How Do We Epidemiologists Decide What Causes What? Well, It Depends…
Saying that X causes Y is not as easy as you’d think. Or is it?
If you’ve ever heard us epidemiologists talk in public, you might have noticed that we tend to wrap our statements with clauses like “may,” “might,” “could,” “on the average,” or “it depends.” The last one is a favorite of ours because, when you think about it, a lot of things depend on the situation, even some things that we know to be true.
For example, if you were to ask me what the color of the sky was, I would probably tell you that it is blue on a clear day between 9 am and 5 pm, but it is black at midnight if you’re not inside the Arctic or Antarctic circle in their respective summers. Essentially, even the color of the sky depends on the time and place when you’re looking at the sky.
That is the case with exposures and outcomes in public health. Generally, we’re comfortable saying things that have been proven by observational and experimental epidemiological studies for decades. We comfortably say that “smoking causes lung cancer” or that “vaccines don’t cause autism.” Again, lots of evidence, little controversy.
(Yes, I know people out there still think vaccines cause autism. That’s a them problem, not a science problem.)
One of the first lessons I teach my young epidemiology students is about the rules and guidelines for establishing causation. Without those guidelines, it is easy to fall for the trap that X causes Y without taking into account other factors that could be at play. So let me impart some of that knowledge to you. Hopefully, you’ll learn to use this knowledge next time someone tells you that coffee causes pancreatic cancer or that sleeping with wet hair causes pneumonia.
