Richard Smith: The case for workshops on publication and research ethics (and an invitation)

“Imagine that you write a scientific paper on the influence of various factors on drug compliance. You are not writing about a specific drug, but a student who is working with you has a small grant from a drug company. Should you declare this as a conflict of interest?”

This is a case that I have been using for years in a workshop on publication and research ethics.

Many people will think that there’s no point in declaring something so trivial that could not possibly have affected your judgement. It’s not clear if the student is a co-author, but even if she or he is it could not have affected your judgement. It’s a general subject, nothing specifically related to the company. Plus, you are aware of evidence that if you do declare a conflict of interest readers will think less of your paper, which seems unfair.

Others take the view that all conflicts of interest should be declared no matter how minor. As an author you cannot know if the conflict of interest has affected your judgement: the bias operates unconsciously. We have to do double blind randomised trials to evaluate interventions not because people are consciously fiddling results, but because bias is subtle and pervasive. We have lots of evidence that conflicts of interest do affect judgements. The policy should be “if in doubt declare” or “declare everything even something that seems irrelevant.” One reason for such a policy is that if you don’t declare a conflict of interest and later an editor or reader makes you do so it looks as if you have been dishonest, hiding something.

The point of the case is that there is no “right” answer and that people make different decisions. After people in the workshop have read the case I ask them to vote, and often the vote is split 50:50. I then get people from each side to try and convince those from the other side that they are right.

Then I ask questions like: What if it’s a large grant rather than a small one? What if the student is a co-author? What if the person is not a student but a shareholder, employee, adviser to the company, somebody who has been paid by the company to give a talk, or somebody who attended the talk given by somebody paid by the company and was given a glass of indifferent red wine and two pens?

My point is that publication and research ethics (like all ethics) cannot be reduced to a set of rules that you simply memorise and follow. There is no substitute for thought and judgement. (Having said that, I do share with the people in the workshop links to the guidelines of the Committee on Publication Ethics and the International Committee of Medical Journal Editors). 

I have a dozen cases that I use, all of which are designed to divide opinion. But often these days I don’t use them at all—because I start by asking if anybody has had a problem. Years ago people would share problems with me after the workshop, but these days somebody usually volunteers a problem at the beginning. The commonest problems relate to authorship, and the single commonest problem is senior colleagues, often supervisors, insisting on being authors on papers where they have done little or nothing. In many cultures it is impossible for a younger researcher to resist this demand. We often conduct a role play with me playing the demanding, unreasonable senior (a role I enjoy and find embarrassingly easy).

Stephen Lock, my predecessor as editor of The BMJ, was one of the first people in Britain to become concerned about research fraud or misconduct—back in the 1980s. The establishment dismissed his concerns by arguing that misconduct was rare, nobody was harmed, and science was anyway self-correcting. We now have ample evidence that all these arguments are fallacious. Misconduct is common, and it’s important that anybody conducting research has some chance to learn about publication and research ethics and recognise the scrape that they might fall into inadvertently. In the US all PhD candidates have to do online courses on misconduct, but there are worries that these have degenerated to “tick box exercises.”

I think that it’s important to discuss cases and reflect on the difficulties, which is why I run my workshops. I’m not looking for more work (and am happy to share my cases with anybody), but I am running a workshop with Sara Schroter from The BMJ, who has done more research into scientific publication than perhaps anybody the London School of Hygiene and Tropical Medicine on 25 March from 2-4pm—and it’s open to doctoral students and staff at LSHTM and doctoral students from the Bloomsbury Postgraduate Skills Network (see below for list of members). If you are eligible, please come.  Contact to book your place. I promise you’ll have fun.

Bloomsbury Postgraduate Skills Network

University College London (founding member)
City University
King’s College London
London School of Economics
London School of Hygiene and Tropical Medicine
School of Advanced Study
School of Oriental and African Studies
Royal Veterinary College

Richard Smith was the editor of The BMJ until 2004.

Competing interest: RS has run his workshop in many countries and institutions. Usually he is not paid, but he has been paid intermittently by the London School of Hygiene and Tropical Medicine, where he has been running the course for around 15 years.