A new joint-study between the University of Colorado and the University of Louisville has been making waves in the medical cannabis community. That study boldly titled, “Daily Marijuana Use Is Not Associated with Brain Morphometric Measures in Adolescents or Adults”, makes a compelling case based on three different physical analyses of the brain in daily cannabis users vs. non-users. Taken alone, there’s nothing particularly advanced about the study from a technological standpoint; the real significance of the study is that it flies in the face of a larger body of evidence that seems to indicate that cannabis *can* cause structural changes. …So who’s right? Does cannabis cause structural damage or not? Furthermore, if this study is correct, how could so many independent studies from around the world be wrong?
According to the researchers, the issue revolves around dependent variables and how to account for those variables. To give an analogy here, I may claim that living near the beach causes people to take up surfing, and I may back up my claim with data that shows that beachside communities have a higher percentage of surfers. At first glance, that seems logical. However, this relationship, between the neighborhood and the percentage of surfers, is not proof of anything. In fact, it’s common sense that many of the surfers living in those neighborhoods moved to those neighborhoods specifically to be able to surf frequently. Without acquiring more data or removing this effect somehow, we can’t possibly know if living near the beach makes people more likely to begin surfing. In other words, when answering a scientific question, we have to be very careful to examine the cause-and-effect relationship between variables or we may very easily misinterpret the results.
In this case, the relationship in question is that between alcohol use and cannabis use. At face value, frequent alcohol users are more likely to be frequent cannabis users and vice versa. It’s also well known that frequent alcohol use can lead to structural brain damage. So if we just compare the brain sizes of cannabis users vs. non-users, without taking into account that increased likelihood of alcohol use, some of the differences we observe may have be caused by the alcohol use alone and not the cannabis use. In traditional studies, the answer has been to control for alcohol use by assuming it is an independent variable. This compares frequent alcohol users in the study vs. non-users and corrects for the difference in their scores. The theory is that removing that difference will leave behind the effect caused purely by cannabis. However, the authors of this paper argue that this is an incorrect assumption – specifically it’s possible that alcohol damage is worse when cannabis is present or vice versa. Removing the effect of alcohol alone wouldn’t account for any interaction there. To achieve more accuracy, this particular study ranked all participants on the Alcohol Use Disorders Identification Test (AUDIT) and then made comparisons between cannabis users and non-users within the same groups of alcohol use.
Unfortunately, while researchers try to remove the effect of as many covariates as possible (like alcohol), some are difficult to quantify. For instance, since a large percentage of cannabis sales are illegal, cannabis users are statistically more likely to be types of people who engage in high-risk behaviors, which could affect results if these people are exposed to more sources of brain damage. How do we account for that? In this study, a specific strength of the adolescent data is that it came exclusively from individuals in the juvenile justice system in Albuquerque. This allowed the researchers to compare both users and non-users with similar high-risk behavior, rather than cannabis users with high-risk behavior to non-users with low risk behavior. Unfortunately, this is not common in many cannabis studies, meaning that the data is consistently skewed to suggest cannabis is capable of structural changes that it may not actually be causing.
After taking both of these issues into account, researchers compared three metrics of brain structure in participants, which included voxel-based morphometry volumetric/density analysis, FreeSurfer surface-based morphometry volumetric analysis, and FIRST shape analysis. These methods are beyond the scope of this article, but essentially researchers wanted to record differences in volumes, densities, and the actual shapes of different parts of the brain, in order to have a well-rounded analysis of brain structure. After computing this data, researchers observed no statistically significant difference in brain structure between daily cannabis users and non-users. None.
Of course, there are a few caveats to this study. For starters, the average age of adolescents in the study (almost 17) was over the age that has been cited as the age under which damage most frequently occurs (16 with the most exacerbated damage under 12). So unfortunately, that data doesn’t hold much weight. Secondly, despite all the talk about the relationship between alcohol and cannabis, researchers did not fully remove that effect. Instead, they compared groups of similar alcohol use between cannabis users and non-users, which is certainly more accurate than a blanket removal of the effect of alcohol, but with a sample size of under 50, this would mean that the groups were pretty small and therefore subject to bias from the specific group of subjects. This is less critical if we make the assumption that that population is fairly representative of the population at large, which was arguably true for the adult data, but clearly not accurate for the adolescent data (since it came exclusively from the juvenile justice system). So again, the adolescent data in this study is questionable at best. Finally, lack of long-term structural damage doesn’t mean that cannabis use is harmless – the researchers themselves are quick to point out that short term impairment from cannabis use is real, and for a daily cannabis user, this effect would exist throughout his/her lifetime, regardless of whether the change was structural or not.
Regardless, the big picture, and the reason this study is making waves, is that the adult data gives a hearty, reliable account of cannabis as a substance that does not alter brain structure. Anti-cannabis activists that claim that cannabis destroys the brain or causes irreversible damage are now finding that their claims are not genuinely supported. On the one hand, this type of finding doesn’t sell papers and is unlikely to receive much mainstream coverage. On the other hand, the medical community is at least finally starting to take notice, and ultimately as more research occurs, the truth will come to light.
Barbara Weiland, Rachel Thayer, Brendan Depue, et al. “Daily Marijuana Use Is Not Associated with Brain Morphometric Measures in Adolescents or Adults”. The Journal of Neuroscience (2015) 25(4):1505-1512.