'Fake' Media (And the Public) Needs a Time Out

Related articles

As a pediatrician, I always advise don’t be fooled by the cuteness. Urging parents to stay strong --especially in those vulnerable moments. See the big picture. Follow through with consequences for bad behavior. Easy to do in an office visit, but hard to achieve day in day out for twenty years— even with the best of intentions.

Sadly, the concept of withholding a lollipop for bad behavior isn’t necessarily transferable to media organizations when they under, over or inadequately inform the public with respect to science and health claims. Even though heightened public anxiety and co-opting of physician office visits to debunk medical myths perpetuated by such imprecise information are very tangible adverse effects, somehow the messenger continues to go unscathed.

Instant gratification of ratings and the nature of our twenty-four hour news cycle creates a pressure to produce quickly, sometimes with insufficient research. When the bottom line doesn't get negatively impacted, no course corrections or significant changes take place. If the train has left the station, then educating the masses is the tool left in my arsenal to shift the paradigm. Or, at least try. 

Newly published research by the Public Library of Science (PLoS ONE) ostensibly investigated what stories are being sold—oops, I mean told—in this sphere. The researchers set out to explore by tangible metrics “the replication validity of biomedical association studies covered by newspapers.” Reproducibility of findings is the hallmark of scientific integrity. Association studies are routinely considered flawed as they often don’t stand up to this type of rigor. So, they especially, should be replicated while consistently showing a relatively high change in risk — as did the multiple associations with smoking and health.

The scientists set out to determine the scope of newspaper coverage for studies that associated a risk factor with a disease in specified realms: psychiatry, neurology and four somatic diseases. These were given  a lifestyle (e.g. modifiable risk factor like smoking) or non-lifestyle (e.g. genetic etiology) designation. (1) Performing an exhaustive review of thousands of primary studies along with subsequent ones to ensure the original work withstood the passage of time, they utilized the Dow Jones Factiva newspaper article database to determine what actually got published. Television reporting was not included though the authors suggest written work likely influences it to some extent. With the help of meta-analysis, they interpreted the following after systematic reviews of large number of studies:

  • “Newspapers never covered initial studies reporting null findings and rarely reported subsequent null observations”— a null finding reflects one that lacks statistical significance or has no consequence 
  • Psychiatric studies were confirmed less often than neurological or somatic ones
  • For example, “This is correlated to an even larger coverage of initial studies in psychiatry. Whereas 234 newspaper articles covered 35 initial studies that were later disconfirmed, only 4 press articles covered a subsequent null finding and mentioned refutation of an initial claim”
  • “Newspapers preferentially reported lifestyle association studies linking a pathology to a risk factor on which each reader can act. Non-lifestyle studies related to brain imaging, genetic factor or other inescapable risk factors were less often echoed.”
  • Lifestyle ones get equal initial and subsequent study coverage, non-lifestyle associations get more initial coverage than subsequent

Basically, if a finding lacks commercial appeal like a null finding, then it won’t be covered. Just because a study results in determining an association is of no consequence does not mean it is of less value than one that finds a definitive--or positive-- correlation. Ambiguity in scientific discovery— though a reality of the adventure—is viewed today as a liability not coveted enough to report. In actuality, this could genuinely be an asset. The major conclusion they drew from their analysis was the fact that “journalists preferentially cover initial findings although they are often contradicted by meta-analyses and rarely inform the public when they are disconfirmed.” (2)

For a shift to greater and accurate understanding, the public needs to assume some level of responsibility by refusing to absorb such messaging without question. Collectively, let’s get empowered and recognize that initial studies are prioritized in the lay press because they tend to be riddled with the most compelling “what ifs?.” (3) Review multiple sources. Appreciate that in the world of medical discovery insight into conditions, treatments and knowledge develops and accumulates over time. Recommendations can be dynamic, not static forever assertions. Conclusions can be ultimately disproven especially when early in the discovery process.

News "curation" as the du jour mode of selecting content—performed by all media forms— where some studies are presented (and others not) merely serves to impose bias. By picking and choosing, a false perception is disseminated that one is of greater import than another, when shiny object syndrome may be truly to blame. If there were no readers or viewers  to capture amidst today’s endless competition, then more measured and equivocal—but precise—depictions of progress would get bolstered. 

Since I’ve personally witnessed toddlers out-negotiate even the most experienced, seasoned CEO-level executive, maybe there needs to be a new approach to battling this form of infection that is misinformation or incomplete knowledge. When a young child is in control of the grown-ups in the room, the immediate impact can be humorous. But, it is the groundwork being put down that lays the foundation for a poorly mannered adult who is not well-adjusted, lacks coping skills and the capacity to be independent. Not any parent's dream for their child.

If collectively we can take a time out and reboot, then we will be reminded of the ultimate goals of invaluable and informative science communication. Good information encourages people to go to the doctor sooner and be treated optimally without delay. Bad information can prompt delay and inevitably poorer outcome. Good information can spur another scientist somewhere in the world to collaborate and speed societal innovation. Bad information can stop forward momentum of a particular field. Good information can draw a respectful and trusting crowd to journalistic endeavors. Bad information can tarnish reputations and diminish the power to maintain a meaningful voice that educates the masses.

With scientists assuming a greater role in articulating the appropriate context and level of enthusiasm for their work, journalists reminding the audience of the nascent or mature level of the research stage, and the public appreciating the nuances of such contributions, there is no limit to what we can achieve in terms of improved health and well-being.


(1) (2) (3) Dumas-Mallet, Estelle; Smith, Andy; Boraud, Thomas; Gonon, Francois. Poor Replication Validity of Biomedical Association Studies Reported by Newspapers. PLoS ONE 12(2):e0172650. doi:10.1371/journal.pone.0172650