You have to wonder why the breakfast tea doesn’t get the same level of attention. Or perhaps whether in France, the humble croissant is elevated to the same status. Or maybe the banana could soon be the star of another media show. But unfortunately it is coffee that headlines tomorrow’s fish and chips papers.
“Drinking three or four cups of coffee a day could have benefits for your health”. As we have seen previously, this kind of headline bears the hallmarks of a media health report:
1) repackaging of common information requiring little or no specialist examination;
2) use of a modal auxiliary verb (could) to conveniently justify or disclaim an
attention-grabbing headline – which, by the way, is point number three.
The health reports in the media also incorporate:
4) a statistically small group of trial participants, whose results are then blown up in proportion as if to be representative of the 7 billion people on the planet.
5) Assumptions. A media report about health could simply include assumptions.
Why dwell on coffee? For starters, it is a commonly consumed drink and so any meaningful research would potentially have bearings on millions of people. It is common media practice to focus on common food and activities because of the relevance to daily life.
But if you examine this carefully, why not tea? Why not write about tea? While conspiracy theories may be slightly far fetched, it is possible that – unless it is a speciality tea – coffees cost more, and any potential health benefits would lead people to spend more, hence generating more for the economy in the forms of tax. Perhaps this is why media writers don’t waste too much ink on researching the potential life-saving benefits of bananas, even though they are widely consumed. The research isn’t going to drive people to buy bananas in bulk, and even so, the extra revenue generated from a low priced item isn’t going to raise much extra tax.
Are there any notable similarities or differences in style across different countries? One wonders whether Parisian newspapers, on a regular basis, churn out headlines such as:
“Eating two or more croissants a day could reduce your chances of heart disease.”
“Pan aux raisins linked with dementia”.
The research done was an umbrella review to potentially examine whether further research should be undertaken into researching the effects of coffee and its role in preventing liver cancer. An umbrella review meant that no actual research was undertaken, but that existing research was examined and analysed to glean insights.
The problem with umbrella reviews is that they are very generalised, no actual research is done, and they are only brief analyses of existing research. This means that first of all, an umbrella review could arrive at a particular conclusion, but in no way should that be taken as the final conclusion.
In fact, the findings of an umbrella review are only the preliminary to more detailed investigation. If an umbrella review suggested that drinking coffee could prevent cancer, then what it is saying is more research needs to be undertaken, and the media needs to be ethically responsible by not reporting “Coffee prevents Cancer”, because there are people that look at newspapers and television as the source of their information and assume just because it has been released in the public domain, it is truth. Who could conceive that newspapers spend time and resources to publish trivial information and that television is pure rubbish?
The second problem with umbrella reviews is that the outcomes are only as good as the original sources. If someone gave you a set of grainy photos, then asked you to make a collage with them, then your collage is going to be as good as the grainy photos will allow. If the original sources were not thorough or exact in their investigation, are any subsequent findings based on these merely just a waste of time?
The third issue with umbrella reviews is that under closer scrutiny, the overall picture is distorted by over focussing on small statistical variances, or sometimes minute errors are magnified and lead one down the wrong path.
If you took a picture on your phone and then blew it up to the size of a mural covering the side of your house, the picture becomes very dotty. You might see big patchy squares. But if you started looking for that big patchy square from the image in your phone… one has to wonder what the purpose of that is.
The fourth is that because umbrella reviews are a prelude to a more thorough investigation, their end results are slightly skewed from the outset. If an umbrella review is bound to provide a few avenues for later time-consuming research then it is fundamentally biased into having to provide one in the first place. Why, in that case, have such reviews in the first place? Some may point out that the flaw in the system is that umbrella reviews are relied on by those in academia and research to warrant the continued longevity of their positions. In other words, if researchers had nothing to research, they might be out of a job, so they best find something to stick their noses in.
Have you ever read the London newspaper Metro and come across some research news such as:
“Going to bed angry can wreck your sleep” (25 Sept 2017)
It is the sort of headline that makes you think “Why bother doing the research in the first place?”
It is likely that you have read a media report of an umbrella review.
What were the findings of the original coffee review?
Drinking coffee was consistently linked with a lower risk of death from all causes and from heart disease. The largest reduction in relative risk of premature death was seen in people consuming three cups a day, compared with non-coffee drinkers.
Now, when an umbrella review mentions drinking coffee is linked with a lower risk of death, it is important to be clear about what it specifically means. And what it is stating is that those who had a lower risk of death all happened to drink coffee. It might have nothing to do with the coffee itself. It might have been that they took a break to slow down a fast-paced lifestyle, and the taking of a break gave them a lower risk of death. By that logic of association, tea could also be linked with a lower risk of death.
Coffee was also associated with a lower risk of several cancers, including prostate, endometrial, skin and liver cancer, as well as type-2 diabetes, gallstones and gout, the researchers said. The greatest benefit was seen for liver conditions such as cirrhosis of the liver.
Again, to be clear, the above link means that those who were at lower risk of those cancers happened to drink coffee. But it is not necessarily stating the coffee had anything to do with it.
And coffee is such a commonly consumed drink, that it is easy to use it to draw links to anything.
If people who died from car accidents happened to drink coffee, an umbrella review might state that drinking coffee is linked with higher incidences of car accidents.
The findings can be summarised by a health analyst:
“Does coffee prevent chronic disease and reduce mortality? We simply do not know. Should doctors recommend drinking coffee to prevent disease? Should people start drinking coffee for health reasons? The answer to both questions is ‘no’.”
We should perhaps add a further third question: Did the umbrella review produce any actionable findings, and should it have been undertaken in the first place?