Retraction Watch

Tracking retractions as a window into the scientific process

“Innocent mistake” leads to bioethics article retraction

with 5 comments

jbiA July article that incorrectly called out nine leading bioethics journals for their lack of availability to researchers in low- and middle-income countries is being pulled after editors of the indicted journals refuted the allegations.

The last author on the article, published in the Journal of Bioethical Inquiry, told us an “innocent mistake” and difficulty navigating a website led the authors to incorrectly note that nine journals had not made their contents available through the World Health Organization’s Health InterNetwork Research Initiative database (HINARI), which gives bioethicists who live in low- and middle-income countries access to research articles either free of charge or at reduced cost. The authors argued that the mistake didn’t affect the paper’s conclusions, but the journal disagreed, and opted to pull the paper entirely.

After searching through the database, first author Subrata Chattopadhyay mistakenly determined that the journals had not made their contents available through HINARI, when in fact they were listed but on a different part of the website.

Even with the error, the authors maintain that their conclusions remain sound and that the field is shaped by a “hegemony of Western bioethics.”

However, once it was revealed that the nine journals in question do participate in HINARI — and two of the criticized publishers actually helped found the database — the journal called that conclusion into question.

Here’s the retraction notice for “Imperialism in bioethics: How policies of profit negate engagement of developing world bioethicists and undermine global bioethics:”

It has been decided by the journal editors and Springer International Publishing AG to retract the above article. The retraction has followed a complaint that identified errors in statements and data presented in the article that were essential to its main conclusions. The errors, which appear to have been made honestly, occurred in the abstract, the body of the text, and Table 1:

-It was claimed that nine leading bioethics journals (The American Journal of Bioethics; Bioethics; Developing World Bioethics; The Journal of Law, Medicine and Ethics; the Journal of Bioethical Inquiry; Medicine, Health Care and Philosophy; Cambridge Quarterly of Healthcare Ethics; The Journal of Medicine and Philosophy; and the Journal of Medical Ethics) do not participate in the Health InterNetwork Access to Research Initiative (HINARI) of the World Health Organization, a program that provides free or very low-cost online access to major journals by researchers in developing countries. In fact, these nine journals do participate in HINARI.

-It was claimed that full-text articles of fourteen bioethics journals are not available through the free digital archive of PubMed Central. In fact, selected articles and/or author manuscripts in these journals are available, which includes the nine journals listed above, as well as The Hastings Center Report; the International Journal of Feminist Approaches to Bioethics; The Journal of Clinical Ethics; the Kennedy Institute of Ethics Journal; and Theoretical Medicine and Bioethics. Additionally, The Journal of Medicine and Philosophy participates as an “NIH Portfolio” journal within PubMed Central, and the Journal of Medical Ethics deposits articles published via its “BMJ Open Access”

-It was recommended that “international bioethics journals published by commercial publishing houses like Springer and Wiley-Blackwell should join WHO’s HINARI and make their published contents accessible to bioethics workers in LMICs.” In fact, two out of the three Springer journals and three out of the four Wiley/Wiley-Blackwell journals listed in the article do participate in HINARI, and Springer and Wiley are two of the initial founding publishers of HINARI.

As a result of these errors, key conclusions of the article relating to the scope of access to bioethics journals and the policies and practices of publishers, journals, and scholars are no longer sustainable, necessitating its full retraction. In accordance with usual practice, the text will remain available as electronic supplementary material marked “retracted article.” The journal and the publisher apologize to readers and other affected parties for any inconvenience or embarrassment caused by the publication of this article.

The paper hasn’t been cited, according to Thomson Scientific’s Web of Knowledge.

Here’s how the mistake happened, according to Chattopadhyay, who is based at The West Bengal University of Health Sciences in India:

The inadvertent error happened because 1) my/our search inside HINARI of “accessible content” in subject heading “ethics” came out with a list of only six journals –  Bangladesh Journal of Bioethics, BMC Medical Ethics, Journal of Education and Ethics in Dentistry, Medicolegal and Bioethics, Ramon Llull Journal of Applied Ethics and South African Journal of Bioethics and Law – and none of the leading bioethics journals mentioned in our work, and 2) I/we could not get access to full-text articles of the journals mentioned in our list through the HINARI, presumably because many developing countries, like India and China, do not qualify for inclusion in the HINARI. In addition, even when some countries qualify for inclusion in the HINARI, this access is limited to bioethics workers working in institutions having access, thus excluding bioethics workers not affiliated with HINARI institutions.

Indeed, other searches – such as by the alphabetical list of journal titles – does turn up the name of the journals in question. We asked Chattopadhyay why he didn’t try different types of searches, but haven’t heard back.

Co-authors Raymond De Vries and Catherine Myser told us in a joint statement that once the authors realized the error, they submitted a corrected version of the article but the journal decided to retract the paper anyway.

De Vries added in a separate email to us that even if Chattopadhyay had found the journals, he is based in India, which is not poor enough to qualify for free or low-cost access to the articles themselves, so he wouldn’t have been able to read anything.

An important part of this story is that we were technically wrong, but we were — I want to say — empirically correct. Even though the journals are enrolled in the program, HINARI does not qualify my colleague in India to view them. He doesn’t make enough money there to spend 30 euros on an article even once per month and although HINARI makes things accessible, it doesn’t make them uniformly available. Even in HINARI-eligible countries, not all institutions have HINARI eligibility. So you can look at say, Mozambique, which is HINARI eligible, but your institution in Mozambique may not be, and you won’t have access. So our conclusion does kind of shift from telling those publishers and journals to enroll, to emphasizing that the HINARI program is a great initiative, yet it has significant gaps and leaves a lot of people without access.

In this sense, De Vries argued that the main conclusion remains accurate:

The conclusion of our article — that bioethicists working in low- and middle-income countries have limited access to the literature of bioethics and therefore can’t make a contribution to the global dialogue of bioethics — remains true. It’s just the technical aspects of our article are in fact wrong and we admit they’re wrong.

Leigh E. Rich, editor in chief of JBI, confirmed that the errors in the retracted paper were honest mistakes:

In the peer-reviewed editorial, the authors erroneously made the claim that several bioethics-related journals that they identified do not belong to the World Health Organization’s Health InterNetwork Access to Research Initiative (HINARI). The authors cited the World Health Organization (WHO) in support of this claim, and thus their data appeared to come from an authoritative source. Although this was an honest error, the authors did not offer further explanations about their methods in determining journal participation in HINARI. The peer reviewers had no indication that this statement was an error of fact.

Additionally, the editors said that there was problematic phrasing in a portion of the paper that criticized the availability of certain journals in the PubMed Central archive:

The authors also claimed that full-text articles of several bioethics-related journals that they identified are not available via the free digital archive of PubMed Central. This statement, which was presented in a table, was not well phrased: the authors likely intended to report that not all articles in these journals are available via PubMed Central, which on its face is true, without further explaining how a journal participates in the archive or the archive’s different levels of participation. Because some articles in the journals are available in this archive (e.g., articles already open-access and/or “author manuscripts”), this point has been clarified in the retraction.

De Vries said that the authors are considering writing a follow-up article to document the entire episode and also hope to publish a corrected version of the original paper.

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.

Comments
  • Karen Shashok January 29, 2016 at 11:07 am

    The authors are paying a heavy (and unfair, it seems to me) social penalty for their honest error.

    Perhaps the reviewers should have checked the websites themselves to see if they were able to find the results the authors were unable to find.

    Perhaps there should be a checklist for authors and reviewers to help ensure that search strategies within websites are reported in enough detail to allow reviewers to check whether the authors have overlooked a link to the information they were seeking.

    Perhaps PubMed Central and members of HINARI should work with their website developers to ensure that users can see at a glance which articles are freely available. (Perhaps commercial publishers that participate in systems like HINARI are not especially motivated to make it easy for users to find free or discounted content.)

    Others have said it before: HINARI is great as far as it goes, but it could go further. Perhaps HINARI should take a more detailed approach to deciding who benefits from this system. Perhaps total GNI (World Bank, GNI per capita (World Bank), the United Nations Least Developed Country (LDCs) List and the Human Development Index (HDI) do not accurately capture how much (or how little) funding is really available to researchers for obtaining access to the literature.

    And perhaps research ethics and access to scientific information is an area of knowledge and research that should be made entirely open access. If it’s not available because it’s not affordable, it’s being wasted.

    • Bobo January 29, 2016 at 2:31 pm

      What social penalty? Their paper was retracted because the editors deemed it to be too far gone to be fixed by a correction. The editors state it was an honest error. The authors intend to publish a corrected version.

      That seems like the way that science should work, and I don’t see the authors being unfairly stigmatized.

      • Karen Shashok February 4, 2016 at 4:35 am

        There’s a study titled “Financial costs and personal consequences of research misconduct resulting in retracted publications”, available at http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4132287/ .

        In a case like this one it’s hard for editors to decide what to do. Current guidelines may recommend taking a hard line, even when there is evidence that the authors did not act with any intention of deceiving anyone. It’s the editor’s decision whether an honest error undermines the conclusions to such an extent that retraction rather than correction is the best course of action. Authors and readers may disagree. As long as the original study remains available to readers, others are free to decide for ourselves whether the work is trustworthy or not.

        • Simon May 16, 2016 at 7:48 am

          ‘others are free to decide for ourselves whether the work is trustworthy or not.’ But surely editors are justified in protecting the trustworthiness of their journal as they see fit, rather than just thinking ‘let readers decide.’

  • udo schuklenk January 29, 2016 at 2:59 pm

    As an the Editor of two of the journals slandered (a second time) by this team of authors, let me just say that the article could not have easily been re-written once the false factual claim had been removed. In fact, one of these authors continued to slander me in internet fora after I had explained to them how their error occurred. Nothing innocent there. Selfrightousness is more like it. For that reason my sympathy as far as the social penalty is concerned is somewhat limited. I have not received an apology yet from these authors either, which goes with being ‘right’ in matters ‘imperialism’.

    Their excitement about OA should have been tempered a bit by a somewhat trivial insight, their Indian co-author (possibly all of them) would not have been able to afford the APC for an OA paper either, so their wonderful prose would not have seen the light of the day unless one such journal had chosen to waive that fee. Really the paper amounted to having-your-cake-and-eating-it prose.

    Short of having major journals produced on a volunteer basis (check how most of such projects fare over time!), the only solution really is to either waive APCs for those who cannot afford them or have schemes like HINARI et al to ensure wider access to subscription based journals.

    It is up to the UN organisations involved in generating the HDI and other such measures to ensure they capture what needs to get captured in order clear the way to access to subscription based journals. I doubt that anecdotes about a particular doctor in India who cannot access a particular journal do much to serve any particular cause. – As an aside, for academics purportedly concerned about matters in the global south, this particular hobby horse of these authors seems among the most inconsequential issues to worry about, considering actual problems in the global south. But then, it’s a matter of academic freedom to freely choose the issues one wishes to work on. Good luck to them.

  • Post a comment

    Threaded commenting powered by interconnect/it code.