Latest / The Biological Psychiatry Podcast / Cannabis Use and Psychosis Onset
Transcript
- Elvisha Dhamala: From the Biological Psychiatry family of journals, this is the Biological Psychiatry podcast, ⁓ where we discuss the latest research on psychiatric disorders and their treatments. ⁓ Thank you for joining us. I'm your host and Social Media Editor, Elvisha Dhamala. ⁓ Cannabis is one of the most widely used psychoactive substances in the world. A growing body of research has linked cannabis use to a range of psychiatric outcomes, including psychosis. One particularly important finding is that cannabis use appears to be associated with an earlier age at onset of psychosis, which itself is linked to worse long-term prognosis. In recent years, cannabis potency has increased, cannabis use has risen, and legal access has broadened. So this raises the question of whether these trends have changed, how we should think about this relationship between cannabis use and psychosis onset. To answer this question, a new study conducted a meta-analysis including 149 studies and more than 70,000 individuals. Joining me to discuss this work are Dr. Matthew Large and Carly Stevens from New South Wales. Senior and first authors of the paper in biological psychiatry titled Cannabis Use in Age Related Acceleration of Psychosis Onset, a meta-analysis. Dr. Large and Carly, I'd like to welcome you both to the Biological Psychiatry Podcast. Carly: Thank Matthew: Thank you so much for Carly: you for having us. Matthew: having us. Elvisha Dhamala: So let's start with some background. What do we know about the association between cannabis use and psychosis and how has that understanding evolved over the past few decades? Matthew: So at the beginning of my career, was a frequently observed association that was not really particularly well understood. And I think if you'd asked most psychiatrists, they would have said things like, well, people with schizophrenia have a hard life. It's understandable they use cannabis. But early in my career, a number of studies started to emerge looking at cannabis use and then prospectively examining for both psychotic illness and psychotic symptoms respectively. there are now a handful of very large high quality studies showing that people that use cannabis earlier in their life have something like a three to five fold increase in the likelihood of being diagnosed and treated for schizophrenia particularly. there are possibly 20 studies of lesser quality, some large. are looking at psychosis outcomes. But that's just one stream of evidence. The other evidence, there's experimental evidence including double blind trials of the infusion of cannabinoids that produce psychotic symptoms in about 40 of people who receive them. And something is known of the genetics of people who hallucinate when infused with cannabis. And then there's the third stream of evidence really is what we're talking about today, which is the effect of cannabis the association between cannabis and the age of onset. Of course, there is other evidence about cannabis use and relapse of people who already have psychosis, their antipsychotic needs and their behavior, all of which point strongly to, and there's no dispute about the association between cannabis use and psychosis. The question is, extent to which it's causal, dose response curves, and none of it, no individual study proves that cannabis causes schizophrenia, but it's impossible to studies that might prove that and there's lots of circumstantial evidence pointing in the same direction. Elvisha Dhamala: before we get into the study, what is the difference between cannabis and marijuana and does that distinction matter? Carly: So the terms cannabis and marijuana are often used interchangeably. Like when you talk to people out in the public, people seem to pick what word they want and roll with that. these days we really try to go with cannabis, especially in a like medical and scientific standpoint. the word marijuana is a little bit messy, especially when you look at its history. So it is a Spanish term, but it was adopted in a bit of a malicious way during prohibition in the US. So it's got a bit of baggage with it and it was often used mostly referring to illicit cannabis. So these days we go with the term cannabis. Matthew: I'd make the distinction between marijuana and cannabinoids. So cannabinoids are a widely varied group of compounds that interact with the cannabinoid receptor and they include endogenous 2-AG and anandamide. Natural cannabinoids which are delta-9-THC and related compounds and actually a bunch of synthetic cannabinoids as well interact strongly with the cannabinoid receptor are extremely varied and rapidly evolving. Elvisha Dhamala: cannabis is often perceived as relatively harmless, especially compared to other substances. Can you tell us about the broader landscape of potential health consequences of cannabis use and particularly psychiatric consequences? Carly: So this is often an an argument that people fall back on when we try to discuss the harms of cannabis. Th people will start talking about heroin, they'll start talking about methamphetamines and they'll focus on the physical harms of those, which of course there are many. But it overshadows the harms associated with cannabis, which are more psychiatric and cognitive than they are physical. There are, of course, some physical things that happen with cannabis. Like if you're someone who's smoking cannabis, you can get things like chronic bronchitis. there's also this strange syndrome that people can get when they smoke cannabis, where they get really nauseous and they have repeated episodes of vomiting. And often the only thing that makes them feel better is a hot shower. So we have some patients that are vomiting multiple times a day, every day, while they're using cannabis. So we do have some physical things that you can maybe compare for other drugs, but by far the biggest thing is psychiatric harms. So there's been some studies across different psychiatric disorders, but by far the biggest volume of evidence is for psycho. And chronic disorders such as schizophrenia. So, in terms of like the health consequences for schizophrenia, there's an increased risk in the development of psychosis. So for example, one paper in the it was in the eighties, there was a great paper that came out, showed that if you smoked cannabis more than fifty times when you were a teenager, you were six times more likely to get schizophrenia. So you're looking at like a significant increased risk of developing it. Matthew: Thank you. Carly: And of course, which we'll get to a little bit more later, it's associated with early onset. and with that comes like a whole bunch of issues in terms of the clinical course of the disease. So people are more likely to have more frequent episodes of psychosis, get hospitalized more frequently, have more severe episodes. so really there's a lot of evidence Matthew: Thank Carly: linking cannabis with severe psychiatric harms, especially for schizophrenia and other severe disorders. and there's still some evidence for other disorders as well. It's just not as well studied. so there's research showing that it exacerbates manic symptoms in bipolar disorder. and also evidence that it increases anxiety, paranoia, and in some cases worsens depression, which is at odds with what we see with the medicinal cannabis Matthew: Thank Carly: industry where a huge proportion of people are getting medicinal cannabis for anxiety disorders. Elvisha Dhamala: So Matthew, you published a meta-analysis on this topic in twenty eleven. Matthew: for the rest of Elvisha Dhamala: Why was an updated meta-analysis necessary and what has changed since then? Matthew: Well, there are two interesting personal stories behind this. The first is that when I was a junior psychiatrist and thinking of doing psychosis research, my then mentor who was a non-psychosis researcher told me that psychosis research was a graveyard for reproducibility. And so I was keen have at least one reproducible finding in my career. And so we published the study in 2011. It had a somewhat unexpected result at the time, but we thought we understood things well. And that was essentially that our cannabis produced left shift in the age distribution of psychosis towards younger onset. And that we assumed that that would be uniform across age groups. In 2024, Carly approached me and said, look, I need to do this research process. I'd like to do something on cannabis. And I said, well, look, this really does need to be reproduced. I've been following the literature a bit. It looks like not much has changed. it's going to be a lot of work and it's going to be boring. And the best we can hope for is that we'll manage to squeeze it into a mid-ranking journal. So Carly said, yeah, that sounds great. And I was very surprised to be honest. I thought this was something that I was going to have to do. And Carly: Yeah. Matthew: but really what has changed in the last 15 years is this absolute explosion of literature about the topic. Some of it almost certainly driven by the initial paper that has now been cited almost a thousand times and has become of well-known factoid actually really in psychiatry. I've even had a taxi driver tell me about it. And this enormous amount of information is really five times more I can't remember exactly how many times more patients but many, many times more patients allows us to dig under the surface of the result that we had in 2011. and that turned out to be quite an interesting story. an additional factor is that, of course, between 2011 and 2026, cannabis has been legalized and medicalized in a host of jurisdictions, both in the US and outside and correspondingly cannabis potency has gone up enormously. It is now possible to ring an online prescriber and have delivered to your door a cannabis product with 35 % THC, which was an unheard of concentration 20 years ago. And so one hypothesis that we had was that we may be able to detect changes in the ecology of cannabis use in studied populations in studies with more recent publication dates. Elvisha Dhamala: Can you tell us about the types of studies that were included in this meta analysis and how you handle the challenge of synthesizing those data? Carly: our meta-analysis was of observational studies. We managed to find 149 studies that were usable. And these were case control and cohort studies. And there was quite a challenge of synthesizing the data because there are These were hundreds of papers that were from all over the world, different time periods, different clinical populations. yeah, there's there was quite a lot of variation in in the papers. But that's One thing that could be addressed reasonably well through meta-analysis. So taking into account that these papers are inherently different in a lot of ways, you could use what's known as a random effects meta-analysis. And that was the model that we use. So you've got that built-in assumption that there's some variability of the papers, which made it easier to put all the data together. and having papers that are different also allows us to look at what features might be causing these differences. Matthew: So I might add that a meta-analysis doesn't synthesize all of these papers into a single big study. That's not the way to look at it. Here pertinent to study, the age of onset of psychosis varies enormously. The youngest person I've seen in my career was nine. The oldest person is in their 90s. And it so happens that studies also vary in their mean age of onset. And that gave us a window into looking at underlying dynamics of age of onset in psychosis as associated with cannabis use. Elvisha Dhamala: studying cannabis exposure can be particularly difficult. What are the key challenges in measuring cannabis exposure and how did those limitations shape your analysis? Carly: So the most difficult thing with cannabis exposure is that there's no universal metric. Like for example, with alcohol we have standard drinks, with tobacco we have a metric known as pachias that we use in medicine. So you're able to compare different populations, different time courses of using those substances. Whereas cannabis, we don't really have that yet. So it can make research a bit more difficult. Like a lot of the papers that we had use different cutoffs for what they deem to be cannabis use. Like some papers were it yeah if you'd smoked more than 50 times in in your life, some were only once, so it was any exposure at all. Others were you maybe smoked a joint a week for more than six months. So there's a lot of variability there that does make it difficult. really, going forward, like one of the best things would be to come up with a standardized metric so we're able to more readily compare papers and synthesize data. it is a limitation of our paper, being aware that most of these studies did have some degree of cannabis exposure in the non-cannabis using group. But nonetheless, our findings were very robust. and it'll be interesting to see what happens down the track when we actually have some kind of real metric around cannabis. Elvisha Dhamala: So let's get into your findings. Your overall finding was that cannabis use is associated with psychosis onset about two and a half years earlier. Can you put that number into context? So how many people actually develop psychosis and what does bringing that onset forward by two and a half years mean for an individual? Carly: about three percent of the population will experience psychosis at some point in their lifetime, and about one percent will develop schizophrenia, which is the most severe lifelong version of psychotic disorders. so we looked at a bunch of people that developed psychosis and we put them into groups of cannabis users and cannabis non-users. And what we did was we compared the average age of onset of psychosis between these two groups. There were different types of psychosis, but the majority were patients with schizophrenia. And what we found was that the difference between those two numbers, so the difference in a average age of onset was about two and a half years. So Because it's a better analysis, we can't sit here and say, therefore, everyone who smokes cannabis will develop it two and a half years earlier. But we found a really strong association there. So if you've got someone who's developing an earlier onset of psychosis, Matthew: and if you've got someone who can help you. Carly: the illness, like the trajectory set tends to be worse. they tend to be hospitalized more frequently. they have a lower level of overall social functioning, they're less likely to hold down a job. So some people might think it doesn't sound like much having a few years of extra illness, but it actually is quite for people that are also going to be missing out on potential milestones such Matthew: Excellent. Carly: as vocational ones, educational milestones. so really although Matthew: So. Carly: it might not sound like much to some people, it's actually quite catastrophic for for individuals. Elvisha Dhamala: One of the most striking findings in your study is the age-related pattern. So you found no association Matthew: great to be part of this. Elvisha Dhamala: between cannabis use and earlier onset and adolescent onset psychosis, Matthew: So. Elvisha Dhamala: but progressively larger effects in older onset groups and up to six years earlier in the oldest group. How should we interpret this pattern? Carly: So this was one of our most surprising findings because there's a big emphasis in research, which is very fair, that because the adolescent brain is very vulnerable as it develops, that you would expect to see some of the most dramatic outcomes with that patient population. And that's what we went in thinking. We thought if if we're looking at accelerated psychosis across the lifespan, the teenage brain is so vulnerable, surely there'll be a finding there. So Matthew and I were very surprised about that to find that for people who developed psychosis in their teenage years, there was no association between cannabis use and earlier onset. But instead we found that this gap widened over time. So when we looked at people who on average developed psychosis in their early twenties, people who smoked cannabis developed it on average about one and a half years earlier. And when we looked at people in their late twenties, that ballooned out to about four and a half years. And then it was over six years for people over the age of 30. So yeah, we got to see this widening gap Matthew: Thank you. Carly: between cannabis users and non-users as they got older. So rather than focusing on the concept of the vulnerable adolescent brain, it seems that it could be more likely that cannabis has a larger effect in people who develop psychosis at a later age. So that can point to several things. One is the cumulative exposure model, which I think is one of the most concerning ones. The idea that the more you're exposed to cannabis and the higher the concentrations, the lower the threshold is for you to develop psychosis over time. And that would explain our gap that we're seeing, our widening gap over time. And also the fact that, and this would link with a lot of research as well, people with earlier onset psychosis, it tends to be Matthew: Thank Carly: Their genetics that play the biggest role. So some people are dealt a bit of a poor hand when it comes to the risk of developing psychosis. And that's likely the predominant Matthew: I'm going to stop. Carly: factor for them. Whereas environmental exposures may play a greater role later in life, and our research seems to support that. Elvisha Dhamala: given that there are these different theoretical frameworks for understanding how cannabis might accelerate psychosis onset, what would you say your results actually support? Carly: So our results seem to most strongly support the cumulative exposure model, because you you would think that if some if someone's being exposed to cannabis during Matthew: So, please, please leave. Carly: adolescence and they have that vulnerable brain, that you would see like some Matthew: Thank you. Carly: some effect, which we weren't able to find. But that widening gap as people got older does seem to support the cumulative exposure model, which is that the more you're exposed to cannabis, the risk accumulates over time and we have that lowering of the threshold for psychotic illness. not to say that we're discrediting Matthew: seconds. Carly: the adolescent neurodevelopment model. That's definitely not the case. It's it's not that it's one or the other. There's probably both at play depending on what sort of outcome you're looking at, whether it's age of onset psych of psychosis or maybe if it's to do with cognitive issues. But Matthew: Okay. Carly: in contrast to what we originally thought our results align more strongly with the cumulative model as opposed to adolescent vulnerability. Matthew: I certainly agree that they're not mutually exclusive but our study has led me to dig into the adolescent vulnerability model a little bit and it seems to have two main pieces of supporting evidence. One is studies about the incidence of psychosis among earlier and later smokers and that seems to be quite good evidence. The other piece of evidence is really from animal models and the exposure of adolescent rats. But I have to say the evidence for the adolescent vulnerability to cannabis is not overwhelming and certainly needs more research. In contrast, cumulative exposure has in way even less evidence, but it does have our study, is the meta-analysis with the single largest number of studies in it by a long way. But it is also in a way more plausible more similar to perhaps alcohol and nicotine with the well-known mechanism continuous exposure is likely to suppress the activity of the endogenous cannabinoid system. Elvisha Dhamala: this endocannabinoid system is one of the most ancient and ubiquitous signaling systems in vertebrate biology. Can you tell us a bit more about the system and what does it do in terms of how cannabis interacts with it? Matthew: You're quite right. It has an amazing evolutionary history, which is perhaps topic for another podcast. The cannabinoid system comprises the cannabinoid receptors, which are ubiquitous throughout the body, but with the CB1 receptor mainly existing in brains of vertebrates. It also includes some mechanisms for producing endogenous cannabinoids that fit onto the cannabinoid receptor, which are a chemical called 2-AG and a longer-acting one, anandamide, and a bunch of enzymes for synthesizing these endogenous cannabinoids and actually metabolizing them in the postsynaptic cleft. Now, the unusual feature of the cannabinoid system is that it has retrograde signaling. So it's released from the postsynaptic membrane. and works on the presynaptic membrane. And it doesn't turn anything on or off, but it acts as a volume control predominantly for excitatory neurons, but also for some inhibitory neurons and definitely also for some dopaminergic neurons, which links cause. theoretical mechanism of psychosis, is hyperdopaminergic function. Elvisha Dhamala: You also examined whether gender, degree of cannabis exposure, and country income status moderated the association. Were there specific effects you expected to find and how should we interpret your results? Carly: there were certain things that we were expecting to find or hoping to find rather. so I'll start with gender. So when I was originally looking at all the primary papers to put together for this meta-analysis, it seemed to be that in papers that were comparing male and female cohorts, the association with cannabis seemed to be much stronger. Like there were some papers that were showing that women were developing psychosis seven years earlier if they smoked. cannabis like there was and it was remarkably consistent so I thought that we would find something here and we actually didn't But we also looked at the proportion Matthew: I'm hooked up. Carly: of males in different samples and we were able to find something with that which I can talk about later The degree of cannabis exposure was a really Matthew: and come to see more of them. Carly: difficult one. As we were talking about before, there's no standardized metric for cannabis exposure, so there's there's so much variability with it. So Matthew and I really had to Matthew: Thank you. Carly: kind of have a bit of a broad definition of what severe or less severe cannabis exposure was. And we think largely because of that extreme variability, Matthew: folks. Carly: if there is a a dose response relationship, which many papers have shown, it's just it's probably likely that because Matthew: Thank you. Carly: of how variable it is that we weren't able to come up with a result. So we definitely don't want people Matthew: Thank Carly: looking at our paper and thinking, there's no association between more severe and and lighter cannabis use. We largely want people to take that particular analysis with a grain of salt. And country income status was Matthew: success. Carly: one that was a like quite a secondary outcome for us. It was just trying to think about other Matthew: Thanks. Carly: method differences and population differences that might contribute to variability. And I would have thought that we would have found something there, and we didn't end up doing that. But at the same time, with all the lines of data that we had, only eleven were from lower-income countries. So would probably also take that analysis with a bit of a grain of salt on that one. Matthew: So with respect to gender, men have an earlier onset of psychosis than women generally in almost every study. It's usually a few years earlier and men are also more likely to smoke cannabis including male people with psychosis. so there's a potential confound there to do with an interaction between gender and we did examine that. We found that to be true. But our main finding of the age cohort is we've demonstrated that it's statistically independent and actually doesn't account for big gap in age between cannabis and age of onset of psychosis in cannabis and non-cannabis users among people over 20. Elvisha Dhamala: your study did identify two significant moderators, sample mean onset age and sample gender proportion ratio. Can you explain what those findings mean and why they matter? Carly: So I'll start with the gender proportion ratio, given that we were just talking a bit about gender before. So this ratio the larger the value was, so the larger the gender proportion ratio, it meant that the proportion of men in the cannabis using group outweighed the proportion in the non-cannabis using group. And we found that the difference between cannabis users and non-users in psychosis onset, it did get larger. as the gender proportion ratio increased. And that makes sense because as Matthew said before, men tend to have an earlier onset of psychosis. So that finding does align with previous research but by far the most interesting one is the sample mean onset age. So we had to do a bit of maths with sample mean onset age, but essentially it's a proxy for the point in life at which people are developing psychosis. So the larger the sample mean onset age value was, it meant that like on average, that sample was developing psychosis later in life. So when the cohort was developing psychosis later in life, there was a more dramatic difference between cannabis users and non-users. Elvisha Dhamala: So in your paper, you're very cautious about the language that you use regarding causality. But if we were to assume that there is a causal relationship here, what are the implications for public health, particularly given some of the things you mentioned earlier, like the trends in legalization, increasing potency, and expanding use in the general population? Carly: So the The public health implications for this are very significant. When you're looking at the number of people that are using cannabis in Australia, there's two and a half million people smoking it or ingesting it through some form every year. So that's a lot of people. And even in one year alone, there were over six million units Matthew: you Carly: dispensed medicinally. So we've got a lot of people using cannabis here. a lot of these people will have a family history of mental illness or psychosis, which can further increase their risk. So we're looking Looking at a higher risk of people developing psychosis and also, as our paper suggests, developing it earlier. And if they're developing it earlier, they tend to have a worse illness trajectory, often with greater rates of hospitalization, more extensive treatment needed. A lot of people need social support. So, for example, only one in five people with schizophrenia are able to work full-time. That's how debilitating the condition is. So we're looking at all of these people that are being exposed to something that increases the risk Of a really terrible psychiatric harm. And on top of that, I always hesitate to talk about finances, because at the end of the day, it's all about our patients. But if you look at the impact on the health system, it's actually huge. About 3% of healthcare expenditure in developed countries is all to do with psychotic illness. And that's a massive number when you're thinking about the fact that about 1% of people have schizophrenia. Matthew: Thank Carly: so if we're thinking about what can happen. Over time as people are smoking or consuming high potency cannabis, the really the the implications if it is causal are are quite terrifying. It's likely I mean I'm interested to see what happens in the coming decades, what happens with the the rates of development of psychosis, with the severity of illness trajectory. And if it is indeed causal, then I think by not communicating the harms properly and by giving cannabis this therapeutic halo by making it medicinal, I think we've made a terrible mistake when it comes to public health. I definitely don't think we should go back to prohibition. I that's not the way either. It really should be treated more the way like tobacco and alcohol are. it's a psychoactive substance. There's very little evidence that it works for most medical conditions. There are a few, like certain types of epilepsy, but really I think in terms of public health, the drawbacks far outweigh the benefits. So I think unfortunately Matthew: Thank you. Carly: we've made a mistake here in Australia in the way that we're Essentially marketing it to people. Matthew: I suppose the reassuring communication is that most people who smoke cannabis will never develop psychosis. And in a way, I'm less concerned about people smoking cannabis than I am about them being uninformed. I think you can assume that most people are rational actors in their own lives. And currently, as an illegal substance, cannabis comes with no meaningful health warnings. And paradoxically, as a medically prescribed compound, it comes with no meaningful health warnings. And so I think the real public health implication of causality is much more to do with communication than anything else. Elvisha Dhamala: So if you had one key message that you would want individuals who use cannabis, their family members, clinicians, and policymakers to take away from your study, what would it be? Carly: like. Matthew: Okay, so I mean the key message really in my view is to be properly informed and to be cautious. Carly: So my my message is along the same lines as well. So really the cannabis is associated with significant psychiatric harms. as Matthew said, like most people who smoke cannabis won't develop psychosis, but it's important for the public to know the risk just for informed use. gonna say. Matthew: one important thing would be that the harms of cannabis don't seem to be restricted to people who use in their adolescence. a note of caution for cannabis smokers of all ages sounded by this paper. Elvisha Dhamala: That's all for today's episode. Thank you to Dr. Large and Carly for joining us to share their work on cannabis use and psychosis. For our listeners, you can find the full paper in Biological Psychiatry, and we'll include the link in our episode description. To learn more about our journals, please visit biologicalpsychiatryjournal.com. You can find The Biological Psychiatry Podcast wherever you get your podcasts. Subscribe now so you don't miss our next episode and follow us on our social media channels for highlights and updates. Thanks for listening. ⁓ This podcast is for informational purposes only and does not constitute medical or professional advice. ⁓ The views expressed are those of the participants and do not necessarily reflect those of the Society of Biological Psychiatry, ⁓ its family of journals, or its editors ⁓ All rights reserved, including those for text and data mining or use in AI systems. ⁓