Vaccination News Home Page                                            subscribe Vaccination NewsLetter

http://bmj.com/cgi/content/full/326/7394/883

BMJ 2003;326:883 ( 19 April )

Letters

Competing interests

    Consent was not obtained
    Problem is greater than editorial indicates
    It's my journal, and I'll write if I want to
    Authors' reply
 

Consent was not obtained

EDITORSo, the editor of the BMJ is happy to coauthor research involving undeclared deception of subjects and publish it in his journal.¹ Three hundred readers were unwitting dupes. They took part without being informed what the real object of the exercise was, no informed consent here, at least not by the standard of being willing to show the subject the research protocol.

This sort of research violates Kant's categorical imperativeact as if this were a universal principle. It seeks payment in a coin that it debases for others.² BMJ readers should be warned: next time a researcher contacts you for an opinion, it is probably a hoax.

What is ironic about the paper is that the results have meaning only if the respondents were more honest than the authors. My warning to researchers is as follows: if you deceive your subjects what right have you to expect they will not do the same to you?³

Stephen J Senn, professor of pharmaceutical and heath statistics. 
University College London, London WC1E 6BT stephens@public-health.ucl.ac.uk

Competing interests: SJS is a consultant to the pharmaceutical industry and an academic. His career is therefore furthered by publication.

Problem is greater than editorial indicates

EDITORThe BMJ should be congratulated for its efforts to avoid bias in the reporting of scientific work.¹ Asking editors and authors explicitly to state competing interests reduces the likelihood of bias. As Smith rightly says, however, we still have some way to go to the fully transparent world. Much of the discussion of conflict of interest may leave the impression that the problem lies in commercial sponsorship, in particular from the pharmaceutical and tobacco industries.

The problem may be more widespread, however. Could researchers in a public health institute be influenced by political pressures when the institute is financed directly by the health ministry? Will the pressure become stronger when the research institute is placed within the ministry? Can researchers who receive honoraria for advising government and courts on tobacco issues be influenced by the fees they receive? If researchers can be influenced by commercial sponsorship, and the evidence here is convincing, why should researchers be immune to influence from other sponsors? Nevertheless, it is not unusual that research financed by government and other non-commercial sources is presented without warning the reader that there is a potential conflict of interest.

My point is not that bias from commercial sponsorship should be belittled but rather to emphasise that the problem is greater than Smith's editorial may indicate. Policies to avoid bias in the conduct and reporting of research should be guided by scientific principles, not by moralism or prejudice.

Ivar S Kristiansen, senior researcher. 
Institute of Public Health, University of Southern Denmark, DK-5000 Odense, Denmark ivarsk@c2i.net

Competing interests: ISK has received salary and honoraria from several public institutions and honoraria from several pharmaceutical firms. He has not received honoraria from the tobacco industry.

It's my journal, and I'll write if I want to

EDITOR

Richard and colleagues just walked Through the door

Like a king with his court.

He says he's had a great idea

And surely to publish they ought.

Sorry, my muse left on holiday after that. Please feel free to add some more verses. Like other respondents, I applaud the BMJ 's crusade to enlighten readers about the issues of conflicting interests.¹ The paper by Chaudhry et al should surely have been submitted to another journal, or if not then someone else should have it.² The BMJ 's peer journals are not the BMJ itself.

Could we be informed as to how long the paper was out at the reviewers and how quickly it got accepted in its final version? It was published within six weeks of acceptance. That's nice. I have had work published in the BMJ, on one occasion after the manuscript (and I am quoting directly from the correspondence at the time) "lay in the top drawer" of a staff statistician's desk for six months. I doubt that would ever happen to an editor's paper.

These are small points but if the BMJ 's crusade is to be credible and successful, inhouse guidelines about staff submissions should be torn up and replaced by a rule that BMJ related work is only submitted elsewhere.

Jonathan O'B Hourihane, senior lecturer, infection, inflammation, and repair. 
Mailpoint 218, Southampton University Hospitals NHS Trust, Southampton SO16 6YD j.hourihane@soton.ac.uk

Competing interests: None declared.

Authors' reply

EDITORMedical researchers like Senn are often very concerned about minor deception, but its use is common in the social sciences. We at the BMJ have debated the use of deception with our ethics committee, and it sees no problem so long as the deception is minor, the study would be hard or impossible to do in any other way, and participants are informed afterwards.

We agree with Kristiansen that all forms of financialand non-financialconflict of interest are important. Our anxiety is not only about commercial sponsorship, and we have had examples in Britain of the government trying to interfere with the publication of research. We disagree, however, with the implication in Kristiansen's letter that the use of scientific principles can avoid the influence of conflicts of interest. The evidence suggests otherwise.

There are strong arguments against editors publishing original research in their own journals and egregious examplesfrom Cyril Burt and Hans Eysenckof editors publishing highly dubious research in their own journals. Nevertheless, it makes sense for editors to try to publish in their own journals when the research is conducted on readers, authors, or reviewers associated with the journals and the results influence the policies of the journals.

That was the case with this research, and we do have a declared method of reviewing research submitted to the BMJ by the editorial staff that excludes editors employed by the journal at every stage. And we have had several papers rejected.

Richard Smith, editor. 
Sara Schroter, research fellow. 
BMJ, London WC1H 9JR

© 2003 BMJ Publishing Group Ltd

Related editorials in BMJ:

Making progress with competing interests .

Richard Smith
BMJ 2002 325: 1375-1376. [Full text] [extra: Correction to BMJ 2002; 325:1375]  

 

Other related articles in BMJ:

PAPERS
Does declaration of competing interests affect readers' perceptions? A randomised trial.

Samena Chaudhry, Sara Schroter, Richard Smith, and Julie Morris
BMJ 2002 325: 1391-1392. [Full text]  

Vaccination News Home Page

ALL INFORMATION, DATA, AND MATERIAL CONTAINED, PRESENTED, OR PROVIDED HERE IS FOR GENERAL INFORMATION PURPOSES ONLY AND IS NOT TO BE CONSTRUED AS REFLECTING THE KNOWLEDGE OR OPINIONS OF THE PUBLISHER, AND IS NOT TO BE CONSTRUED OR INTENDED AS PROVIDING MEDICAL OR LEGAL ADVICE.  THE DECISION WHETHER OR NOT TO VACCINATE IS AN IMPORTANT AND COMPLEX ISSUE AND SHOULD BE MADE BY YOU, AND YOU ALONE, IN CONSULTATION WITH YOUR HEALTH CARE PROVIDER.