© 2005 American Thoracic Society doi: 10.1164/rccm.2501005
Primum Non NocereDoes the Current Research Publication System (or the Lay Press) Harm Our Patients?
a Royal Brompton Hospital London, United Kingdom Out of the blue, our unit received a telephone call from a patient, unknown to us at the time, requesting recommendations on the exact dose and formulation of a particular herbal remedy. He knew of several other patients with the same disease who were already taking this compound. The disease in question was cystic fibrosis (CF) and the "treatment" was curcumin, the active component of turmeric. Many of our own patients had been asking about a high-profile study demonstrating the efficacy of curcumin (1), and, illustrating the widely held belief that such herbal compounds are completely natural and benign, the view of this patient was, "What have I got to lose?" Against our advice, he is (and likely others are) now self-administering curcumin-containing formulations in large quantities, of varying and undetermined purity, despite no evidence of benefit in humans and the theoretical possibility of side effects, in particular interactions with conventional medications. Such scientific basis as existed to support the use of curcumin in CF has now largely disappeared, but the consequences, like the memories, linger on. What lessons can be drawn from this episode? We are not at all sure, but what is clear is that we should try to consider if a repetition can be prevented. The basis of this editorial is the belief that all scientists concerned acted impeccably at all times, but the outcome has been disappointing.
The commonest class of mutations in the gene causing CF leads to misfolding of the CFTR protein (2) which then fails to traffic to its site of function. Because it was recognized that CFTR could act as a chloride channel if encouraged to reach the cell surface, for example after cell cooling (3), effort has rightly focused on the search for new pharmacologic therapies to facilitate intracellular trafficking (4, 5). In April 2004, Egan and coworkers published a study in which they had administered high doses of curcumin orally to
Unfortunately, since the original publication, several groups have failed to demonstrate any effect of curcumin on Each person involved in original research has differing responsibilities: those of the scientists are to generate accurate and valid data and to reach appropriate conclusions. The latter should be enhanced greatly by the input of good reviewers. However, no matter how rigorous a job these reviewers attempt, they have only the submitted data to aid their judgment. One way around this would be the requirement for authors to submit all raw data with the manuscript. This could certainly allow the identification of flaws in analysis or interpretation, but would greatly increase the workload of the reviewer and would not prevent deliberate scientific fraud. Another solution is for objective investigators to observe repetitions of the experiments, or for results to be duplicated before publication. The former approach was adopted for a paper allegedly providing proof of principle of homeopathy (12). The report of this visit was highly critical of the original experiments (13), and the matter ended with accusations of McCarthyite persecution and witch hunting comparable to Salem. The massive public concern surrounding the MMR vaccine, and subsequent catastrophic drop in immunization rates in the United Kingdom after the publication of a single paper (14) (the interpretation of which has now been formally refuted by several of the original authors [15] and repudiated by the Editor of the journal in which it was published), could possibly have been prevented altogether by such an approach. It is unlikely to be necessary in the majority of cases, in which the work submitted will have little immediate impact. However, in principle, it ought to be possible to revive this approach at the discretion of the editor. The downside, in addition to the increased work load, is the delay in publication and the chance of being "pipped to the post" by another group. As long as scientific careers depend on publications for their progression and even survival, such an approach is therefore inherently limited. Finally, but perhaps most importantly, papers in scientific journals are targeted at other members of the scientific and medical community, most, it is hoped, possessing the skills to interpret the data and set them in the correct context. This may be beyond the scope of many lay writers and editors, who nevertheless elect to condense and "spin" the data in their own publications, feeding sensational morsels to a hungry public. It is from these lay publications that our patients with CF and their families gain most of their information. How do we reconcile "responsibility toward patients" with the motivations of newspaper and magazine editors? Toning down the content of a report and emphasizing caution in interpretation is unlikely to enhance sales. But we consider that the lay press should not be allowed to evade their share of responsibility; power without responsibility is the prerogative of the harlot. Perhaps some lay editors might consider the scholarly and balanced tones of the reviews on curcumin that came out in the scientific press. Are we overreacting? This specific problem is probably not widespread within the CF community and, to our knowledge, nobody has yet come to any harm. Patients have recovered from dashed hopes before and manage, in general, to maintain their faith in the medical and scientific professions. However, we would suggest that now is the time to debate these issues and to question whether there are useful changes that could be made to the research publication system before either significant harm does result, or we finally lose the goodwill and trust of our patients. FOOTNOTES Conflict of Interest Statement: J.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; A.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
F508 CF mice (
