It Doesn't Matter: The Non-Utility of Medication in Childhood Behavior Disorders?

A surprising new study compared medication vs medication free treatment and found absolutely no difference.

When it comes to the epidemiology of healthcare, thank God for Northern Europe. Countries like Finland and The Netherlands, have national health systems that allow for extremely robust science at a population level. Now, it's pretty easy to argue that places that are not Northern Europe are different from Northern Europe, and make snarky comments about generalizability. However, it's not like we have large scale data from a vast array of other entire countries on the health outcomes across the entire populations to choose from.

Therefore, a lot of the world’s large scale epidemiological and cost-effectiveness data comes from this part of the world.

Today's article is about a surprising study. Not every study shows something is different. Some studies failed to reject the null hypothesis. This means there's no difference between condition A and condition B. Usually, in an experimental design, this is because the drug or intervention doesn't work for the condition being studied. In order to keep you reading, here's a picture of my daughter who probably has ADHD, because her dad has ADHD, and this is a highly heritable condition.

It's always a bit more surprising when we get a result that shows no difference between things that have many studies showing that in more carefully controlled settings there is a difference.

This paper looked at 209 kids who were Dutch. These young people were treated, in their communities, either with a medication free approach to their behavioral disorder, which all of them had, or with a combination of medications and other treatments. Previous studies, the landmark MTA trial, demonstrated that behavioral treatments were inferior to oral medication's in the treatment of ADHD, but that study wasn't specifically including oppositional define disorder in the data set. It was designed to show that a very high-quality behavioral treatment was just as good as methylphenidate, and it was actually stopped in the middle of the study. So the study, with its negative result, challenges orthodoxy, even while it asks a different question about a similar population.

And given I've already, you're not gonna be surprised to hear there wasn't no difference. But it wasn't just there was no difference in terms of outcomes. There was no difference in terms of … anything. There was no difference in terms of cost-effectiveness. There was no difference in terms of quality of life. There was no difference in terms of impairment in the life of the child. It literally didn't matter at all, other than side effects, which are not meaningfully addressed in this paper. No, although this is an observational study, it's not retrospective, it's prospective. Thus the Sample Size is not thousands or millions, it's 100s, but they're closely followed. Here is the thrilling title of the paper:

Let’s dig into the paper. In the process, we're gonna learn a little bit about how to deal with missing data and cost-effectiveness studies, in case you ever needed to know how to do that.