24 February 2016

So you believe what the media say?

RedHerring

Sensational reporting of medical news poses a danger to public health literacy and behaviour


 

Sensational reporting of medical news poses a danger to public health literacy and behaviour.

“If it bleeds, it leads,” is the mantra of the news editor. Drama, conflict, adversity, or triumph over it – these are the stories that make the front page of the lay media.

And reporting of the biomedical news just as sensational, according to a US cancer specialist who follows this closely.

“There were news segments last year on how to avoid contracting Ebola virus that played on people’s fear. But even at the peak of the Ebola outbreak, that risk was orders of magnitude lower than the risk of getting in a car accident on the way home,” Dr Vinay Prasad, Assistant Professor of Medicine at the the Knight Cancer Institute, Oregon, told The Medical Republic.

The pen may indeed be mightier than the sword at stirring acts of mass foolishness, particularly when the language gets emotive. Professor Prasad gives an example from Kansas, where in 2013 there was a flurry of news stories that described streptococcal infection as a “flesh-eating bacteria”. Rates of testing for the infection doubled even though there was no change in infection rates.

As well as killer diseases, the media equally loves a “miracle cure”.

In a recent paper in JAMA oncology Professor Prasad counted the overuse of superlatives such as “breakthrough” “game changer” and “revolutionary” in relation to new cancer drugs. Worryingly, the use of such language was equally attributed to approved and non-approved drugs, and more than one in 10 drugs mentioned by the media in these terms hadn’t even been tested in humans.

“I have cancer patients coming to my clinic all the time with some breakthrough they read about in the news,” said Professor Prasad. He doesn’t enjoy having to explain the actual likelihood of hyped-up cures becoming available. “I feel like I’m taking take the wind out of their sails. They get very disappointed.”

And then there is the impassioned victim. “Brain on fire” is the headline for a Sydney Morning Herald Good Weekend article describing the “hellish headaches” of people suffering hypersensitivity to electromagnetism from Wi-Fi and mobile phones. The personal anecdotes are moving, but to date there’s not a scrap of clinical evidence for this phenomemon.

But who wouldn’t read a David and Goliath battle of victims not heeded by the big establishment, or a Lorenzo’s Oil narrative of patients coming up with their own unconventional cures?

A LONG BOW TO DRAW

One of the reasons the public gets a skewed picture of biomedical research, is that journalists don’t always cover most watertight papers. The lay media are much more likely to report about observational studies than the big journals are, according to an analysis by Professor Prasad in PLoS ONE.

In comparing the top US papers with the highest-impact general medicine journals, it turned out that the newspapers published stories on cross-sectional studies twice as often as the journals published original research on those types of studies. For observational studies, the ratio was 50% more newspaper articles than journal coverage.

And it’s no wonder, given that observational studies usually generate more surprising results and exciting headlines. Which would you rather read, “Coffee may ward off breast cancer”, or “New Cancer Drug Ibrutinib Outperforms Chemo For Some Patients”?

But observational studies don’t necessarily prove cause and effect and very often don’t warrant the dramatic spin they get in the media, says Professor Prasad. Scientific research doesn’t always provide clear-cut findings and research results are often conflicting.

Take for example recent media coverage of a JAMA Pediatrics study showing that there was an increased risk linked to maternal use of antidepressants. Within two weeks, Translational Psychiatry came back with another study demonstrating no such association. The general media focussed exclusively on the first of these studies, however.

Bad news is good news when it comes to the editors and journalists of the Fourth Estate. As are genes that tell us what to do. “Can’t get out of bed? Blame your genes”. “Having trouble giving up smoking? Blame your genes”. “Mean drunk? Blame your genes”. The way such studies are presented in the mainstream media, it’s a surprise we have any free will at all. But genetic association studies are, just that – associative – and there are often countless steps between gene expression and behaviour that the public will never hear about.

Not all the blame should go to overzealous journalism, however. Researchers themselves aren’t shy about playing up the implications of correlations they’ve observed. Professor Prasad has found that half of observational studies published in high-impact journals make suggestions about clinical practice that are entirely speculative. As soon as a new biomarker or association for disease is found, researchers talk of “routine screening”. But Professor Prasad says this risks promoting expensive follies for healthcare until prospective studies fill in the gaps.

Sometime the causal evidence does exist, but it’s from experiments in animals or in cells in a dish. Not that that doesn’t stop the evocative headlines. “There’s so much reporting of something that worked in a mouse to cure cancer or whatever,” says Professor Prasad. “I don’t for the life of me understand the public relevance of that, because we know from people that work in drug discovery that for a 1000 drugs that work in mice, we’re lucky if one makes it to the clinic.”

KEEPING IT IN CHECK

And the media offices of research institutions also need to shoulder some of the blame. A 2012 analysis in the BMJ showed that if a press release didn’t mention absolute risk or adverse events from a research study, then the media stories were five times less likely to discuss them.

The hype may, sadly, be necessary in the competitive climate of attracting research grants, says Professor Prasad. It’s well demonstrated that broad media coverage of a paper will greatly increase the likelihood of its being cited academically.

“But it’s also human nature to talk up the significance of one’s research,” says Professor Prasad. “Human beings want to feel like their life’s work has some importance, when in reality, most of our work is modest and incremental.”