June 27, 2014, by Brigitte Nerlich

Science, sensationalism and the dangers of over-selling research

This is a GUEST POST by FREYA HARRISON. Freya works in Steve Diggle’s group in the Centre for Biomolecular Sciences at the University of Nottingham, where she researches the ecology and evolution of cooperation. She spends most of her time exploring how communication and cooperation help bacteria to cause chronic infections, but she is also interested in social behaviours in vertebrates, including humans.


I’ll admit it, I get very excited when my research is picked up and reported on by the news media. But when I read online that my study of parental care in birds clearly indicated that governments should stop giving welfare payments to single mothers, excitement was quickly replaced by consternation. I was unnerved that a blogger had misinterpreted my article so egregiously to bolster his own political views (views with which, incidentally, I disagree). The research explored what types of behaviour make biparental care evolutionarily stable. At no point did I or my co-authors use this as a basis to make moral judgements about human parenting.

The above might be an extreme example of extrapolating beyond the results of a study to draw conclusions that are not justified, but it’s part of a wider pattern of how the research pudding can be over-egged in the media. We have some wonderful science journalists in the UK, aided by an army of STEM professionals who engage with and support the popular reporting of research. But at times, a combination of the journalistic imperative to write an exciting story and the increasing pressure on researchers to ensure their work has impact can lead to the implications of a piece of research being oversold.

There are notorious examples of scientific discoveries being thrust into the media spotlight before they have been thoroughly checked for veracity and repeatability (the MMR-autism debacle, the arsenic life saga and, most recently, the possibly premature excitement over BICEP2). Less serious but perhaps more insidious is the temptation to report every finding as a “breakthrough,”every discovery of a genetic allele linked to a disease as a “potential cure”, every highly-parameterised implementation of a problem from game theory as a “new way of understanding evolution.”

Lipopolysaccharides and antibiotic resistance

Let’s dissect a recent article that received coverage in the mainstream news media. Last week’s edition of Nature carried an article which reported key new information on how some bacteria make a protective outer membrane. So-called “gram negative bacteria”produce a tough outer layer of molecules called lipopolysaccharides or LPS.   Because the LPS envelope protects the bacteria from many conventional antibiotics, infections by gram negatives can be hard to treat. In humans, LPS also triggers an inflammatory response and, if it reaches the bloodstream, can cause fatal toxic shock. Many key human pathogens, including E. coli and Salmonella species, belong to the gram negative group. The new work revealed the structure of a protein that shunts freshly-made LPS from inside the cell to the outer membrane. To steal a lovely metaphor from the BBC, it acts as a molecular “bricklayer.”

This paper was widely reported as having identified the “Achilles’heel”of antibiotic-resistance bacteria (e.g. in the Telegraph and the Independent) – a phrase taken directly from the authors’press release. Vice went as far as to announce that the discovery may “end antibiotic resistance forever.” All of these articles relied on a quote from the lead author, again taken from the press release, stating that because drugs designed to target the “bricklayer”protein would not need to enter the cell “we hope that the bacteria will not be able to develop drug resistance in future.”

Now, this is a very odd thing to say. It is certainly true that most antibiotics need to enter the cell to kill it, and one very common way bacteria evolve resistance to these drugs is to pump them out before they can do any damage. These pumps are often encoded on mobile pieces of DNA that can be hot-swapped between bacteria, allowing resistance to rapidly sweep through populations. This avenue to resistance is clearly not an option for drugs that work outside the cell. However, bacteria can evolve tolerance to many stresses that act externally (high salinity or high pressure, for example) and resistance to some antibiotics relies on secreting enzymes that destroy the drug molecules before they can enter the cell (this is how many bacteria become resistant to penicillin). It may well be the case that drugs designed to target the LPS “bricklayer”are hard to evolve resistance to, but if this is true it’s more likely to be because any mutations that make this crucial protein less susceptible to the drug also stop it functioning properly and reduce bacterial viability, not simply because they act externally. (The use of the word “develop” rather that “evolve” adds another element of confusion, but that’s a blog for another day).

Most of the news articles on this work ended with an eminently sensible quote from Prof. Mark Fielder of Kingston University. While recognising that the protein structure is an important step forward in finding targets for novel antibiotics, he cautioned against suggesting that resistance would be impossible. If studying evolution –particularly the evolution of antibiotic resistance –has taught us anything, it’s that we should never say “never.”

Headlines, impact and trust

So why the sensationalist headlines? I’m not going to bash the journalists here, they were going on what they were given by a university press office and in most cases they did temper their initial excitement with Prof. Fielder’s down-to-Earth comments. I think the real problem here is a growing pressure in academia to sell every research article as a high-impact scientific breakthrough that that will change the world as we know it. Doing this, we are told, will boost our CVs, our altmetrics and ultimately that crucial but rather nebulous measure of our worth and fundability known as “impact.” And I worry that this can cause us to misrepresent the incremental process of science to the media and therefore the wider public. Not only are we setting ourselves up for failure if we expect our every result to have maximum impact and so ratchet up the expectations of funding bodies and tenure panels, but if we can’t follow through on media-friendly announcements of evolution-proof antibiotics, cures for untreatable diseases or paradigm-shifting technology, we risk crying wolf. While a majority (just!) of British adults say they trust academic scientists, few think the media is a trustworthy source of information about science. Thus by presenting our work as headline-grabbing soundbites, we may sacrifice its credibility in the eyes of the general public, whose taxes fund our research and whose trust is vital to the continued development of science and technology in the academic sector.

Image: Noticias, Wikimedia Commons



Posted in antibioticsHypeScience