Psychosis, Hype And Baloney

As the month began, the worldwide press jumped all over a study in the March issue of the journal Addiction purporting to show a causal link between marijuana use and psychosis. "Drug Doubles Mental Health Risk," the BBC reported. "Marijuana Increases Risk of Psychosis," the Washington Times chimed in.

Such purported links have lately become the darling of prohibitionists, but a close look at the new study reveals gaping holes unmentioned in those definitive-sounding headlines.

Before we look at the study itself, let's consider some basics: If X causes Y, it's reasonable to expect a huge increase in X to cause at least a modest increase in Y, but this has not been the case with marijuana and psychosis. Private and government surveys have documented a massive increase in marijuana use, particularly by young people, during the 1960s and '70s, but no corresponding increase in psychosis was ever reported. This strongly suggests that if marijuana use plays any role in triggering psychosis, that effect is weak, rare, or both.

For this reason, researchers should approach "proof" that marijuana causes serious mental illness with great caution. The researchers in this case, a New Zealand team led by David M. Fergusson of the Christchurch School of Medicine and Health Sciences, seem to have done just the reverse.

Fergusson's team looked at a group of 1,265 New Zealand kids who were followed from birth to age 25 and assessed at various points along the way for a variety of physical, mental and social problems and issues. At ages 18, 21 and 25 they were assessed for both marijuana use and supposed psychotic symptoms. Having found a correlation with daily users reporting the highest frequency of psychotic symptoms, they then applied a series of mathematical models. These models are designed to adjust for possible variables that might confound the results and to assess whether the marijuana use caused the symptoms or vice versa.

Whatever model was applied, the correlation held up. But the reported "growing evidence" that "regular use of cannabis may increase risks of psychosis" depends completely on the validity of the underlying data, and those data raise some screamingly obvious questions.

Psychotic symptoms were measured using 10 items from something called Symptom Checklist 90. Participants were asked if they had certain ideas, feelings or beliefs that commonly accompany psychotic states. The researchers did not look at actual diagnoses, and the symptom checklist is not identical to the formal diagnostic criteria listed in the DSM-IV manual. Perhaps most important, they only used 10 "representative" items from a much larger questionnaire.

These 10 items focus heavily on paranoid thoughts or feelings, such as "feeling other people cannot be trusted," "feeling you are being watched or talked about by others," "having ideas or beliefs that others do not share." This presents a big methodological problem, because it is well known that paranoid feelings are a fairly common effect of being high on marijuana.

But the article gives no indication that respondents were asked to distinguish between feelings experienced while high and feelings experienced at other times. Thus, we are left with no indication at all as to whether these supposed psychotic symptoms are long-term effects or simply the normal, passing effects of marijuana intoxication. While it's possible the researchers had these data and didn't see a need to report them, the failure to do so is downright bizarre. It's like reporting that people who go to bars are more erratic drivers than people who don't, without bothering to look at whether they'd been drinking at the time their driving skills were assessed.

Even if these were long-term effects, the researchers seem not to have considered that what might be an indication of psychosis in other circumstances could be an entirely normal reaction for people who use marijuana. Consider: Someone using a substance that is both illegal and socially frowned-upon almost by definition has "ideas or beliefs that others do not share." This is not a sign of mental illness. It's a sign of a rational person realistically assessing his or her situation.