Meddling Kids

¬†In the last few months,¬†I’ve found myself using the “correlation does not necessarily imply causation” argument a tremendous amount. It’s come up largely in conversations about the swine flu vaccine (and vaccines in general), but it’s an important thing for people to understand on a myriad of issues. Simply put, just because something happens around the same time as something else does not mean that it caused it.

Steven Novella over at Neurologica has a great post talking about this particular logical fallacy, giving examples and turning it from one of those things nerds say to something people can understand, and I’d really recommend you read it.

For my own experience, I’ve found this to be a fairly easy argument to explain just by giving examples of both sides. When people say that they ate at McDonalds last night and wound up with food poisoning, it’s likely true that in this case the correlation is associated with causation. Of course, without understanding the full situation it’s entirely possible that this is not the case; a person could have caught salmonella from any number of household surfaces, previous meals, etc. ad nauseum, but it’s also entirely possible that they caught it from improperly prepared food.

But the more I think about it, the more that underlines my point. We don’t know for certain that it was the cause unless we fully understand the variables. Remember those from science class? Variables and constants? Yeah, they count in any interpretation of information. For instance, I could give someone a dose of liverwurst mixed with strawberry jam and then isolate them in a clean room, and then say that they didn’t catch herpes, and therefore that is the proof that liverwurst and strawberry jam prevent herpes. The fact that they were not exposed to the herpes virus is clearly an important variable, and relates directly to the validity of the conclusion.

A two year old child develops autism. The parents are understandably devastated, and they seek to understand what has gone on. They know their child had the MMR vaccination, and all sorts of information on the internet tells them that this is a secret cause of autism. It’s normal and natural to assume that to be the case, but it’s not. We’ve seen plenty of examples that shatter this claim, and the official anti-vax position on the subject has been to move away from autism and towards other neurological claims that haven’t been so thoroughly shattered.

It’s easy to see the correlation, though. It’s probably the first time the child receives an MMR vaccination, and also quite possibly their first direct contact with thimerosal and all the other ingredients. But is that the only variable that changes around two years old?

Did your son have his first sleepover? His first trip to the ball room? His first haircut? His first exposure to Enya? His first McDonalds Happy Meal? His first chiropractic adjustment? His first girlfriend? His first glass of Kool Aid? There are an infinite number of other things, common things that are just as easy to correlate to the display of symptoms, and there is absolutely no information on how these things might affect your child.

But more importantly, we don’t know what causes autism yet. We can rule out the MMR vaccine because places that have banned the vaccine have not seen a decline in new cases of autism. Even if the MMR vaccine was just a part of the equation, removing it from the equation would have an impact on the result. And what we know about autism is that the symptoms are based on social interaction and communication, and that the symptoms are all¬†begin by the time the child hits about three years old. This is the time when we as parents are intently watching every aspect of their development, and thus the symptoms are often quite obvious as soon as they begin to show up.

It’s normal to see the behaviors developing shortly after the vaccination shot because that’s when the symptoms show up. But correlation doesn’t guarantee causation.

At any rate, read the article. Steven Novella is a fantastic skeptical writer, and really nails the explanation.

Jim