. So, if the paper had included all the right declarations is would still be there and deemed to be “right”, and therefore citeable.
    But because two declarations were omitted, it is now “wrong”?
    Are papers not judged on their technical merit any more?

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.