Recently, through an unrelated chain of events, I came across the author guidelines (PDF) for the international journal of general medicine, The Lancet.
All of you are, of course, aware that The Lancet is a high impact factor journal with significant global readership of both of its off-line and online versions. The journal —no stranger to controversy&mddash; has come under severe criticism from the medical and scientific community, ever since it published the Wakefield article on Autism and the MMR vaccine in 1998. The conclusions and interpretation of this study were retracted by ten of the co-authors in 2004 (Murch et al., Lancet 363 (9411): 750), and the editor, Richard Horton, went on to explain that the retraction was due to “revelations about conflicts of interests” (reported on the Beeb here). Wakefield was found to have been paid a large sum of money by lawyers trying to prove that the MMR vaccine was unsafe, and he did not declare this clear conflict of interest. The rest, as they say, is history. I needn’t elaborate; you are aware of how Wakefield was censured for this non-disclosure, as well as for scientific and ethical misconducts.
Given this turn of events, it is understandable that The Lancet would want to take the issue of disclosure of conflict of interest very seriously. And indeed, in their author instructions, they issue this admonition towards prospective authors:
Conflict of interest
A conflict of interest exists if authors or their institutions have financial or personal relationships with other people or organisations that could inappropriately influence (bias) their actions. Financial relationships are easily identifiable, but conflicts can also occur because of personal relationships, academic competition, or intellectual passion. A conflict can be actual or potential, and full disclosure to The Editor is the safest course.
Here The Lancet appears to recognize that even without financial gain, non-monetary matters can potentially introduce biases to a personal viewpoint that may lead to a restricted, inadequate or selective interpretation of scientific observations. But is it realistically possible to eliminate all the biases, or even all the biases listed?
Failure to disclose conflicts might lead to publication of a statement in our Department of Error1 or even to retraction. All submissions to The Lancet must include disclosure of all relationships that could be viewed as presenting a potential conflict of interest (see Lancet 2001; 358: 854-56 and Lancet 2003; 361: 8-9). The Editor may use such information as a basis for editorial decisions, and will publish such disclosures if they are believed to be important to readers in judging the manuscript.
It doesn’t specify whether only the editors, or even the reviewers, will have access to the said disclosures. The following rules and regulations —all mentioned in the author instructions— seem to be geared towards preventing a Wakefield type of situation from happening again:
- … All authors must disclose any financial and personal relationships with other people or organisations that could inappropriately influence (bias) their work. Examples of financial conflicts include employment, consultancies, stock ownership, honoraria, paid expert testimony, patents or patent applications, and travel grants, all within 3 years of beginning the work submitted
- All authors are required to provide a Conflict of Interest Statement and should complete a standard form.
- For Comment, Seminars, Reviews, and Series, The Lancet will not publish if an author, within the past 3 years, and with a relevant company or competitor, has any stocks or shares, equity, a contract of employment, or a named position on a company board; or has been asked by any organisation other than The Lancet to write, be named on, or to submit the paper (see Lancet 2004; 363: 2-3).
In a perfect world, scientific observations would be interpreted based on their scientific merits alone. The conclusions should follow the trail of evidence. However, biases are as much a part of the human condition as is anything else. A complete disclosure on part of the authors is certainly a step in the right direction, because it would possibly help in the evaluation of the interpretations by the reviewer or reader. But I don’t know what that would achieve in the long run. Would the concept of scientific integrity be obsolete? Would we, say, automatically discount the results of a clinical study or a drug trial merely because the lead author or one of the authors has an industry-sponsored grant? Would it be warranted to expect that in order to avoid such fatal conflicts of interest a prospective author should recuse himself or herself from a proposed study, if the said author has any, close or remote, direct or indirect industry ties?
Lancet asks the authors to fill out the Conflict of Interest Declaration form adopted by the International Committee of Medical Journal Editors (ICMJE; PDF here). What I found highly intriguing in that form is the requested Non-financial Disclosure at point #5.
Please report any personal, professional, political, institutional, religious, or other associations that a reasonable reader would want to know about in relation to the submitted work.
Really? Political and religious affiliations? Are the scientific conclusions expected to change because the lead author is a Democrat or Republican, or supports Tory or Labor, or is an atheist or a born again Christian? What possible value can this disclosure add to the scientific merit of the study?
Shall we discuss this? As always, I invite your opinions and suggestions.
1: Heh! I found it amusing that they have a “Department of Error” – it was reminiscent of the Improper Use of Magic Office of the Department of Magical Law Enforcement, Ministry of Magic! Oh, JKR! How have you corrupted my muggle mind…
A full disclosure to the editor about possible conflicts due to personal relationships etc?? Many times, the author is not aware of this at all. I think that apart from financial relationships other conflicts, though relevant, will be difficult to keep track of.
About your question reg political and religious affiliations…the answer is yes, while the truth is in the data, conclusions do (unfortunately) lie in the eye of the beholder. We do not live in an ideal world, there are several scientists who are simply in the wrong business, e.g. Jan Hendrick Shoen.
conflicts can also occur because of ,,, intellectual passion
I see what they mean, but I’d have thought some intellectual passion was an essential ingredient of being a scientist? How do you measure when it is in excess?
But do these guidelines refer only to primary research papers, or to authors of any Lancet article? The Lancet publishes many types of article, including commentary, opinion, news, and reviews. I would certainly expect an author to declare any relevant political ties if they had contributed an opinion piece on science policy, for example. Likewise, one could imagine certain articles touching on religion (people abstaining treatment on religious grounds) and, in some cases, we would want to know whether the correspondent has a close interest in the religion discussed.
@Sangita: Exactly. I don’t know how this would work in practice. Re your point about truth vs. conclusions, I don’t know if I necessarily subscribe to that. You’re right that the truth is in the data, and I think that is the strongest point. Conclusions may well vary from investigator to investigator, but the observations pertaining to a biological fact won’t. Therefore, one would find enough corroboration from multiple sources and it is possible to reach a scientific consensus about experimental observations. It is acceptable to have divergent view points – ‘scientific conclusions’ is not an immutable quantity. As long as the conclusions are based on available evidence and bolstered by empirical corroborations, they are valid… The disclosure of political and religious affiliations becomes irrelevant.
@Frank: Or, how do you determine when it is the right kind of passion, as opposed to the wrong kind… These questions are very valid, and I doubt that The Lancet has the relevant answers.
@Matt: It appears that these guidelines apply to all the Lancet group of journals, and perhaps to any journal that adheres to the ICMJE guidelines. I was hoping someone from the Nature magazine would comment on this.