Misinformation, coffee, and statistics: How a viral news story highlights the spreading of poor information
by Brad Hines 8-29-17, 3:45 pm
You may have seen this news on coffee that's gone viral since a few days ago, about research from the European Society of Cardiology that drinking 4 cups of coffee a day will significantly increase longevity. While I love coffee, am drinking some right now, and otherwise have no opinion on how true the research is beyond what the study actually does claim, I want to step back to note how it's worrisome how studies like this get interpreted, reported, and in turn– altered and re-interpreted/reported incorrectly ad infinitum.
| "I would advise drinking plenty of coffee. It could be good for your heart. I think it's a good idea to have about four cups a day." |
You can search this topic of the week and see that one publication writing news about the researchers' finding is already seriously using the title "Drinking Coffee In Excess Is Actually Great For Your Health". To say nothing of that a person should read the research itself and understand the inherent biases in a study, a derivative news story like the aforementioned is irresponsibly worded and certainly misleading those who wouldn't otherwise examine the study. Consuming anything in "excess" is of course by definition unhealthy. "Excessively" would be particularly bad with something like caffeinated coffee which is linked with adrenaline increase. More concerning, the study had said nothing of the sort to begin with. So it is, like the game of "telephone", information gets butchered as it is passed along.
The "four cups a day" part of the news, an arbitrary figure, is frequently mentioned in each publication's article as you scroll through Google results. There's a false authority here in mentioning a specific amount, even though there's no mention of what four cups even is equated to in most of the articles. There's also no explanations of the cultural bias that in Spain–where the study was conducted–people drink coffee with less sugar and more milk than say the United States. I don't have quantitative info on how they drink it differently but suffice it to agree that they do based on their tradition of serving "Cafe con leche", yet there's nothing pointing this out to readers how it would impact the findings. The "four cups" part of the research all over the Internet, it turns out is actually a comment from the lead researcher Dr. Adela Navarro. She states:
"I would advise drinking plenty of coffee. It could be good for your heart. I think it's a good idea to have about four cups a day. I think it's the polyphenols, they have an anti-inflammatory effect".
Sure, that's all good and well as it were. Her comment is a reasonable conversational statement. But this says nothing about who the drinker is, their caffeine tolerance based on things like genetics and body mass, how they take their coffee, or what constituted the portion of "a cup". The study mentions 400mg, but I only found one article actually sharing that figure which is a bit irresponsible (think people who view a cup of coffee as a Starbucks 20oz Venti). This quote from her also strikes me as a researcher getting too close to their work and exhibiting results bias, where a researcher stops trying to disprove their own findings, a crucial element to scientific practice. I don't know what she thinks of her finding of course, but I wonder why not also point out the polyphenols known to be in tea for instance? Or the apparent 2013 study suggesting the opposite of her longevity finding, that there indeed was an increased link with death after consuming 4 cups of coffee a day in people under 55? It misleads people when the researcher admittedly doesn't know what the link actually is and then the finding(s) get encapsulated as though it is known. The blame as to why this happens can often be shared, although I'd contend the misinformation is actually mostly enunciated by the news publications. (And this is as well to say nothing of financial incentives, in the coffee study there reportedly was no coffee industry related funding, but so often this is not the case, and then the reader of the news isn't necessarily privy to this.)
So what do we do when we are exposed to premises like "4 cups of coffee a day can help you live longer"? Analogously, some studies on red wine at least suggest red grape juice for a healthier alternative of the same benefit. So in this case there's not any real guidance here and plenty of grounds, no pun intended, for misinterpretation of the research, such as the notion of drinking even more coffee than you already do. If "an apple a day keeps the doctor away" was once an oft-quoted adage, the digital age has certainly not helped deter "pop" buzzy statements that get rendered almost useless. We should at least perpetually ask what something means, find out who said what, why, and then keep on going, for veracity's sake.
In the age of online content farming, junk viral sites, and so-called "listicles", media stories increasingly seem to rarely point out the logical fallacies and biases in the underlying research: things like publication bias, self-selection, simultaneity bias, sample set size/randomness, etc. It's up to us to do these things. If an underlying article does declare or hint at any of these informational caveats, the headlines almost never do of course, those are in turn all that gets digested of the story heuristically, and it's almost as though the writing actively discourages critical thinking. Almost every article about this [mortality study in actuality] mentioning coffee is regarding the part of "4 cups a day makes you live longer", a cherry picked quote it turns out, from Navarro herself, likely because it was the most byte-worthy. In the upcoming months, you could picture a statement like this becoming like "common knowledge" neatly bottled up in an aphorism, complete with accompanying blurry Photoshop memes.
"Did you hear they say we're supposed to have four cups of coffee a day?" will perhaps be a frequently overheard small talk. It's about as informative and exact as the "walk 10,000 steps everyday" rule of thumb.
Simply look to the past for such instances, like misinformation on how much iron was in spinach (infamously the researcher accidentally put a decimal place in the wrong spot, which was in turn an untrue story in and of itself), or more recently the incorrect information of fats' role in our diet that reigned throughout the 80s and 90s.) Many of these articles on the coffee news have "science" and "scientific study" in the title. Scientific procedure is very important, until it's being cited ironically so as a false authority. Things like the millennial expression "because science" come to mind. Saying something is because of science is fine, when you intend to accurately explain why, and the known limitations.
With the mortality "coffee" study, the exact wording of the ESC studies' conclusion in their slide presentation as it presently stands is: "Coffee could [my emphasis] be part of a healthful diet." But a quick social media and search engine query shows this is already widely being declared as near-fact based and with poor context on the meaning. Coffee is a somewhat banal topic, so now imagine emotionally and politically-charged research topics like abortion, human rights, global warming, and back.
As I sip my coffee typing this, I think about how I drink the beverage daily, and would indeed love it to be healthy as suggested; but will refuse to not strive to seek to disprove what is reportedly true or false about it just because I enjoy it and want something to be so. Similarly, if I hear someone in the up coming months state "have four cups of coffee to live longer", I'll quietly think in my head, "Maybe..."
Have a thought? Leave it in the comments below.
HiNES
