PUBLIC HEALTH:

SUMMARY: Both scientists and the media are to blame for extreme misrepresentations of ADHD neurobiology in the scientific literature and the lay press.

Attention deficit hyperactivity disorder (ADHD) causes a person to be overly active, insufficiently attentive, and impulsive. It's present in a few percent or more of the population (children and adults), although the exact prevalence is hard to define due to diagnostic difficulties.

What causes ADHD? This issue is also controversial, but may be some combination of genetic, environmental, and social factors.

In order to rigorously identify the causes of ADHD (and thus arrive at effective treatments), an unbiased evaluation must be made of many scientific studies. It's essential for the scientific manuscripts reporting on these studies to fairly report the results.

Francois Gonon (University of Bordeaux, France) and coworkers show that these manuscripts typically hype their findings, and this hype is commonly amplified in the media as well as in the scientific literature. This study highlights the great potential for scientific confusion and social harm that can arise when scientific research is hyped and taken out of context, e.g. as with genomic medicine.

Study procedures.

Gonon and coworkers studied a large set of ADHD-topical research articles. They were obtained through PubMed and ISI Web of Knowledge searches.

Each article was published in an English-language journal, with abstracts, by 2009. Dow Jones Factiva and Google search were used to find a large set of English-language media reports on these articles.

This research should be viewed as qualitative, and is not necessarily representative of the ADHD scientific literature (e.g. because the study of factual omissions only covers one aspect of ADHD genetics). A more detailed description of the scientists' research methods can be found in the original article (open access).

Research article internal inconsistencies are not presented.

Gonon and coworkers found that only 2 of 360 research articles presented conclusions that were clearly inconsistent with the results. Since the scientists' literature review was not systematic, this is not a quantitative estimate of internal inconsistency (although it does suggest that internal inconsistency is infrequent).

However, the media has almost never fairly reported on either of these two articles' conclusions. Ninety-eight percent (60 out of 61) of media reports on these two articles only discussed the articles' conclusion that dopamine levels are reduced in the brain of those with ADHD, despite the results section reporting that the opposite conclusion is still a possibility.

This error is repeated in the scientific literature. Sixty-seven percent (20 out of 30) of research articles citing these two internally-inconsistent results reported on the dopamine conclusion with no further comment, and only one (written by Gonon) mentioned the inconsistency.

Research article factual omissions are not presented.

Gonon and coworkers found that research articles often present definitive statements in the summary that do not match the tenuous data in the results section. The media often plays up on the summaries, i.e. 82% (137 out of 168) of the time.

A representative report was published in the New York Times, stating that a genetic variant is found in a high proportion of people withh ADHD, when in fact the research article cites a prevalence of 23% compared to a 17% prevalence in the general population. This is clearly a low risk.

This error is repeated in the scientific literature. Only 26% (19 out of 74) of research articles citing this "genetic link," and which stated that the results were statistically significant, mentioned that the risk was indeed small.

Review articles were even worse, at a rate of 14% (6 out of 43). In fact, 30% (13 out of 43) stated that the risk was high.

Research article future therapeutic benefits are hyped.

It's natural for a scientist to want readers to think that fantastic scientific advances are on the horizon, as a direct result of his or her research. On the other hand, it's imperative to not overhype, e.g. the constant drumbeat of news stories on "synthetic life" and "alien life" that never match the underlying science.

Ganon and coworkers found that 55% (56 out of 101) of research articles overstated the findings' relevance to human ADHD, e.g. extrapolating findings in mice to humans, and 23% (23 out of 101) unjustifiably hyped therapeutic potential. These 23 articles were more likely to be in a journal of higher impact factor, probably because such journals require a bit of sensationalism.

The media uncritically parroted such hype 83% of the time (52 out of 63 reports). Scientific hype clearly has a corresponding negative impact on news reporting.

Implications.

Scientific hype has negative consequences on the progress of basic science, and consequently on medical progress. As Ganon and coworkers note, as of 2010, there are far more studies on the role of dopamine (1314) rather than histamine (38) on ADHD; although both have a role in attention, it's reasonable to suspect that scientific funding unjustifiably favors dopamine research.

This scientifically-narrow worldview has negative social consequences, e.g. those resulting from reports on "definitive findings" that ADHD can be traced to biological rather than social or environmental causes; medical rather than psychological therapy or environmental remediation may be unjustly favored. Furthermore, if the public is erroneously led to believe that a "cure is on the way," they may question scientists as a group when the "cure" never materializes.

Ganon and coworkers propose three remedies. Specifically, scientists need to be more responsible when reporting their research, scientific journal editors need to weed out hype, and neuroscience research funding needs to be independent of therapeutic applications.

I agree with all three recommendations. However, since scientists have a natural tendency to hype themselves, scientific journals are often in constant competition with one another, and "groundbreaking science" is more likely to be funded, I personally feel that none of these excellent recommendations stands any realistic chance of implementation.

NOTE: The scientists' research was funded by the Centre National de la Recherche Scientifique (CNRS), the University of Bordeaux, the Institut des Sciences de la Communication du CNRS, the Agence Nationale de la Recherche, and La Region Aquitaine.

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) DOI: 10.1371/journal.pone.0014618