Issue 47 / 15 December 2014

‘TIS the season to be jolly … or jolly loose with scientific claims anyway.

Actually, it may always be the season for that judging by some thought-provoking research from Cardiff University in the UK.

When exaggerated stories about medical “breakthroughs” appear in the general media, it’s all too easy to blame the journalists. But is that really the whole picture?

These researchers, including one from Australia, examined health-related press releases issued by 20 leading UK universities over the course of a year and found a strong correlation between exaggerations in the release and in subsequent news stories.

A whopping 33%‒40% of university press releases exaggerated the findings of research when compared with the original journal article (33% made exaggerated causal claims, 36% made exaggerated claims for effects in humans based on animal research and 40% contained exaggerated advice).

On top of that, there are the possible exaggerations in the journal articles themselves, but that’s beyond the scope of this research.

Here’s how one journal article describes its results: “This observational study found significant associations between use of antidepressant drugs and several severe adverse outcomes in people aged 65 and older with depression”.

By the time the university issued its press release, that correlation had become a causal relationship: “New antidepressants increase risks for elderly”.

Another journal article described an animal model of an “important component of memory consolidation”.

The press release claimed: “Scientists have shed light on why it is easier to learn about things related to what we know”.

It probably is, but what do we know about that kind of scientific hyperbole?

Universities, scientific journals (I’ve written about their press releases before), and scientists themselves, all have an interest in achieving maximum media coverage for their “ground-breaking” research – and sometimes scientific rigour goes out the window as a result.

An earlier online survey conducted by the researchers found scientists overwhelmingly blamed media outlets for exaggerated science reporting, although a smaller proportion did acknowledge some responsibility on the part of scientists and press officers.

An extraordinary 40% of scientists who had press releases issued about their work acknowledged their most recent release had been exaggerated.

This figure declined with greater levels of involvement from the scientist, but remained above 30% even when the scientist had actually written the release.

That’s one-third of scientists admitting they had written exaggerated claims about their research in a press release.

“… as a group we appear to engage in doublethink”, the Cardiff researchers wrote in their analysis of these results, “colluding in producing exaggerated [press releases] but mainly blaming the media for the shortcomings of science news”.

The researchers stress they are not seeking to shift the blame for exaggeration in medical news stories “from one group of non-scientists (journalists) to another (press officers)”, suggesting most of the responsibility must lie with the scientific authors who approve releases before they go out.

And then there is the bigger picture.

“The blame — if it can be meaningfully apportioned — lies mainly with the increasing culture of university competition and self promotion, interacting with the increasing pressures on journalists to do more with less time”, they write.

Those are tricky problems to solve, but an accompanying editorial by that tireless campaigner for evidence and accountability in health care, Dr Ben Goldacre, does make some suggestions.

Press releases should be treated as an integral part of the scientific publication, he argues, linked to the paper and referenced from it. And journals should publish commentary and letters responding to misrepresentations in the press release just as they would about the paper itself.

On top of that, both press officers and the authors of the original scholarly paper should be named authors of the press release, creating “professional reputational consequences for misrepresenting scientific findings in a press release, which would parallel the risks around misrepresenting science in an academic paper”.

They’re good suggestions but, in the meantime, perhaps we should all join in a carol to celebrate the season:
Hark the herald PRs trill
Glory to the new-found pill
Works in mice, in humans too
Take it now, it’s good for you

A happy festive season to you all.
 

Jane McCredie is a Sydney-based science and medicine writer.

2 thoughts on “Jane McCredie: Happy hyperbole

  1. Randal Williams says:

    I have been banging on about this for years — media breathlessly announces the latest “breakthrough” in cancer treatment, building up hopes, then we find that human trials are (at least ) five years away so the actual treatment is proabably ten years away, if it ever happens. This situation is rife across TV, print media and ( increasingly) social media. Researchers are partly to blame in their desire to get publicity and further funding. It  is cruel trick for those sufferers of the diseases in question, having their hopes raised then dashed on a regular basis.

  2. Donald W. Light says:

    When it comes to prescription drugs, an overview of our research at Harvard’s E J Safra Center for Ethics found that 90 percent of new drugs each year (for the past 30 years) have been found by independent review bodies to offer few or no clinical advantages over existing ones.  So hype press can mislead patients and doctors.

    Alongside this is an epidemic of serious harmful side effects. The risks that “great new drugs” cause enough serious harm to warrant to serious warning or withdrawal is 1 in 5 – a very high risk. Prescription drugs have become a major health hazard. They are the 4th leading cause of death and a major cause of avoidable hospitalizations. 

    All the data and references can be found in this overview: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2282014

     

Leave a Reply

Your email address will not be published.