One month we hear “Hard exercise is good for you” (paraphrased headline), and literally the very next month we hear “Moderation as the sweet spot for exercise” (actual headline) from the same news outlet. Gee, how do you actually make any sense of the conflicting nature of the news and actually know what do?
There are a few important things to keep in mind: First, people in the news business are in the business of selling news (as in NEW, they need new stuff). If the only health article they ever ran was 100 percent consistent and uncontroversial, then it would be: “You should exercise, you should eat better, and you should get enough sleep”. Nobody would buy that more than once, and as a society we’d stop sharing information and we’d do a lot less research, which I think are bad things.
Second, newspaper budgets and staff are being/have been slashed. Reporters and editors have to be responsible for more and more content with less and less staff. They rarely have the time to look up the study, check the “methods,” and write up their analysis of what actually happened. Very often what you read is a press release sent out by the authors of the study who have their own agenda. (That’s not an accusation, it’s a fact of life. Everyone has their own agenda.)
Third, slow down. If you read “New research: Twinkies are actually health food,” don’t go out and immediately buy a box of Twinkies. (If it’s Twinkies, you don’t need me to tell you that, but look at all diet and exercise headlines the same way, even if you agree with them.) Take charge of your own health, and do a little bit of follow-up.
This last part (follow-up with scientific research) is what people seem to loathe, so I made a super simple two-step checklist that’s a pretty good BS-filter because most people seem to hate math and science. (I think some researchers take full advantage of this to pass off bunk to the public.) It boils down to just two questions:
1. Did they even test what they were trying to test?
2. Does the way they collected their data make sense?
Did they test what they were trying to test? Sadly, this does not go without saying. In 2009 there was a diet study that made big waves by proclaiming that low-carb and low-fat diets are equally effective, the only thing that matters is the number of calories in the diet. Based on the press release for this study, you’d be under the impression that they had a “low-carb group” and a “low-fat group” who they fed the same number of calories, and they then compared their results.
Nope. This kind of information is usually in the “methods” section. And for this study it turns out that “Carbohydrate-rich foods with a low glycemic index were recommended in each diet.” In other words, there were two groups of people eating the same stuff in the same quantities, and they achieved the same results. (Well, duh.)
If you do nothing else, read the “methods” section of the study.
Does the way they collected their data make sense? About a month ago there was a study in the news with the headline, “What do egg yolks and cigarettes have in common?” The “results” were that eggs clog your arteries two thirds as fast as smoking does. Based on the press release you would think that they had some reliable data for a large group of people about egg yolk consumption and cigarettes smoked.
Not really. They asked middle-aged to elderly stroke victims how many egg yolks they’ve eaten in their lives, and also how many cigarettes they’ve smoked in their lives. Do you know your numerical answer to that question? If you actually have an answer, do you trust your number(s)?
I’ve never had a stroke, pay close attention to what I eat, and have a great memory, but I have no idea how many egg yolks I’ve had in my life. I studied dietetics (to become a dietician) in college, and food questionnaires were a part of our labs. Basically, food questionnaires ask people to accurately and honestly report what they’ve eaten, how much of it they’ve eaten, and how it was cooked. This is usually for a six-month period, a year, or even more. I call this multiple choice guessing.
Furthermore, at best, if you buy the data, all they could establish was a correlation. By definition, a correlation is not cause and effect. It only suggests directly studying what happens when you feed people egg yolks (intervention), and try your best to keep them from doing anything else different at the same time so you can see what actually happens.
We already know the answer to the above intervention. Outside of two drug trials (of thousands), no one has ever shown that reducing cholesterol intake prevents anything.
Your health is too important to trust to a press release. Very often making sense of “new research” is as simple as answering the above two questions.
Josef Brandenburg is a Washington, D.C.-area certified fitness expert with 11 years of experience.