Scientists’ responsibilities in the public understanding of science – The case of neuroscience and ADHD
Most of us rely on the media to learn about scientific innovations and breakthroughs. Where health is concerned, the media are often accused of looking for sensational stories, presenting incomplete, unbalanced, uncritical or misleading information. Yet, scientific journals and experts constitute, by and large, the main source of information for science journalists. Would it not be possible that misleading conclusions in the media stem, at least in part, from misrepresentation of data in scientific papers?
This is what a recent article examining the neuroscience literature on Attention Deficit Hyperactivity Disorder (ADHD) suggests (Gonon, Bezard and Boraud, 2010). The authors examined how neuroscience data were presented in scientific papers and media articles. Interestingly, they identified three types of misrepresentation in the scientific literature on ADHD.
Internal inconsistencies between results and the claimed conclusions
Fortunately, inconsistencies between results and claimed conclusions in the neuroscience papers examined were rare. Out of 360 articles, only two studies presented such discrepancies between results and conclusions. What is challenging however is that these two studies were significantly echoed in the media. Respectively 40 and 20 articles cited them and in almost 100% of the cases, the media reported on the claimed conclusion, ignoring the results mitigating the claims. Even more puzzling was the fact that the scientific literature was no more critical. Out of 30 scientific articles citing the above mentioned studies, 20 cited the conclusions without further comments.
Fact omission occurs when researchers put in the abstract of their paper a fixed conclusion while they provide raw data strongly limiting its relevance in another section of the paper. Reviewing scientific articles that discussed the link between ADHD and the DRD4 gene, Gonon et al showed that the omission of relevant facts limiting the impact of the claim occurred in 86% of review articles (n = 52). This misrepresentation was similarly echoed in 82% of media articles (n = 170) reporting on ADHD and on the DRD4 gene.
Extrapolating basic and pre-clinical findings to new therapeutic prospects
According to the authors, unjustified overstatements in the conclusion of studies with animal models are frequent. The authors rated ADHD-related studies of mouse brain as “(…) overstated when the link between ADHD and the studied mice only relied on their behavioural similarities with ADHD symptoms and when the conclusion stated that findings provide novel insights into neurobiology of ADHD”. They found that 55% of the scientific articles (n = 101) surveyed were overstated and 23% extrapolated to new therapeutic prospects. But also important is the fact that the 23% papers previously mentioned were published in journals with a high impact factor, thus in journals likely to be consulted by science journalists. And indeed, 83% of media articles (n = 63) faithfully reported these misrepresentations. The authors note that this is all the more troublesome when one acknowledges the fact that “In the biomedical literature, scientific articles, which are echoed in the media, are more likely to be cited in subsequent scientific articles”. Such misrepresentations are thus likely to spread rapidly and irreversibly.
The social consequences of misrepresentations
According to the authors, misrepresentations of neuroscience data bear social consequences. In the case of ADHD, misrepresentations often consist in emphasizing a link between certain genes and ADHD while downplaying the data limiting the role of genetics in this disorder. Not only do they corrupt the message the general public receives, but they also emphasize the idea that ADHD is primarily caused by biological factors, ignoring other possible avenues such as a combination of biological, social or environmental factors. This contributes to spread the idea that solutions to mental disorders lie mainly in drug treatments rather than prevention and psycho-social interventions. The third type of misrepresentation (extrapolating basic findings to new therapeutic prospects) also emphasizes this idea that ADHD is primarily caused by biological factors. As the authors state “(…) animal models are not suitable to develop psycho-social interventions.” Moreover, it feeds families and patients with illusory short-term hope, rarely acknowledging that, should a concept developed in an animal study be validated, still 10 to 15 years will be needed to develop new treatments.
It is likely that neuroscientists are not the only ones in the academic community who overstate their conclusions and downplay or make less visible the limits and relevance of their findings. Scientists are increasingly encouraged to promote their work to the media. In order to do so, they must have something to offer that is of significant interest to the public (www.phcris.org.au). In this context, it can be tempting to simplify the message or embellish the conclusions. Editors of scientific publications also bear responsibilities for the papers they accept to publish. “If they collectively reject sensationalism and clearly condemn data misrepresentation, we may expect rapid improvement” (Gonon, Bezard and Boraud, 2010). In short, it is the scientific community’s responsibility to acknowledge the risks of data misrepresentations and reflect on their potential social and public health consequences.
Should you want to learn more about the relationship between the scientific community and the media, you may have a look at our dossier Medical innovation and the media or our communication workshop entitled How to communicate the stakes of medical innovations to the public? (In French only).
|Author :||Myriam Hivon, Ph.D.|
Gonon F, Bezard E, Boraud T (2011) Misrepresentation of Neuroscience Data Might Give Rise to Misleading Conclusions in the Media: The Case of Attention Deficit Hyperactivity Disorder. PLoS ONE 6(1): e14618. doi:10.1371/journal.pone.0014618
Primary Health Care research & Information Service. (2010). Engaging the Media: Promoting your work to the media. Fact Sheet. www.phcris.org.au.