Declaring Conflict of Interest - Current State of Affairs in the Ophthalmic Literature.

TLDR: Consistency and completeness of declaration of COI in the ophthalmic literature is analyzed and journals and editors may need to take a more active role in ensuring accurate and consistent COI reporting.

Abstract: The importance of transparency with financial ties in biomedical research is widely recognized, and most peer-reviewed journals require declarations of Conflicts of Interest (COI). Nonetheless, variability in the consistency of declarations of COI has been sparsely assessed. To assess consistency and rates of COI declarations in the ophthalmic literature and the effectiveness of journal COI policies. We analyzed consistency and completeness of declaration of COI in the ophthalmic literature and compared the levels of completeness to specific journal requirements. Six-hundred forty-two peer reviewed articles satisfied the inclusion criteria. In 64%, COIs were unreported, in 25% declaration of COI was incomplete, and 11% of the articles reviewed had complete declaration of COI. Of the 33 journals in which the most frequently published authors' articles appeared, 10 required the authors to complete the International Committee of Medical Journal Editors (ICMJE) form or an equivalent form, but this did not affect the rates of COI declaration. In a random sampling of the most frequently published authors in the field of ophthalmology, declaration of COI was low and highly inconsistent. Requiring a standardized COI form has no significant effect on the rate of accurate COI reporting. Our findings lend support to the growing body of literature that shows that journals and editors may need to take a more active role in ensuring accurate and consistent COI reporting.

Full text

Made available by Hasselt University Library in https://documentserver.uhasselt.be
Declaring Conflict of Interest - Current State of Affairs in the Ophthalmic Literature
Peer-reviewed author version
Schaefer, Jamie Lea; AUBERT BONN, Noemie & Craenen, Geert (2017) Declaring
Conflict of Interest - Current State of Affairs in the Ophthalmic Literature. In:
ACCOUNTABILITY IN RESEARCH-POLICIES AND QUALITY ASSURANCE, 24(7),
p. 375-383.
DOI: 10.1080/08989621.2017.1357474
Handle: http://hdl.handle.net/1942/27528

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gacr20
Download by: [Australian Catholic University] Date: 05 August 2017, At: 04:49
Accountability in Research
Policies and Quality Assurance
ISSN: 0898-9621 (Print) 1545-5815 (Online) Journal homepage: http://www.tandfonline.com/loi/gacr20
Declaring conflict of interest – current state of
affairs in the ophthalmic literature
Jamie Lea Schaefer , Noemie Aubert Bonn & Geert Craenen
To cite this article: Jamie Lea Schaefer , Noemie Aubert Bonn & Geert Craenen (2017): Declaring
conflict of interest – current state of affairs in the ophthalmic literature, Accountability in Research,
DOI: 10.1080/08989621.2017.1357474
To link to this article: http://dx.doi.org/10.1080/08989621.2017.1357474
Accepted author version posted online: 26
Jul 2017.
Submit your article to this journal
Article views: 1
View related articles
View Crossmark data

Accepted Manuscript
Declaring Conflict of Interest – Current State of Affairs in the
Ophthalmic Literature.
Jamie Lea Schaefer MD
1
, Noemie Aubert Bonn MSc
2
, Geert Craenen MD MBE
1,3,4
University at Buffalo Ross Eye Institute
1
; Hasselt University, Faculty of Medicine and Life
Sciences
2
; University at Buffalo Center for Clinical Ethics, Western New York VAMC
4
.
CONTACT
Geert Craenen MD MBE
Dept. of Ophthalmology; Room 627 D
WNYVAMC
3495 Bailey Avenue
Buffalo NY 14215
Email: Geert.Craenen@VA.Gov
Abstract:
Context: The importance of transparency with financial ties in biomedical research is widely
recognized, and most peer-reviewed journals require declarations of Conflicts of Interest (COI).
Nonetheless, variability in the consistency of declarations of COI has been sparsely assessed.
Objectives: To assess consistency and rates of COI declarations in the ophthalmic literature and
the effectiveness of journal COI policies.
Methods: We analyzed consistency and completeness of declaration of COI in the ophthalmic
literature and compared the levels of completeness to specific journal requirements.
Results: 642 peer reviewed articles satisfied the inclusion criteria. In 64%, COIs were
unreported, in 25% declaration of COI was incomplete, and 11% of the articles reviewed had
complete declaration of COI. Of the 33 journals in which the most frequently published authors’
Downloaded by [Australian Catholic University] at 04:49 05 August 2017

Accepted Manuscript
articles appeared, 10 required the authors to complete the International Committee of Medical
Journal Editors (ICMJE) form or an equivalent form but this did not affect the rates of COI
declaration.
Conclusions: In a random sampling of the most frequently published authors in the field of
ophthalmology, declaration of COI was low and highly inconsistent. Requiring a standardized
COI form has no significant effect on the rate of accurate COI reporting. Our findings lend
support to the growing body of literature that shows that journals and editors may need to take a
more active role in ensuring accurate and consistent COI reporting.
Introduction:
The November and December 2014 issues of Ophthalmology and the American Journal of
Ophthalmology feature a joint editorial addressing ongoing issues with declaration of authors’
conflicts of interest and recent changes in the recommendations of the International Committee
of Medical Journal Editors (ICMJE) on financial disclosure
1,2
. In the editorial Liesegang and
Bartley, the respective editors, commit to maintaining physician and public trust through
enhanced transparency. Amongst other things, both editors argue that disclosure of interests
should include all ties to industry encompassing the 36 months preceding submission, even if not
specific to the submitted work. The two editors also submit that terms such as “potential conflict
of interest” and “relevant financial information” lead to confusion and form an impediment to
transparency.
Liesegang and Bartley advocate for universal adoption of the ICMJE form to effect this
transparency. To the ICMJE, improved transparency takes the form of requiring all authors to
Downloaded by [Australian Catholic University] at 04:49 05 August 2017

Accepted Manuscript
disclose all financial ties and interests, and letting the readers decide on their significance for
themselves. “Confusion and subjectivity in reporting financial conflict” the editors argue, now
including continuing medical education presentations in their scope of comments, “would be
lessened simply by disclosing everything and having the same disclosures at each presentation
and each manuscript submission.”
1
We applaud these editors of two of the leading journals in ophthalmology for taking this
strong public stance on an issue vital to the future of scientific reporting, and this leads us to
wonder: How well does the literature in our own field of ophthalmology live up to these high
standards?
The same December issue of Ophthalmology, for example, contains back-to-back articles in
which the same author, writing on the same subspecialty topic, using the same methods and
materials, reports very different financial interests. Illustrative of the findings reported in our
paper, such discrepancies demonstrate the need for a better understanding of the declaration of
conflict of interests (DOI) in practice.
DOI presumes an understanding of which interests may be considered conflicting, biased, or
requiring special mention. Agreement on what should be declared, even amongst the publishing
arm of the research enterprise, is far from universal, and DOI requirements differ dramatically
from journal to journal.
3,4
In the past few years, increasing efforts have been made to arrive at a
consistent, common DOI format. The International committee for Medical Journal Editors
(ICMJE), for example, probably provides the most highly endorsed DOI form – the ICMJE Form
for Disclosure of Potential Conflicts of Interests – intended to be completed by all authors at the
time of submission and to be confirmed before publication. This form ensures that each author
declares any payment or services received for any aspect of the work submitted, but also any
Downloaded by [Australian Catholic University] at 04:49 05 August 2017

Frequently asked questions

Q1. What are the contributions mentioned in the paper "Declaring conflict of interest – current state of affairs in the ophthalmic literature" ?

The authors analyzed consistency and completeness of declaration of COI in the ophthalmic literature and compared the levels of completeness to specific journal requirements. In 64 %, COIs were unreported, in 25 % declaration of COI was incomplete, and 11 % of the articles reviewed had complete declaration of COI. Of the 33 journals in which the most frequently published authors ’ 

Q2. What is the effect of a conflict of interest on the patient?

In medical sciences, conflicts of interest have the potential to undermine their genuineunderstanding of medical mechanisms, impede research, and ultimately harm patients. 

Q3. What is the way to improve the self-regulation of medical literature?

As Liesegang and Bartley suggest, improved transparency through declarations of interests by every author, whether they are deemed relevant or not, is one step towards improved self-regulation. 

Q4. What type of DOI form was used for the analysis?

In the 524 articles included in the journal analysis, approximately a third (N=186) were published in journals either requiring or endorsing the ICMJE form for DOI or an equivalent slightly modified DOI form. 

Q5. What is the main reason why the medical literature is disturbed?

The medical literature continues to be disturbed by withdrawals of articles following revelations of financial impropriety, ghostwriting, and even data fabrication. 

Q6. What are the two editors’ views on financial disclosure?

The two editors also submit that terms such as “potential conflict of interest” and “relevant financial information” lead to confusion and form an impediment to transparency. 

Q7. How many articles were not declared in the ICMJE?

Of the remaining 338 papers published in journals in which a formal DOI form was not mandatory, the declaration of known conflicts of interest were complete, incomplete, and unreported in 12%, 27%, and 61%. 

Q8. What does the editorial say about disclosure of financial conflicts?

“Confusion and subjectivity in reporting financial conflict” the editors argue, now including continuing medical education presentations in their scope of comments, “would be lessened simply by disclosing everything and having the same disclosures at each presentation and each manuscript submission. 

Q9. What is the view of the authors?

Their view is that, as much as authors have a duty to declare all COI and comply with journals' policies, journals have a responsibility to endorse complete and coherent COI guidelines and to enforce them. 

Q10. How many peer reviewed articles did the authors find?

A random sampling of the most productive authors in ophthalmology yielded 642 peerreviewed articles which satisfied their inclusion criteria. 

Q11. Why is transparency put forward as the summum bonum?

in order to preserve the authenticity of the knowledge captured by researchers, transparency, rather than independence, is put forward as the summum bonum. 

Q12. What was the number of journals listed as 'following ICMJE recommendations'?

Of the 33 journals investigated, only seven were listed on the ICMJE website as 'journalsfollowing ICMJE recommendations' (two of which were members through the association of journals they are part of rather than at the individual journal level). 

Q13. What is the highly endorsed form for disclosure of potential conflicts of interests?

The International committee for Medical Journal Editors (ICMJE), for example, probably provides the most highly endorsed DOI form – the ICMJE Form for Disclosure of Potential Conflicts of Interests – intended to be completed by all authors at the time of submission and to be confirmed before publication.