The Washington Post, Christian Science Monitor and Time were among the respected media outlets that recently covered the sensational alleged scientific finding that Oreos are as addictive as cocaine. Some journalists contacted me for my reaction, and I said I would get back to them after I read the study. I searched on PubMed and it wasn’t there. I looked through the “on line early” section of leading journals and it wasn’t there either. Neither could I find it in the published proceedings of any conferences.
Relapsing to a simpler approach, I Googled “Oreos and cocaine” and found the sole source of information available: A press release about an undergraduate research project at Connecticut College in New London, Connecticut. It is to the credit of the mentor on the project, Joseph Schroeder, PhD, that he’s teaching bright young people how to do science. I wish, however, he had also taught them something else: The role of peer-review in the promotion of scientific quality and credibility.
Putting out a sensational press release before experts in your field have had a chance to evaluate your scientific work is bad for science and bad for society
A leading scientist in the addiction field, Edythe London, PhD, of UCLA, was able to identify fatal flaws in the Oreo research even on the basis of the limited information in the press release. You can read Dr. London’s discussion of those problems if you wish, but my purpose here is to address a larger point: Putting out a sensational press release before experts in your field have had a chance to evaluate your scientific work is bad for science and bad for society.
The Oreo news stories were upsetting to people struggling with cocaine addiction as they trivialized the problem they face. Judging by the comments on many of the news websites covering the story, the study also fueled many people’s bias that taxpayer dollars are being wasted on studying silly topics. Granted, these harms are less serious than those that occurred when Andrew Wakefield, MD, intentionally deceived many journalists into reporting that the MMR vaccine caused autism, but that still doesn’t make it good practice to report initial findings as facts in the media before one’s work has been peer-reviewed.
Credibility-straining press releases are unfortunately becoming more common in science. And even when a finding is real, a press release can distort its meaning. For example, the recent Qesem cave discovery of fossils from homo sapiens was considered a genuine achievement by experts in physical anthropology and was published in a peer-reviewed journal, but the accompanying press release made wild claims about human evolution that were nowhere to be found in the article. When the shocking claims in that press release were disseminated by journalists, the public was misled and the science was cheapened.
Sensationalized press releases about scientific findings unfortunately can be much more widely disseminated today than in prior eras. Science journalists once had the time and resources to do more shoe leather reporting, conduct follow-up interviews with the scientific team, and interview independent scientists to obtain a careful critique of the findings. But today, as Paul Costello, chief communication officer for Stanford’s medical school, told the Columbia Journalism Review, the “shift to new media Web site traffic” is putting added pressure on reporters, leading some to cut corners in the name of more copy, “often writing right off press releases, even at the good papers.”
If science journalists can no longer be depended on to sort the wheat from the chaff, scientists are going to have to assume more responsibility for self-restraint when they release findings to the press. All of us would like to believe that we never make mistakes in our scientific work. All of us are capable of becoming overly enthusiastic about a pet theory before our colleagues have subjected it to criticism. And all of us can be tempted to hype our findings as a way to draw some gratifying attention to ourselves. But when we give in to those human frailties in our interactions with the media, we undermine our collective credibility and may also harm members of the public who act on the assumption that our claims have been rigorously evaluated.
Addiction expert Keith Humphreys, PhD, is a professor of psychiatry and behavioral sciences at Stanford and a career research scientist at the Palo Alto VA. He recently completed a one-year stint as a senior advisor in the Office of National Drug Control Policy in Washington.
Previously: The influence of medical press releases on news coverage quality and The problem with “science by press conference”