Conflict of interestBMJ 1996; 313 doi: https://doi.org/10.1136/bmj.313.7071.1555a (Published 14 December 1996) Cite this as: BMJ 1996;313:1555
Conflict of interest statement should be abolished
- C K Connolly
- Consultant physician Department of Medicine, Darlington Memorial Hospital NHS Trust, Darlington DL3 6HX
EDITOR,—Since the BMJ introduced the requirement that authors should declare any conflict of interest at the foot of papers and other articles1 such conflicts have rarely, if ever, been admitted. I wonder whether this reflects the truth.
The purpose of scientific publication is to present the results of observations or experiments in such a way that readers can assess their veracity and importance in terms of scientific knowledge. In the case of a medical journal, this includes their relevance to clinical practice in cost-benefit terms in their widest sense, and hence their relevance to the allocation of resources. Only when there are absolutely no outside influences or pressures that might modify the presentation of results in such a way as to compromise their interpretation can there be said to be no conflict of interest. Scientists must be objective to a saintly degree if they can truly state that their wish to justify a substantial proportion of their life's work or what they regard as their most brilliant innovation has never influenced the mode of their presentations. This might merely be to emphasise those results that prove their theory or to press harder for publication of those things that they feel most deeply about. Indeed, it could be said that it is the innovator's duty to do these.
If an author has accepted funding from any pressure group or from any charity that restricts itself to funding particular diseases or groups of the population, let alone any money from a government agency for its own agenda, then it is difficult to avoid conflict of interest. This might arise from an agency's funding policy or, where a “popular” charity has particular ease in raising money, from distortion of the perceived need, because the greater the funds available the more the publications about that particular subject.
A change in the wording to “no undeclared conflict of interest” would solve the second group of problems but not the first. Perhaps it would be better to end the charade altogether and return to the old fashioned virtue of trusting the integrity of authors. This might allow a few charlatans to get away with it, but one should not forget that mistrust, like other forms of pessimism, is a self fulfilling prophecy.