Anyone who has ever seen “Reefer Madness” or survived a middle-school drug education program has probably figured out that there are a lot of exaggerated fears and warnings when it comes to marijuana. And all that hyperbole has evidently not done a lot to reduce the number of kids smoking pot, according to the latest survey.
But recent research has revealed a possible danger in pot smoking that really ought to worry potential users: It might increase your chances of becoming schizophrenic, or otherwise psychotic. Of course, this does not refer to the temporary effects that cause intensive sensory experiences, the illusion of profound thoughts, and an insatiable craving for brownies. The fear is that in some people, marijuana use may trigger a predisposition to a serious mental illness.
Not everyone is buying the argument that the association between marijuana use and psychotic disorders is cause and effect. A student commentary from Drexel University, for example, points out that the purported correlation is small, and that correlation does not imply correlation anyway. Bloggers Paul Armentano and Mitch Earleywine add that there is no new data in this study, only an analysis of previously published data. They also point out that the rate of schizophrenia is remarkably similar from one country to another, despite different patterns of drug use. But I would not go lighting up a joint to celebrate these counter arguments just yet.
Consider first the argument that the relationship is not a whopping correlation, but rather a small but significant correlation. This is exactly what we should expect, if only a small percentage of the population is vulnerable to schizophrenia and similar mental illnesses in the first place. And because the numbers are small relative to the population, we can’t expect big variations in the overall rate of schizophrenia, drugs or not. And the fact that they found this relationship in old data that looked at the broad population, rather than a new study looking at a targeted at-risk population should make us more afraid, not less afraid.
What this meta-analysis is saying is that the relationship between marijuana use and psychosis is strong enough to show itself even in studies that were never designed to look for it. I’m sure someone out there is running a study right now that focuses on teenagers with a family history of psychosis. If there really is a cause and effect relationship, in which marijuana use triggers a pre-exisiting vulnerability to psychosis, this is the type of study that should show big effects.
And what about the argument that correlation does not imply causation? I’m certainly glad that Mr. Gero and others have kept this in mind. Many science news writers seem not to be able to figure out the difference between a cause and effect experiment and a correlational study. But we should also remember that correlation does imply some sort of relationship. I tell my students to think of the two variables in a correlation as X and Y, and to get in the habit of considering three possibilities: X causes Y, Y causes X, or another, third variable (Z) causes both X and Y. In this story, the X causes Y hypothesis is that pot smoking causes psychosis (but probably only in those with a pre-existing vulnerability). The Y causes X hypothesis is that psychosis causes pot smoking. I think we can probably reject this, since the people in question were smoking pot before they became schizophrenic or otherwise mentally ill. A variation on this is not out of the question, though. Perhaps people who are destined to become schizophrenic are already unusual in a way that makes them more likely to smoke marijuana. This is very similar to the Z causes X and Y hypothesis, that some people have a pre-existing condition that makes them vulnerable to both marijuana use and schizophrenia. I see no reason why the first and last hypothesis could not both be true.
So what is the take home message? Should you worry that you might be one of those people for whom pot smoking could cause a lifetime mental illness? If there is a history of schizophrenia in your family tree, the answer is undoubtedly yes. The problem is that it is impossible to be sure that you are safe. We don’t know what might cause the vulnerability, but it is probably in the genes. Even if you don’t know of any schizophrenic relatives, that does not mean that none of your relatives were carrying some of the genes that lead to vulnerability. I hate to sound like a doomsayer, but it is difficult to rule out the possibility that you may be carrying genes that make you vulnerable. In the end, it is a roll of the dice. To me, the payoff seems pretty small in comparison to the risk.