Many a slip… between actual science and Press Release clip?

To whomever reading this first post of a shiny, brand new year, A Happy New Year To You. May the year ahead bring you joy, peace and accomplishments galore.

Within the past week from today, two notes related to science communication – from two people I follow on Twitter and elsewhere – came to my attention. And the funny thing is, the two notes (I say ‘notes’ for want of a better descriptor) are related to each other. One note was from Hilda Bastian, noted Blogger and Science Communicator currently associated with PubMed and PubMed Commons; the other was from Jonathan Eisen, another noted Blogger and Professor of Evolutionary Biology at University of California at Davis. Interestingly, both the notes happened to talk about an issue that has been much debated in various fora lately, an apparent disconnect between scientific research and journalistic communication of scientific research.

Allow me to explain.

On November 21, Bastian wrote a note at PubMed Commons, commenting on a paper that was e-published several days earlier in PNAS, containing research performed at the University of California at San Diego (UCSD). Working with Triclosan, an antimicrobial additive present in various popular consumer products of daily care (including soap, hand sanitizers, and cosmetics), Yueh et al. discovered that in their mouse model, this chemical substance was capable of stimulating liver cell growth and division (‘proliferation’, leading to a rapid increase in numbers) to the point of causing fibrosis and oxidative (related to biochemical reactions that produce oxygen free radicals capable of causing tissue damage if not modulated) stress. In mice with chemically induced tumors, Triclosan appeared to hasten the process of liver tumor formation. The authors titled the article as “The commonly used antimicrobial additive Triclosan is a liver tumor promoter” (DOI link); highlighting this paper, the UCSD Press Department brought out a public release titled “The dirty side of soap“, with the subhead declaring “Triclosan, a common antimicrobial in personal hygiene products, causes liver fibrosis and cancer in mice“.

Bastian’s critique took issue with the fact that [I quote] “The title and abstract of this article focuses on the positive finding in tumor promotion, without emphasizing that the findings were negative on causation, in a way that is clearly accessible for non-specialist readers.” [End quote] She also voiced her serious concern about the UCSD Press Release, whose title she criticized – justly in my opinion – as misleading and alarmist; she further commented on this on her blog (then-hosted at the Scientific American). Not only in the title, the press release stated definitively, “The study, published Nov. 17 by Proceedings of the National Academy of Sciences, shows that triclosan causes liver fibrosis and cancer in laboratory mice through molecular mechanisms that are also relevant in humans.” (Emphasis mine. I, too, commented on this at PubMed Commons a few days ago when I came across the PR and Bastian’s comment while searching for information on Triclosan use.)

In her comment, Bastian provided example from a 2010 study on Triclosan safety, which failed to find a causal link with cancer in any species except liver cancer in mice, and raised a few most germane points focusing on the relative lack of information about the outcomes of the oncogenicity studies with Triclosan. She lamented the fact that [I quote] “Further areas of uncertainty relate to the experiments here [N.B. meaning, in the paper as published]. The article does not report sufficient data and methodological information to enable adequate assessment of the level of uncertainty associated with the experiments…” [End quote] and went on to ask for some key pieces of data that would aid a proper evaluation.

In my reading of the paper, what struck me immediately as odd for a claim of this magnitude was that their sample sizes were rather small (as small as 6 mice in each group in one experiment); the exposure route for Triclosan was either oral, or a direct injection into the peritoneal cavity at a high amount – neither of which is likely to apply directly to humans, whose Triclosan exposure is mostly restricted to skin surfaces at low concentrations. (To provide a context, liquid and bar soaps contain 0.1-1%; the hospital-grade liquid soap used in my lab has 0.5%; current recommendations for removing persistent colonies of various antibiotic resistant bacteria from the skin and nasal cavity of critically ill patients include showering with 2% Triclosan.) I also would have liked to see, in support of the claim in the paper, some epidemiological data regarding the dosage effect, the exposure dose at which Triclosan is likely to cause the hepatocellular effects. However, even the lead author, quoted in the PR, didn’t seem inclined to emphasize any of these points.

To clarify, it is not that the widespread use of Triclosan in all manners of consumer products is without controversy, especially since its intended efficacy may at best be questionable. In a consumer bulletin dated December 2013, the US FDA has indicated that it is re-evaluating the evidence around Triclosan safety in view of the consumer concerns. In contrast, for Triclosan present in certain toothpastes, the evidence appears favorable. In a Dec 2013 report, a meta-analysis by the Cochrane Collaboration concluded:

There was moderate-quality evidence showing that toothpastes containing triclosan/copolymer, in addition to fluoride, reduced plaque, gingival inflammation and gingival bleeding when compared with fluoride toothpastes without triclosan/copolymer. These reductions may or may not be clinically important, and are evident regardless of initial plaque and gingivitis levels, or whether a baseline oral prophylaxis had taken place or not. High-quality evidence showed that triclosan/copolymer toothpastes lead to a small reduction in coronal caries. There was weaker evidence to show that triclosan/copolymer toothpastes may have reduced root caries and calculus, but insufficient evidence to show whether or not they prevented periodontitis. There do not appear to be any serious safety concerns regarding the use of triclosan/copolymer toothpastes in studies up to three years in duration.

But the science is not what Hilda Bastian is objecting to. What is reprehensible is the needless sensationalism and fear-mongering in journalistic communication of scientific results, achieved by sacrificing scientific accuracy and over-hyping incomplete parts of results – thereby misleading the general public and guiding their understanding of the research into a specific direction. I, echoing Bastian, see this as a failure of science communication, and an example of irresponsible journalism.

To gauge the impact of the PR, I searched Le Google with the keywords “triclosan liver caner concerns”, and indeed, within a couple of days of the article and the PR coming out, at least 10 different major news outlets across the US and the UK, as well as various New Age/organic/wellness websites, some law firms, and the NHS UK blog, already picked up and repeated the same lines from the PR – and only a few of them (including the NHS UK blog and Bastian’s post at SciAm) actually offered any kind of reasoned scientific analysis of the data and tempered evaluation. The NHS UK blog, NHS Choices, concludes: “This small mouse study raises the prospect that triclosan may have tumour promoting-properties that could be relevant to humans but, on its own, does not provide any conclusive evidence that it does.”

Primed thusly, I found it no surprise at all when yesterday (January 1), Jonathan Eisen, wrote a post in his blog (The Tree of Life) excoriating what he termed as a ‘shameful overselling of the microbiome’ in a Press Release (quelle surprise, non?) on by none other than the venerable American Society for Microbiology (ASM), the premier professional body in our field.

Discussing the study – wrote Eisen in his blog post – the PR was doing a good enough job to describe the researcher’s work with the microbiome (the sum total of the community of microbial residents in the gut) present in mouse models of lupus, an autoimmune disorder.

The researchers characterized the microbiome is mice with and without the lupus model disease and also compared over time and between sexes. And they found some interesting correlates of microbial patterns that are found during flare ups of lupus for example and also in drug induced worsening of symptoms.

But thereafter the author of the PR (or perhaps the editor) possibly succumbed to the inevitable lure of sensationalization to attract eyeballs. While the researchers (continues Eisen)…

… never showed ANY causal connection between any of the mcirobes (sic) and the lupus like disease. And the (sic) never showed ANY benefit of treating the lupus-like symptoms in the mice. Yet amazingly they go over board in making such claims including

No – the study did not suggest that at all. The authors suggested that, yes. And the study is consistent with that. But it is also consistent with altering gut bacteria having NO EFFECT on lupus. So this title is simply deceptive.

These are just two examples in the recent past. But if this kind of inaccurate and sloppy science communication – especially from people who should really know better – becomes a trend, it would indeed be worrisome for the field, and for science and the public understanding of scientific research in particular.


  1. Paige Brown Jarreau

    January 3, 2015 at 1:55 am

    Thanks for highlighting these examples. Kausik. I hadn’t followed these, even though I did see a tweet about Eisen’s post on the ‘shameful overselling of the microbiome’. The overselling of scientific results via intermediary PR materials is, unfortunately, a common story line. I think it is long overdue that we re-think the role of press releases in the communication of science, as PR materials are increasingly becoming direct reading materials for the public (see Science Daily!) and copy-pasted by too many overworked and under-experienced journalists. I think the negative impact of hyped science PR materials is much less of a concern when these studies are covered by people like Bastian and Eisen – if only journalists would take some of their news judgements from discerning and scientifically experienced bloggers!

  2. Over the weekend, three other informative and interesting essays from well-known scientist-bloggers came out, commenting on the same issue of a discrepancy between the actual science and press releases – in relation to the recent Science paper on cancer that appears to attribute the risks for a majority of cancers to ‘bad luck’. I highly recommended reading all of them.
    1. Critique of the paper from a statistical perspective – by Bob O’hara and GrrlScientist at The Guardian
    2. Critique of the paper with clarifications on the points made in the Guardian essay – by PZ Myers at Pharyngula
    3. Comprehensive critique of the paper and synthesis of the other concerns – by David Gorski at Science Based Medicine

Leave a Reply

%d bloggers like this: