Whenever you see a news story that contains imbalanced and inaccurate information about health care, you’d be justified in wondering whether the content originated with a poor-quality PR news release.
We’ve been demonstrating how these promotional documents can mislead readers about animal studies, phase 1 research, and minimally invasive surgery — to name just a few of the themes that we’ve hit upon in our problematic PR series.
But there’s another theme we haven’t called attention to and which deserves increased scrutiny from journalists and the public: Many of these problematic messages are broadcast to the world via Eurekalert!, the PR newswire service run by the American Association for the Advancement of Science (AAAS).
AAAS is the world’s largest multidisciplinary scientific society. It organizes conferences, publishes peer-reviewed journals, and generally advocates for science literacy and the responsible use of science in all areas of public life. But in a troubling contradiction, it’s also home to a global clearinghouse for unvetted PR messages from universities, medical centers, journals, drug companies and other organizations who pay the EurekAlert! service to broadcast their new releases.
The messages put out by these organizations sometimes contain egregious levels of spin and exaggeration designed to attract the attention of journalists and the public. That spin and exaggeration can find its way into news stories or get picked up by aggregrators and rebroadcast to the public verbatim. Here are a few choice recent examples:
Again, these are just a handful of the most shoddy recent examples. We also encounter a daily stream of more subtle and insidious messages that arguably do just as much damage — for example by overstating the conclusions that can be drawn from observational research.
Why is AAAS undercutting its own mission by also spreading inaccurate and incomplete information that, in many cases, grossly misrepresents science to the public?
I posed this question to a number of journalists, public affairs officials, and academics. They all responded that EurekAlert! has an obligation to do better.
“I think it’s entirely fair and reasonable to ask EurekAlert!, and other similar services, to apply more active quality control mechanisms when publishing or re-publishing promotional material about scientific research,” said Andrew Williams, PhD, a senior lecturer at Cardiff University’s School of Journalism, Media and Cultural Studies. Williams was one of several co-authors on a seminal BMJ study documenting how exaggeration in health-related news releases is associated with exaggeration in related news stories in the popular press.
“As a platform with unprecedented reach, and a go-to resource for journalists and public relations officers all over the world, [EurekAlert!] is in a unique position to exercise serious influence over the quality and standards of public information about science,” he said.
Matt Shipman, a public information officer at North Carolina State University and author of the Handbook for Science Public Information Officers, agreed that there is much room for improvement here. “I think it would be great to see EurekAlert! and, by extension, AAAS take a stronger leadership role in facilitating the dissemination of health research news that more fully captures the limitations of new research findings. This would certainly be more in line with AAAS’s stated goals of both enhancing communication between researchers and the public; and promoting and defending the integrity of science and its use.”
For its part, AAAS disclaims any responsibility for the messages that it distributes via EurekAlert! via the following statement that appears on its website:
AAAS disclaims responsibility for the accuracy of material posted to EurekAlert! by contributing institutions and for the use of any information obtained through EurekAlert!. Support from sponsors does not influence content or policy.
But Sharon Dunwoody, PhD, a professor of journalism and mass communication at the University of Wisconsin, Madison, suggested that this position is increasingly untenable in today’s digital media environment. “I suspect AAAS does not view EurekAlert! as falling under their ethical purview, that the organization would define the site as a kind of neutral ‘transmitter’ of information,” she said. “But that claim of neutrality is being tested in a host of other sites these days, Facebook for example.”
In response to mounting evidence that the social networking site has been used to spread political disinformation to sway elections, Facebook has expanded its efforts to fact-check news stories and root out the fake accounts that promote them.
Williams seconded the suggestion that, like Facebook, EurekAlert! has a responsibility to safeguard its users from misleading content.
“Public relations newswires might try to present themselves as neutral platforms rather than publishers of information. But as we’ve recently seen with Facebook and other social media organisations, this argument doesn’t really wash,” Williams said. “In contemporary digital communications there’s really no such thing as a neutral platform, and everyone who publishes, or re-publishes, information should be bound by standards of truth, honesty, and integrity. Especially so for scientific information.”
What would an expanded effort to improve the quality of PR news releases look like?
Shipman suggested, for example, that news releases describing an animal study should be clearly tagged as such, perhaps with a flag or button at the top of relevant news items. “This could be useful,” he said, “particularly if reporters (or other readers) could click on those tags to go to an explanatory note articulating both the benefits and limitations of research using animal models. This model could also be used to identify other aspects of a given study, such as whether it’s an observational study, whether it was a double-blind interventional study, and so on — again, with pop-up windows or links that could provide additional information about the pros and cons of these approaches.”
Such a system of voluntary labeling has already been adopted by some UK institutions, Williams noted. “This has prompted lots of useful discussion among practitioners, the agreement of professional standards, and it’ll no doubt be useful in improving practice. Eurekalert! seeking a voluntary agreement like this one from those who publish on the site would raise awareness of the issues, would be useful for accountability, and would probably improve practice too.”
A much greater impact would be possible if EurekAlert! acted more like a traditional publisher, Williams added, actively employing editorial resources and standards to police its content. “That would be a real game-changer,” he said.
Earle Holland, who was the senior science and medical communications officer at Ohio State University for 35 years before retiring, believes it would be a relatively simple matter for EurekAlert! to adopt a rudimentary labeling system like the one Shipman and Williams propose. “All it would take was for their programmers to add a half-dozen new fields to their template that submitters needed to fill out and that would incorporate the kinds of answers you suggest,” he said.
But as someone who was involved with EurekAlert! for more than a dozen years as a national advisory board member, Holland doesn’t expect these suggestions to be embraced by the organization. He described the EurekAlert! team as “largely an IT shop” and said they “don’t see it as their job” to provide editorial oversight.
“My skepticism is based on the numerous times we on the advisory committee suggested improvements that required modifying their model and were rebuffed, being told that their clients hadn’t requested it and they were responsive to who paid them,” he said.
When told that experts in health-related communication were calling for more robust quality controls at EurekAlert!’, the Director of Editorial Content Strategy, Brian Lin, suggested that the newswire is already providing adequate oversight of its content and has little need to do more. “Even though the responsibility to ensure the accuracy of a news release ultimately lies with the issuing institution, EurekAlert! as a distribution platform has always taken an active role in achieving a basic level of integrity in the content we host, precisely because we are part of a scientific society and because we value the trust that reporters, public information officers (PIO), and members of the public place on us,” he said.
Lin pointed to user and content guidelines that he said are “some of the strictest among similar services” and noted that each of the release submissions they receive is reviewed by at least one staff member against those guidelines. “For research-related news releases, for example, we require that the associated paper is published in a peer-reviewed journal, and we encourage PIOs to submit additional information, such as links to the paper, DOI, and third party comments that would support independent reporting.” In addition, he said that EurekAlert! is “looking into the possibility” of incorporating elements of the UK news release labeling system in future system updates, and that it “absolutely” supports basic quality standards for news releases such as those advocated by HealthNewsReview.org.
But Lin’s claims ring hollow to HealthNewsReview.org publisher Gary Schwitzer, who noted that EurekAlert!’s standards haven’t been adequate to stop the blatant misinformation that we’ve encountered on an almost daily basis on our blog and in our news release reviews. “If their requirements are, indeed, ‘some of the strictest among similar services,’ then the entire field is in bad shape – which is, of course, what we’ve documented with our nearly 600 release reviews,” Schwitzer said.
He also took issue with Lin’s claim that they “encourage PIOs to submit additional information, such as links to the paper, DOI, and third party comments that would support independent reporting.”
“Encourage?” Schwitzer responded. “How’s that working out? How many PIOs have supplied third party comments to support independent reporting? They could make that a requirement, but they don’t.”
EurekAlert! isn’t the only PR news service engaging in the distribution of misinformation. Private, for-profit newswire services such as PR Newswire and Newswise enage in similar, possibly worse practices that confuse and mislead the public. But EurekAlert! is the only such service that’s part of a non-profit scientific society dedicated to educating the public about science. As such it seems like a good place to start demanding higher standards.
We’ve laid out criteria upon which these standards could be based and showed how they could be applied in nearly 600 reviews of news releases. Now EurekAlert! has an obligation to take the next step and help put a brake on the health news spin cycle.
“The increasing need and opportunity to ‘sell’ science and scientific institutions, combined with changes in journalism, means that there are now some pretty big structural barriers to the routine production and circulation of sober, accurate science news,” Williams said. “Given their stated aims and growing influence as publishers, representative bodies like the AAAS clearly have a role to play here.”