SEARCH

SEARCH BY CITATION

It is highly damaging to issue press releases on purported addiction research findings before these have been checked by peer review. Journals and fellow scientists should take a stand against this.

In a witty essay whose central concern sounds almost quaint just 15 years later, Babor [1] defended the decision of the Project MATCH research group to not broadly share their findings until they had been published as a paper in a peer-reviewed journal. In the internet-connected, blog-saturated, tweet-generating and journalistically slapdash world of today we should be concerned about behaviour at the other end of the publicity-seeking spectrum: scientists actively distributing non-peer-reviewed findings through the media. The recent media frenzy concerning the scientific ‘proof’ that cocaine is no more addictive than Oreo-brand cookies provides a vivid example.

A few days after the sensational story began spreading on internet news sites and blogs, other scientists were contacted by responsible journalists for their views of the study. This quickly made apparent that the research in question came from a press release arising from an undergraduate student project that had not been published or even presented at a conference [2]. The contents of the press release were passed along uncritically by many news outlets (e.g. [3, 4]). Facebook, Twitter, Reddit and the like did the rest.

Then a round of news stories and commentaries appeared which noted serious flaws in the research and questioned its astonishing conclusions [5, 6]. Following the backlash, the college news service changed the sensational title of the press release and added a new line to its text: ‘The results are preliminary and subject to further scientific review’. Indeed, but at that point the closing of the stable door was but a distant sound in the ears of the horse that had galloped away.

What is the harm of scientists' releasing conclusions that have not been subject to peer review directly to the media—and is there anything the scientific community can do to reduce its incidence?

In the case of the Oreos study, one can imagine specific harms from the trivialization of a serious addictive disorder, but arguably the greater potential damage of sensationalized press releases is the erosion of the standing of scientists with the public. The authority of scientists is a hard-earned and precious commodity and can literally save lives, such as when the public believes the chief scientist in a health agency who announces that an effective vaccine is available for a deadly new infectious disease. Every time a scientist makes headline-grabbing claims which are then debunked, this authority is diminished; the public comes to see us as just one more group spouting unsubstantiated opinions that are no more credible than those of any others.

Science reporting by press release also sets a terrible precedent that could be exploited by the willfully malign (which the Oreos cookie researchers were clearly not). If it is acceptable for academic researchers to distribute non-peer-reviewed claims to the press, then the designer of a new addiction treatment can surely feel comfortable in making extravagant claims of success in the same way. Similarly, an industry-funded tobacco researcher could cite prevailing academic standards as a reason to release ‘shocking new evidence’ that cigarettes are not that dangerous after all.

One common response to the problem of scientifically inaccurate news stories is to blame journalists for writing stories directly from press releases without bothering to verify that the release is credible. Because of the economic collapse of much of traditional news media, fewer journalists are available to cover science carefully [7]. Although this is lamentable, it is not something that researchers can change. That said, if scientists respond to this situation by directly feeding unpublished, non-peer-reviewed results to the media they can hardly cast themselves as innocent victims of the decline in journalistic standards. That link of the chain is indeed something scientists can influence, both by the example they set themselves and by the norms they establish collectively.

There must be costs to deter responding to the siren call of media attention without the discipline of peer review. One such cost is the opprobrium of colleagues (e.g. [8]), but this is probably too inconsistently imposed and too small of consequence to be a significant deterrent. A more effective alternative would be to establish a policy at our journals that derives from the ‘Inglefinger Rule’ [1]. As editor of the New England Journal of Medicine, Inglefinger admonished authors not to ballyhoo the findings of accepted papers prior to publication, and threatened to withdraw acceptance if they did. The parallel rule for the present situation would be for Addiction and the other members of the International Society of Addiction Journal Editors to refuse to accept for peer review any scientific results that have already been actively disseminated to the media. Beyond its worth in communicating a set of shared values (i.e. peer-review is important), it should also serve to elevate the caliber of science that ultimately attracts wide attention, which will benefit both the public and the scientific community.

Declaration of interests

None.

Acknowledgements

  1. Top of page
  2. Acknowledgements
  3. References

Preparation of this paper was supported by a VA HSR&D Senior Research Career Scientist Award. Kristy Nielsen made helpful comments on an earlier draft of this paper. The views expressed are the sole responsibility of the author.

References

  1. Top of page
  2. Acknowledgements
  3. References