 This is going to be one of the most challenging videos to defend. You, the viewer, have very likely been immunized against the evidence I have to present. If you've watched any documentary on cannabis use made in the modern era, reefer madness will have been held up as an object of ridicule, a hyperbolic overreaction by drug policymakers to a rising tide of drug use by the late 30s youth culture. It's been satirized and mocked endlessly, and I think justifiably so. What I need from you, my audience, is that you put that prior prejudice aside and fairly assess the latest scientific evidence on this topic. None of the primary peer-reviewed sources I'm using here are more than 10 years old, and all of them are by legitimate researchers from around the world. There's a body of evidence recently emerged that links cannabis use, particularly at young age, to risk of psychosis and the schizophrenia spectrum. I'm going to explore the data and I invite you to do the same. I'll only be using sources that you, the viewer, can read and confirm. Because I know what a typical reaction will be to this content, I want to ask you to watch the whole video. Pick at least one primary source and confirm that I'm representing it fairly. Share your findings in the comments if you like. My first source is my most important. It's a comprehensive review of all evidence on this topic up to about a year ago. It's written by a psychiatry group at Yale that specializes in schizophrenia research, and the authors are all very well published in this field. This video will be a summary and paraphrase of that paper. In fact, I'll start with their conclusion, taken word for word from that paper. Exposure to cannabinoids in adolescence confers a higher risk for psychosis outcomes in later life, and the risk is dose related. Individuals with polymorphisms of Compti and AKT1 genes may be at increased risk for psychotic disorders in association with cannabinoids, as are individuals with a family history of psychotic disorders or a history of childhood trauma. The relationship between cannabis and schizophrenia fulfills many but not all of the standard criteria for causality, including temporality, biological gradient, biological plausibility, experimental evidence, consistency, and coherence. At the present time, the evidence indicates that cannabis may be a component cause in the emergence of psychosis, and this warrants serious consideration from the point of view of public health policy. That's a bold statement, probably a surprise to you. Let's start with the three minute guide to schizophrenia. For one thing, it's probably not a single condition, but a collection of abnormalities in the brain that are grouped as following on a spectrum of behavioral symptoms. The common element on a behavioral level is grouped into positive and negative symptoms. Positive symptoms, behaviors exhibited by patients with schizophrenia, include delusions, hallucinations, disorganized thought processes, and sometimes movement disorders like constant fidgeting or agitation. Negative symptoms, normal behaviors lacking in patients with schizophrenia, include neglected hygiene or inability to complete routine daily tasks. It may include lack of emotional response or inability to speak in social situations or make decisions. Patients on the schizophrenia spectrum may have only a few of these behaviors or distinct subsets, and their diagnosis of schizophrenia would include where on that spectrum they fall. Underneath the behaviors in the underlying neurobiology, the cause or causes of schizophrenia are still unknown, but it's fair to say that the dominant working theory is around dopamine metabolism. Dopamine is a neurotransmitter found in your brain, and there are several receptors that use it to send signals between brain regions or neurons. I tend to associate dopamine with the reward response, a little burst of pleasure we get when we do something we enjoy. As such, it's involved in most conditioned behaviors and habits, both good and bad. Dopamine is also involved in motor functions. Abnormalities in the brain's dopamine levels in specific brain regions may be involved in Parkinson's and ADHD, and quite likely in psychosis and schizophrenia spectrum disorders. Why is this significant to cannabis use? We know that cannabinoid receptors interact with dopamine receptors in the brain, and we know that high potency cannabinoids can induce the symptoms of schizophrenia by disturbing the way your brain handles dopamine signaling. You're probably familiar with the stories of K2 or spice drugs causing psychosis in first-time users. This is probably the origin of the reefer madness idea, using a lot of pot in a short period of time, can cause some of the behaviors that define schizophrenia spectrum disorders, and through the same mechanisms that we suspect actually cause it. But those people go back to normal a few hours later. At this point, at least a few of you are already saying correlation doesn't imply causation, and I applaud your skepticism. The problem is that all causation is correlative. So if we want to look for causation, we have to start with two factors that correlate. How then can we go from two factors being related, in this case early cannabis use and development of psychoses, and determine whether one causes the other? This was the very problem facing Sir Austin Bradford Hill, a British epidemiologist in 1965. He was working on the correlation between smoking and lung cancer. Inspired by the work of Sir Karl Popper, famed philosopher of science, Hill wanted to create a series of predictions that could be falsified by experimental evidence in the absence of a causative relationship. What he created became known as the Bradford Hill criteria, a series of nine factors between two correlated variables, that while never being quite definitive proof, greatly increased confidence in a causative role. I want to go through those nine criteria for cannabis and psychosis. The first factor is temporality or time dependency. We know that people who develop psychoses are significantly more likely to use or have used cannabis than a control population. But which came first? In 69% of cases, cannabis use precedes onset of schizophrenia by at least a year. In only 11% of cases did the onset of psychosis precede the cannabis use. There's reason to be cautious here. Schizophrenia is a disease defined by its symptoms, and it's possible that we could miss the early signs of a silent psychosis. But it can be confirmed that full symptomatic psychosis usually follows cannabis use, not preceded. Next, biological gradient. Does more cannabis use, defined in terms of joint smoked in one's lifetime, put someone at higher risk of the disease state? The evidence is pretty strong that the more pot you smoke, the higher your risk and the earlier the onset of the condition. The earliest work on this was done in a population of 45,000 Swedish conscripts followed over 15 years. Among those who had previously smoked pot at least 50 times, the risk of schizophrenia was 6 times higher. Among intermediate users, those with between 11 and 50 uses, the risk ratio drops to 3 times higher. Other more recent studies have confirmed similar findings, even after all the potential confounding variables have been removed. And when patients smoked five or more joints a day, the risk of developing certain schizophrenia symptoms was elevated 11 fold. Consistency is also a criteria. In the last 20 years, the vast majority of studies on this subject have found the same elevated risk ratio. Seven longitudinal studies, characterized by their large study populations, all of different populations by different authors, have all converged on similar findings, although not all of them reach statistical significance. The next factor is biological plausibility. Since we don't definitively know the cause by which patients develop schizophrenia, it's hard to say if daily cannabinoid stimulation of your brain can lead to a permanent dysregulation. If the dopamine hypothesis is true, then we have a pretty plausible mechanism. Again, we know for a fact that pot smoking is interacting with dopamine metabolism and in regions of the brain that are associated with schizophrenia progression. In another video, I'll talk about all the progressive loss of white and gray matter visualized by MRI in long-term cannabis users. I think, possibly more strongly than the authors of the review, that the biological plausibility is quite high. Coherence refers to the agreement between experiment and observation in real populations. It's hard to do experiments on humans that involve exposing them to marijuana for multiple years, or exposing young people to see how it affects their mental health later in life. That's neither ethical nor feasible. However, in acute exposure, that is a lot of pot in a short time, or high-potency cannabinoid drugs, we do see a good agreement between experimental prediction and the psychosis-like behaviors exhibited. And we know that in patients with schizophrenia, especially those early in the progression of the illness, using cannabis is predictive of worse outcomes, more debilitating symptoms later on. The criteria of experimental evidence is likewise satisfied. So, what's missing from a perfect Bradford Hill criteria test? Certainly specificity. If every person who ever smoked pot developed schizophrenia, the 60s and 70s would have filled the mental health wards. As it stands, the rates of diagnosis of psychosis haven't much changed. Likewise, in one perspective study, only 21% of patients diagnosed with a psychosis ever previously smoked cannabis. What's very instructive is to compare the specificity of linkage to other mental disorders and other drugs. Cannabis users, for example, aren't at significantly more risk of depression or anxiety in the same longitudinal studies. Conversely, alcohol use wasn't significantly linked to psychosis risk, while amphetamine use was, but at a lesser risk than cannabis use. So, the specificity is low but consistent. We're going to come back to this. The strength of the association is likewise low. The relative risk increase is not very much, probably between 40% and 300% increase of psychosis risk with any cannabis use, based on the longitudinal studies we've discussed. At the highest levels of use, multiple joints a day, that risk elevation climbs to 11 fold. Also really striking were patients admitted to treatment with early signs of psychosis induced by cannabis use. Fully half of them progressed to a diagnosis of schizophrenia. The last criteria is analogy. Is this sort of relationship consistent with other similar causative agents? We're trying to falsify whether cannabis use is causative for schizophrenia. We could compare similar component causes. Eating a high sodium diet puts you at risk for heart disease, for example. Not everyone who eats more than 2 grams of salt a day develops heart disease, and not all hypertension is found in people with high salt diets. Likewise, some pachyday smokers live to their 90s and never develop lung cancer, while some lung cancer patients are never smokers. A component cause is one that by itself can't explain all disease, and all cases of the disease are not caused by that agent alone. Tobacco smoke is responsible for 21% of all types of cancer related deaths worldwide. By comparison, it's estimated that 8-14% of schizophrenia is the result of cannabis use. The analogy to all cancer and all schizophrenia is chosen intentionally. Cancer is a diverse set of diseases caused by diverse causes all contributing and cross-promoting each other. Likewise, schizophrenia is highly complex and diverse, with multiple causes and risk factors that interact in complex ways. I want to address three of the most common beliefs about cannabis and psychosis or schizophrenia using what we've just discussed. Belief 1. Cannabis can't cause psychosis. It can only bring it out in people already predisposed to mental illness. As we've just addressed, cannabis is a component cause. There's no single cause of schizophrenia. It's a host of factors. From this video, you should be more prepared to understand that this relationship is low strength and low specificity, but that it still represents a serious threat to community mental health. We do know of genes that magnify the psychosis risk of cannabis use. However, like cannabis use, genetics are only a component cause as well. Belief 2. If cannabis causes psychosis and cannabis use is increasing, why are the number of people affected relatively flat? This is a good question and not one that's likely to be answered easily. It's possible that we will see more people with schizophrenia as cannabis use becomes more prevalent. There could be factors that are decreasing some types of schizophrenia while the cannabis-induced schizophrenia diagnoses are increasing. Belief 3. The association between cannabis and mental illness is because people are self-treating their condition. This has been tested multiple times. What's called reverse causality. What we can say with confidence is that reverse causality fails most of the Bradford Hill criteria, while forward causality meets most of them. We can take a random group of people and, from their cannabis use, make a statistically accurate prediction of which of them will progress to schizophrenia spectrum disorders that is based on the frequency of their use alone. As I mentioned, in only 11% of cases does cannabis use come after symptoms start to show themselves. One of the frustrating things for people new to scientific research is the lack of certainty. You never get to peek to the back of the book to see how it ends or look for spoilers on the internet. In the case of the link between cannabis and schizophrenia, we're still not 100% certain and we may never be. At this point, I can only say we don't have enough evidence to let reefer madness go down in the history books as just silly paranoia. The expert in this field have expressed concern based on solid epidemiology and experimental science. Any public policy decisions we make about how cannabis is handled will have costs and benefits. It doesn't benefit anyone to have false information, propaganda, occupying the debate about which is the best course of action. I hope I've informed and educated you a little bit about a potential risk, and regardless of your choices, I hope for you good health. Thanks for watching.