BMJ  2003;327:503 (30 August), doi:10.1136/bmj.327.7413.503-a

Letter

Passive smoking

Agreeing the limits of conflicts of interest

Editor—The paper by Enstrom and Kabat¹ raises the issue of how much conflict of interest can editors reasonably allow before the findings and interpretation of a particular study are rendered unsafe or, at the very least, too uncertain to be a substantive scientific contribution?

If we think that there really is a limit to the degree of conflict that we judge reasonable, as some responses to the Enstrom and Kabat paper seem to imply, then criticism should be directed to the medical community for having such imprecise thinking over conflicts of interest. In pharma sponsored studies, we mostly allow conflicts provided they are reported accurately. We deplore them in tobacco sponsored research. But there are many examples of how both industries have tried to undermine the independence and rigour of research, bias policy makers, and gouge huge profit from disease.

In papers from the pharma industry we publish a statement about the role of the funding source in the design, conduct, analysis, and reporting of the data for all primary research, irrespective of who the sponsor might be (for-profit, not-for-profit etc). No such statement appears in the Enstrom and Kabat paper—would this have helped readers judge the safety and reliability of their research?

Could this paper therefore provide a useful opportunity for us all to clarify what is an acceptable conflict, for readers, researchers, and editors alike, and how that conflict should be reported? Could we agree also about how to handle these matters during prepublication peer review (should the extent of the conflict be a factor, in addition to the science, in deciding acceptance or rejection?)—well before they might confuse an already difficult scientific issue of great public concern?

Lancet London NW1 7BY richard.horton{at}lancet.com

---

Competing interests: None declared.

References

1. Enstrom JE, Kabat, GC. Environmental tobacco smoke and tobacco related mortality in a prospective study of Californians, 1960-98. BMJ 2003;326: 1057. (17 May.)[Abstract/Free Full Text]

Related Article

Environmental tobacco smoke and tobacco related mortality in a prospective study of Californians, 1960-98

James E Enstrom and Geoffrey C Kabat
BMJ 2003 326: 1057. [Abstract] [Full Text] [PDF]

Rapid Responses:

Read all Rapid Responses

Passive smoking: Non-smoking researchers’ personal values cause conflict of interest.

Erik Nord
bmj.com, 10 Oct 2003 [Full text]

Passive Smoking Risks: From Pollution, Not Bias

James L. Repace
bmj.com, 2 Dec 2003 [Full text]