Why Scientists Seem Like They’re Always Changing Their Minds

Is coffee good or bad for us this week? Butter is still okay, right? Are we in a “diet coke will kill you” or a “diet coke is fine” cycle? It can be hard to keep track. But headlines don’t tell the full story. Behind the scenes, scientists aren’t constantly disagreeing and flip-flopping.

Photo by Regina Roig-Romero

Remember, health is complex — there are so many different foods and different people and different things we worry about. No study can answer all the questions about, say, coffee. So each team of scientists looks at the problem space — I envision this as a big sparkly cloud filled with question marks — and cuts off a tiny slice that they think they can handle. (Even “large” studies still address a small question; they just do it by studying a large group of people.)

There are a ton of scientists working on butter, and coffee, and wine, because those are common enough foods that a lot of people care about the answer. So you have hundreds or thousands of teams taking little slices off that cloud, and each coming back with their own answer. They might not always agree, but they weren’t asking the same question in the first place.

Take wine for example. It contains an antioxidant called resveratrol that might have some health benefits, so some scientists will study that. They may even do studies using pills containing resveratrol, rather than actual wine. But then wine also contains alcohol, which has its own potential risks and benefits, so other researchers will study that.

And each team of scientists will take their own approach as they design their experiments. Maybe one will feed resveratrol to mice. Maybe another will survey people about how much wine they drink. And the scientists may be focused on different outcomes: some may look at premature deaths, or some may count heart attacks, or some may just take blood samples to check cholesterol levels.

With all that, you would expect a variety of positive and negative results, right? But if you’re not a researcher, you’re probably going to skim each news article and use the information to put red wine into either the “good for you” or the “bad for you” box in your brain. You’re not lazy, you’re just an ordinary person trying to figure out if you should feel good or bad about having wine with dinner.

Even when scientists study the exact same question, the results aren’t going to line up all of the time. Think of your favourite sports team: they don’t win every game. You might feel a little rollercoaster of emotion from their wins and losses, but you don’t look at a win and say, aha, that proves it! The Raiders are the best team in football ever!

You might hope that’s the case, but you won’t believe it until you start seeing more wins than losses. In science, the equivalent of a win-loss record is a systematic review that evaluates and tallies up previous studies. And sadly, those rarely make headlines.