hospital drip

Journalism in absentia: Reporting medical research in the mainstream media

A version of this piece ran in the first edition of the Michael Smith-edited reboot of Village Magazine in December 2008. 

A common rumour concerning the contemporary print media is that it is heavily reliant on the public relations industry to supply it with story ideas. The mass austerity drives of newspapers all over the world from the 1990’s onwards; brought on by declining sales and advertising revenue, led to massive staff cutbacks, even as newspapers increased their pagination. In 2006, a group of researchers at Cardiff University set out to discover if the rumours of a newspaper industry in hock to its public relations cousin were true. Their findings, originally published in 2006 but brought to wide public attention in Nick Davies’s 2008 book Flat Earth News, showed that they were: of the 2,207 stories they analysed, they found that less than half of them were entirely independent of traceable PR.

The reliance of the media on press releases and press officers to give them both story ideas, and, too often, the wording of a story too is explained in one way by the the increased pressures on ever smaller numbers of journalists to fill an ever larger news hole, in less time than ever before. A second contributing factor is the increased popularity of lifestyle supplements. These are easy and cheap to produce, consisting as they do of pages of thinly-veiled advertorial and news of celebrity exploits, both of which categories come pre-packaged in press releases from companies and agents eager to see their brand promoted.

And what harm, perhaps, if they do: if their advertorial is indeed thinly-veiled, then flicking through the Sunday Times Style magazine is as potentially damaging as flicking through an Argos catalogue. Similarly, despite occasional attempts to invoke lofty conceptions of the right to privacy, much celebrity interaction with the media is a carefully contrived game of cat and mouse, with both sides fully aware of the rules of the game. There is one area, however, that has been co-opted into the PR-driven lifestyle sections in which reliance on press releases can have far from benign real world effects.

Newspapers love health stories. Most newspapers have at least a weekly health page, many have a health supplement, and the news pages of every newspaper will have at least one ‘Boffins today claimed…’ story every day. These stories are almost always based on research that has just been published in one of the thousands of medical journals now on the market. The health pages and supplements divide their time between reporting on academic medical research and ‘alternative’ health – vitamin pills, homeopathy, crystals, magnetic bracelets and so on. More often than not, no distinction is drawn between these untried, untested and frequently useless alternative approaches and the academic stuff.

The blurring of the lines between quackery and established medicine has become par for the course in recent years. Distrust of Big Pharma, despair at the limited time an overworked GP can devote to each patient, thus precluding a consistently holistic approach to patient health, married to a more widespread disillusionment with an underfunded and inefficient health service have all generated a climate of approval for alternative approaches. All well and good, and if the placebo effect of the ‘whole life’ approach of alternative therapists makes patients feel better then it can only be saluted. However. While on an individual level turning to alternative therapies isn’t terribly problematic, something else is going on in our attitudes to medicine: it’s dangerous, it’s worrying, and it’s the media’s fault.

On 13th September 2008 the front page of the Guardian reported the collapse of a libel case taken against the paper by Matthias Rath. Rath is a qualified doctor and the head of a nutritional supplement empire, which, it is claimed, is worth millions. He is also behind campaigns in the US, UK, Germany, the Netherlands, Spain, France, Russia and South Africa to persuade HIV sufferers to stop taking anti-retroviral drugs and take his nutritional supplements instead. The case was sparked by a series of articles in January and February of 2008 in which Guardian columnist Ben Goldacre condemned Rath and his practices, saying ‘He aggressively sells his message to Aids victims in South Africa that Rath vitamin pills are better than medication.’ There is a good deal of evidence to suggest that Rath’s campaign in South Africa caused people there to stop taking their medication, or to never start, leading to hundreds, if not thousands of preventable deaths (Guardian, 13th September). Commenting on the collapse of the libel case in an editorial, Alan Rusbridger said ‘In the west, wacky claims often seem harmless enough. But the Rath case provides a terrible illustration of the potential consequences of treating evidence with contempt.’

The media has no time for evidence. Copy must be turned around quickly, and space is limited. With limited space, only headline claims are included and qualifications or caveats that might undermine their apparent certainty left out. Scientists have lamented this situation for years; it is the basis of what is usually characterised in the public understanding of science literature as a called the ‘clash of civilisations’ between scientists and journalists. Both have very different aims in communicating information: the latter must make their copy short, readable and interesting, the former is dedicated to careful explication of findings, in context, at length. There is a relatively simple solution to this problem, one suggested by Gary Schwitzer. Rather than running health stories as news in briefs, with the aim, as he says, to prove the paper’s ‘breadth of coverage’ (flick through any copy of any newspaper and you’ll find a report on medical research in 200 words or less), ‘If a news organization cannot give sufficient space or time to a story to cover the necessary issues, then it would be better if the story were not published or broadcast at all.’

Imagine for a moment a 200-word article dedicated to the findings of the Mahon tribunal. That’s the kind of feat of contraction performed every time a piece of academic research is reported in this way. Why, then, do newspapers persist in reporting it like this? There is one principal reason: a complete failure to understand that not every piece of medical research is the same. There are single case studies – where a patient presenting with a particular illness has that illness and her experience of it written up for publication. The findings of the study are anecdotal, as meaningful as someone down the pub telling you about their bad back. The case study may come in useful to point to a future research direction, but that is all. There are studies with small samples – the smaller the sample, the less robust your study. There are studies that have controls, and those that don’t. Each study’s author will have a rationale for their choice of study design, but the findings of the research can only be judged when the study design has been scrutinised. For example: the Sun reported in July that the more oily fish one ate the less chance one had of having a stroke. The study in question was not comparative – i.e. it didn’t look at rates of oily fish consumption in those who did have and those who didn’t have strokes. Had the journalist who wrote the story looked at the original research (s)he would have found that information right near the beginning of the text.

But the media is not interested in evidence, only in claims, and each claim is only as strong as its proponent’s voice. There is a fundamental lack of understanding that a claim made by a vitamin pill merchant, or a homeopath, or a magnetic bracelet peddler, unsupported by any evidence, or supported by evidence derived from studies with methodological flaws so deep as to render them irrelevant, is not the same thing as a ‘claim’ (and I use the quote marks advisedly) derived from clinical research that has both a strong design and has been through a process of peer review and found acceptable. The failure to grasp this can have terrible consequences: Andrew Wakefield’s claim for a connection between MMR and autism, based on a study with a tiny sample size and a flawed design, was allowed to gain currency because of this fundamental failure of understanding. Wakefield’s paper was published in the Lancet, but going through the process of peer review and being published in a journal does not allow claims unsupported by evidence to be made by its author. Wakefield’s claim did not stand up, could not stand up on the basis of the evidence he had. And yet, but for a very few dissenting voices, the media allowed this claim to fill their pages, and fuelled a debate based on no evidence, with the effect that MMR vaccination rates dropped sharply throughout the late 1990s and the early years of this century. The decline in vaccination rates precipitated a measles epidemic in Ireland in 2000 in which three children died. Had anyone investigated the evidence for Wakefield’s claim when it first emerged, that epidemic, and those deaths, would have been averted.

Confusion about the difference between empirical evidence and unsubstantiated quack claim is partly cultural. Partly it devolves on few journalists coming from science or medical backgrounds and being simply unaware that there are different kinds of evidence out there. This allows for medical research to be treated with as much reverence as any other lifestyle topic. Medical journals are in the habit of sending out press releases about upcoming issues to journalists. Combine a poor or non-existent grasp of the process of clinical research with an overworked journalist on a tight deadline and these press releases translate as one word: ‘godsend’.

A study (unpublished) by this author carried out in July and August 2008 found that 70% of all medical stories in Irish daily newspapers had a press release as their source. Only 32% of these contained any information above and beyond that contained in the press release. In other words, the press releases had been copied and pasted into the news pages with only some minor tweaks to accommodate house style. Across the board, there was very little to suggest that any journalist reporting these stories had given the primary source of their information, the original academic paper, even a cursory once over. Because it comes prepackaged in a press release, health news is cheap – a handy filler for news pages and health supplements that has the bonus of going some way to demonstrating a paper’s intellectual credentials. It doesn’t look like public realtions bumpf (no glossy picture), doesn’t sound like public relations bumpf (no company or celebrity name attached), so it allows a measure of credibility to be attached to whoever publishes it.

Does it matter? It could be argued that journalists trust that the press releases issued by medical journals are reliable. If they do, this trust is unfounded. Medical journals are commercial enterprises (the scientific journal market is estimated to generate several billion pounds in revenue each year) and they issue press releases with the aim of promoting their brand. That they are commercial enterprises does not quid pro quo undermine the quality and value of the research contained in them, and it should be noted that there is a distinction between a journal’s editorial and its commercial arm. Virginia Barbour, who worked for four years as an editor at the Lancet has said: ‘At the Lancet, editors and managers seem to have different aims; the former to promote the free flow of information; the latter to generate revenue.’ It cannot be assumed, then, that a journal press release will not give in to the temptation to sex up research findings in the hope of more media coverage. A 2002 study found that only 23% of press releases issued by the most high profile and well respected medical journals in the world (including the Lancet and the British Medical Journal) contain any information on the limitations of the research they’re flagging – in other words, they often exaggerate the significance of research findings.

Journalism is supposed to be about accuracy, and facts are supposed to be sacred. That ideal has been jettisoned when it comes to health news. Press releases are treated as primary sources, despite their being anything but.
A press release copied and pasted into the Sun and the Star in July asserted that pregnancy can increase the risk of heart attack three-to-four fold. A look at the study this assertion is based on shows, first, that the study did not find this; was, in fact, constructed in such a way that this conclusion was impossible (it was another non-comparative study). Both papers contain the claim that ‘although acute myocardial infarction is rare in pregnancy, it can increase the risk of heart attack three-to-four fold.’ Anyone who has seen more than half an episode of ER should have some idea that myocardial infarction and heart attack are the same thing.

In August, the Irish Independent and the Star both reported on a study into the alleged effects of the contraceptive pill on a woman’s choice of partner. The pill, they claimed, causes women’s preferences in men to change. The major histocompatibility complex (MHC) is a part of the brain which carries a number of genes. The stories in the papers reported that the study, carried out at the University of Liverpool, found that women can ‘sniff out’ the kinds of genes a man has on his MHC, and are more attracted to men with genes different from their own. There are problems with the research itself – while the authors assume women can tell by a man’s smell what kinds of genes he has; that evolution has programmed us to be attracted by the smell of men with genes different from our own; and that this kind of differential gene combination in offspring is adaptive, the evidence for all of this is in fact equivocal, with the consensus falling rather on the side of women not being able to detect from smell what kind of genes a man has (lower animals can, but the part of the nose used for this is, like our appendices, there but not functioning); and also that having different genes in the MHC is not always adaptive. The study had been carried out at the University of Liverpool, and the lead researcher, Craig Roberts, was quoted in a press release as saying: ‘Not only could MHC similarity in couples lead to fertility problems but it could ultimately lead to the breakdown of relationships when women stop using the contraceptive pill, as odour perception plays a significant role in maintaining attraction to partners.’ The press release also said that ‘Disturbing a woman’s instinctive attraction to genetically different men could result in difficulties when trying to conceive, an increased risk of miscarriage and long intervals between pregnancies.’ Those are pretty bold claims, and they were run without any qualification in the newspapers. A quick look at the study shows them to be unsupportable by the research that was actually done. What the study actually found was a change in odour preference in a small number of women after they’d gone on the pill. That’s it. It didn’t look into how their relationships were progressing since they’d been on the pill; it didn’t look into their fertility or lack of it; it certainly didn’t look at the number of miscarriages these women had had; it didn’t look at anything except what kinds of smells women liked before and after they’d gone on the pill.

And yet, nobody interrogated these dramatic claims. The story is still running – the Sunday Times did a two-page feature on it in October. And so women now have to worry that their contraception is going to harm them and their babies, all because journalists are happy to recycle nonsense given to them by public relations officers.

What can be done? Medical research is often complicated, and it’s difficult to get across its meaning and significance in accessible and readable prose. Few journalists have time to read full studies, and are compelled by time restraints to rely on press releases. As mentioned above, there’s a fairly simple way around all of this. To paraphrase Gary Schwitzer, if you can’t do something right, don’t do it at all.

Image: Allan Foster