Sunday, 14 December 2014

Beware the inflated science related press release!

I'm not normally minded to post on a Sunday (day of rest and all that) but I did want to bring your attention to the results presented by Petroc Sumner and colleagues [1] (open-access) concluding that: "Exaggeration in news is strongly associated with exaggeration in press releases" when it comes to the media reporting of [some] health-related science news.

The idea behind this particular study - which has been summarised pretty well in some of the accompanying media and in an editorial in the publishing journal [2] - was to look at the contribution of the press release (the 'look at me' part of some science publishing) to those sometimes 'inflated' health science headlines which we all encounter on a day-to-day basis. The results suggested that whilst [some] journalists and editors might need a refresher course on some of the basics of science (including the concept of probability and risk), academic press offices and even the very academics behind said research might also have to shoulder some responsibility (bearing in mind correlation is not the same as causation!) So when for example, a mouse study correlating the presence of compound X with improvement in the [mouse] condition Y is interpreted as meaning that compound X is the elixir of life for Homo sapiens and that we should instantly rush out and buy as much of it as we can afford, look to the press release before blaming the newspaper. I exaggerate of course with that example.

So as not to keep you from your Sunday breakfast or that magnificent Sunday lunch you no doubt have planned/are eating, I'm not going to say much more about this topic aside from making two additional points:

(a) Post-publication peer-review is mentioned somewhere in one of the texts. Although this is traditionally meant to imply that researchers who have a beef about some study or interpretation of results send a letter to the publishing (or other) journal outlining their issues, there are quite a few other mediums these days which seem to get the job done just as well. You are reading one of those mediums - yes, bloggers of the world unite - and how on occasion, blog entries have talked about press releases and peer-reviewed results not quite tallying together. I dare say someone somewhere might eventually start a blog titled something like: 'pressing ahead: press release vs. actual results' if it hasn't already been done. Other social media might also play an important role in highlighting discrepancies.

(b) Although interesting, the Sumner results perhaps leave out one very important variable in this scientific producer - consumer relationship: the consumer. You might well scoff that anyone without a PhD or related qualification 'doesn't understand science' but I would say that you are wrong. Although there are people out there who actually believe the term 'scientifically proven', as if God himself endorsed the product, I'd suggest that there are enough people who don't believe every headline they read and are vocal enough to say so. Search engines such as Google do a pretty good job of ensuring that anyone researching a scientific claim important to them (which most people do these days) quickly get both sides of the story. In that respect, an inflated press release (which are also generally detectable on the web) will probably not stay inflated for too long. Probably less so, after this and other research [3] starts to percolate through the web...

Without further ado, I'll leave you to finish of your 'super-food' packed Sunday lunch. And just in case anyone is interested, a few things to bear in mind when reading and interpreting science...

----------

[1] Sumner P. et al. The association between exaggeration in health related science news and academic press releases: retrospective observational study. BMJ 2014; 349: g7015.

[2] Goldacre B. Preventing bad reporting on health research. BMJ 2014; 349: g7465.

[3] Woloshin S. et al. Press releases by academic medical centers: not so academic? Ann Intern Med. 2009 May 5;150(9):613-8.

----------

ResearchBlogging.org Sumner, P., Vivian-Griffiths, S., Boivin, J., Williams, A., Venetis, C., Davies, A., Ogden, J., Whelan, L., Hughes, B., Dalton, B., Boy, F., & Chambers, C. (2014). The association between exaggeration in health related science news and academic press releases: retrospective observational study BMJ, 349 (dec09 7) DOI: 10.1136/bmj.g7015