E173 - 1 - The Future of Cannabis Research: AI, Big Data, and the National Cannabis Registry with Johannes Thrul
Meet our guest
Johannes Thrul, PhD, MS
Research Links
Music By
Desiree Dorion
Marc Clement
(Yes we have a SOCAN membership to use these songs all legal and proper like)
Episode Transcript
Trevor: Kirk, we're back.
Kirk: Hey, Trevor, how's it going?
Trevor: Really good. So you've had some interesting conversations lately. And this one, hopefully I'm saying it right, Dr. Johannes Thrul?
Kirk: \Yes, correct. I was trying to sit here and remember how did I find him and I can't remember.
Trevor: Fair enough. You're getting old.
Kirk: Well, yeah, that is a problem. I do have that limited memory thing going. But no, I was listening to a podcast, I think, and his name came up or the study came up and I reached out and they immediately got back to us. And I was having a conversation with him over like sort of with Dr. Glad that you had, Vlad, Vlad?
Trevor: Vlad
Kirk: Yeah, within within a day and a half, I had the I had the interviewer lined up. So I mean, we I sent out an email. He said, Yeah, I want to talk to you. We set a date all within three days. So, I hadn't I had his voice in our library waiting for us to talk about him within three days of meeting him. And that doesn't usually doesn't happen that fast. It usually requires a lot of waiting.
Trevor: No, that that worked great. So I know you guys talked about other things, but let's just sort of mention off the top, the main thing we're talking with him about is the cannabis and health research initiative. Do you want to talk a little bit about that now?
Kirk: Well, we should introduce ourselves first since we always seem to forget to do that.
Trevor: All right. All right, all right. Fair enough. I'm Trevor Shewfelt. I am the pharmacist. Who are you?
Kirk: And I'm Kirk Nyquist, I'm the registered nurse and you have found Reefer Medness, the Podcast. And if you're listening to this, you realize we've got over 170 episodes. You can find us at Reefermed.ca. Basically, Trevor and I about eight, nine years ago said to each other, let's learn more about cannabis and let's do it by talking to people from around the world. And I now remember how I met this gentleman. I was reading a paper about cannabis use and cannabis use disorder. So anyways, so what I discovered is that this guy, this guy belongs to team, of other researchers that are trying to set up a national database of medical cannabis users with the hope that they can have a library that they can distill out to people on cannabis. It's a fascinating story, national cannabis study that they're doing and their hope, their hope to gather information about medical cannabis from around America and the world. So that's basically the story. We did talk a little bit about the study, the adolescent chronic use of cannabis use disorder, but we're going to save that for another time. And we're gonna dive into his cannabis and health Research Initiative.
Trevor: Okay, sounds good.
Johannes Trul: My name is Dr. Johannes Trul. I'm an associate professor in the Department of Mental Health at the Johns Hopkins Bloomberg School of Public Health. I am a psychologist by training and have pretty much spent my entire career investigating substance use and increasingly behavioral addictions as well. And my niche, my intersection is with digital technology. And so I use a lot of technology methods to understand substance use behavior, develop interventions, test interventions. I'd say that's about me.
Kirk: Are you mostly a clinician? Are you a clinician or a researcher or have you done both?
Johannes Trul: I don't see patients. I spend all of my time doing research and teaching and university service, but I would say predominantly my time goes toward research.
Kirk: So let's talk about your... So where did cannabis come into your life then? How did that intersect? Because we're gonna get into your relationship with the Cannabis Health Research Initiative, but how did you get the cannabis?
Johannes Trul: Yeah. So I would say just based on my upbringing, I grew up in Germany, in rural Bavaria, and cannabis was around when I was growing up. And so I seen some folks develop what I would say pretty problematic relationships with cannabis early on. And then other folks were completely fine. And so in that sense, it's been kind of around me growing up, but then professionally, in my graduate work, I became interested in prevention really, I become maybe a bit disillusioned with the clinical model of treating mental disorders after the fact, and wanted to do my PhD work on prevention. And at the time in Germany, when I was looking for PhD positions and I was looking for prevention, most of the prevention that was funded at the time was substance use prevention. So that was my entry point into substance use. And I started working on mostly tobacco, worked on a clinical trial for a smoking cessation intervention for teenagers. And that way was kind of my entry point into the substance use and addiction research. Incredibly fascinating. It's what I feel like I want to do for the rest of my life. But um. So starting with tobacco, but then when you, when you actually start digging deeper into substance use patterns and behaviors, you quickly realize that we have maybe constructed silos of substance use of like, we're either looking at tobacco, we were looking at alcohol, we are looking at cannabis, but frequently a lot of what's going on is substance co-use in the real world. People are using multiple substances. They use substances to modify the experience. And they use whatever kind of substances are available to them based on what they want in the situation in the moment, right. And so in that sense, I became interested in all kinds of substances that young people use, including cannabis.
Kirk: Cool. You know what? I'm changing. I'm pivoting. I said to you, we're going to start with the database. Let's go right into the paper. But what is the difference between tobacco addiction or tobacco misuse? I like that better. Or alcohol misuse and cannabis misuse. Can you explain to me what the difference of those disorders are?
Johannes Trul: Yeah, so the main difference, I would say, is just a substance that is involved, right? I mean, for tobacco, maybe we wouldn't say tobacco is not intoxicating, so you wouldn't have that kind of acute harm, intoxication, like, you know, somebody gets drunk, climbs on a ladder, falls down the ladder, has an acute injury, or driving is a good example, right? Driving under the influence. So with tobacco, you wouldn't necessarily have that. But if we're looking at just substance use disorders broadly, like what does that mean to us? In general, my kind of lay interpretation is always it's continued use despite negative consequences. So people have serious impairment and consequences in their life because of their substance that they know about, but they're unable to stop. Right. And so that goes along with loss of control, craving, withdrawal when they when they try to stop or are in a situation where they can't use tolerance that they have to use more and more to get the desired effects. And then the negative consequences in maybe their professional life, in their social context, in their hobbies, what have you, right? So continued use despite negative consequences and this kind of loss of control over use.
Kirk: Can somebody be functional while being considered misusing? Like, is a person not competent to be a disorder or functional? Like, how does that work?
Johannes Trul: Yeah. So based on the way we make, we give a diagnosis is based on in the U.S on the DSM. DSM-5. And the DSN goes through essentially, it's a checklist. So it looks like in the past 12 months, have you had fulfilled any of these criteria? And then if you meet two or more, then you get a diagnosis. And obviously, there's a kind of, there is a range, right? You can meet two, you can meet more than two. So there's a gradation there. It doesn't really say like whether or not you're completely functional or not functional as a member of society. So I would say there's still a gradations that doesn't go into, I mean there's severity obviously. So you assume that somebody who has a more severe substance use disorder is more impaired. Um, but so, so I think there's no, there's not one good answer to that question other than that it's not a, the way we do give a diagnosis is a yes, no, either you can have a substance use disorder or not, but then when you have it, it doesn't mean that it all the same experience, right? People can then still be on a spectrum of more or less impacted, impaired, negative consequences due to the disorder.
Kirk: Okay, I got a couple questions. I guess I'm going to just take a step sideways for a minute. When somebody misuses a substance in your research, your personal belief, do you view them as losing control of the behavior or they ultimately always in control and the reason why I said that because there's really two types of therapies Right, there's the AA model where you put your hands in the Lord and God you follow God And you you know and one day at a time and then there's a philosophy that basically you're responsible for your behaviors. We need to show you how to correct these behaviors. So is there Is there, from your perspective, when someone has a disorder, are they ultimately in control of themselves or are they lost control because of the substance? Does that make sense?
Johannes Trul: Yeah, yeah, yeah. I'm just trying to see like, I'm trying to figure out how to best respond to this. I would say as with the previous question, there's no clearly yes or no answer. I think based like for the treatment models, I think it just you know, certain treatment models resonate more with one person versus another. I've seen. 12-step approaches work really well for people that's good evidence and for other people the idea of this higher power doesn't resonate and they're looking for something different. Harm reduction is maybe a different approach. Where the goal isn't complete abstinence of from intoxicating substances, but it is, the goal is to reduce the negative consequences due to substance use by whatever the means necessary. And it's maybe more of a treatment person fit kind of question, I would say. And loss of control, again, I think it's a very subjective thing.
Kirk: No, I appreciate that answer. The reason I'm asking is because I'm trying to put myself back in, you know, in my nursing room when I have a client coming in and I'm very much the harm reduction guy and it's a lot of the care mapping I was following was that you know the care mapping didn't necessarily agree with my approach to what you know what how I would approach the patient. So I appreciate that response because it's One thing Western medicine doesn't always allow, it doesn't allow for individualized care. We seem to care map, and insurance programs tend to fund for care mapping. So I just thought I'd ask. So let's jump into this study. "The association of cannabis use disorder versus other substance use disorders with psychiatric conditions and a propensity to match retrospective cohort." Can you explain this study for me? What... What were you trying to accomplish with this? What was your thesis statement when you started it?
Johannes Trul: Yeah. So before I get into the study, I'll just say that this is the work primarily driven by Dr. Ryan Nicholson and Dr. Una Choi. Ryan was a master's student getting an MPH at the Bloomberg School and this was initiated as his master's capstone project that I was his advisor on.
Kirk: Okay.
Johannes Trul: And he's now. Um, a medical resident here at Hopkins in anesthesiology. Um, so the goal of this was to investigate the risk of developing different psychiatric conditions by substance use disorder exposure. Or so does a certain substance use disorder lead to other mental disorders, psychiatric disorders, with a large sample of electronic health records across the U.S. So we use trinities data, which is an aggregator of electronic health records, across many institutions in the U S. And I believe internationally for this we used US data.
Kirk: When we, again, looking at this paper, is there a concern, the thing that I wrestle with is the lack of understanding of cannabis in Western medicine. And if a young child, I mean, at 15 years old, if my mom and dad had taken me into the doctor at the time, I'm sure if they had that diagnosis, I would have been caused a cannabis use disorder kid because I discovered alcohol was not good for me. I was a very bad drunk as a teenager. So I turned to cannabis and I found cannabis much easier and softer for me So, when we look at young kids, how do we determine, who's determining that these children have cannabis use disorder? I know you go through the tick sheet, but is it a fair diagnosis? Do you believe that they're a fair diagnoses? Do physicians know enough to label somebody with authority that they have cannabis use disorder. Is there a flaws in that?
Johannes Trul: Yeah, so with a data set like this, obviously, it's very difficult to speak to everybody who landed in the data set and how the diagnoses were made. In general, I think, you know, providers are probably fairly competent in making those diagnoses. Now, I can't really speak to every single case here. But but I would say, in general, if you follow the DSM right to it to its definition, then we would assume that that these folks are using cannabis to an extent that causes them negative consequences. Okay, that's now how you know how nuanced that is in every case is hard to say.
Kirk: Okay. We, Trevor and I can discuss the results. I'll get into your, but I guess one question is how, just one question more in this study, how would you apply this study clinically? What would be your recommendation to clinicians who read your paper and say, oh my goodness, look at that. People with cannabis use disorder have doubled the threat of doing a psychosis. So how do we apply it clinically?
Johannes Trul: Yeah, I would say it's a call for the screening and early intervention as much as possible. Because we know that there is a particular risk window for development of negative consequences for cannabis use. As you know, right, the brain is still undergoing pretty substantial changes up until maybe the mid-20s or so. And there seems to be maybe heightened risk in adolescence before the brain maturation is finished with cannabis exposure. Now, what's tricky here is there's some assumptions you have to make here with cannabis use disorder that presumably most of these folks are going to be using THC dominant products, right? The THC is driving this. You wouldn't necessarily get CBD. People don't experience this reinforcing. Obviously the other cannabinoids now in the U.S. They're more and more minor cannabinoids that are part of the products. But just because these data are, it's always historic, right. Like we're dealing with data always from the reflection of the past and the market differentiated recently with a lot more different products being available. Now for teenagers, not so much, right? Because there's age restrictions and you can just walk into a dispensary and buy a product. But we would assume that THC is driving those effects. And so, so early exposure to THC in a developing brain seems to be a risky endeavor.
Kirk: Brilliant, brilliant answer. I completely agree with you. Again, clinically, one of the things that bothers me is how cannabis as a plant is always the problem. Cannabis use disorder, cannabis this, cannabis that. And it's like saying, well, we don't talk about barley that way. I mean, barley produces alcohol. We don't blame the barley plant. So, Trevor and I have this ongoing debate about cannabis overdose. Well I don't believe somebody cannabis overdoses because as an emergency nurse, ABC is antidote. If you come in and say I've overdosed, I wanna know what you overdosed on and I wanna if I need an antidote, cannabis overdose isn't a true term. You know, it should be, you use too much THC. And as one of my guests said, well, that's not a sexy way to describe it. You know, cannabis overdose is just so easy, but you know, you took too much THC, takes too long to say. So I completely agree with that answer, I think that's a good answer.
Johannes Trul: Let's say overdose is also tricky because you're not really dealing with if you're if you are overdosing on opioids, if you're overdosing, on alcohol, right, we would be very acutely concerned with loss of life.
Kirk: With brainstem. You're looking at the brainstem, right? yeah.
Johannes Trul: Like really core bodily functions like are you still able to breathe for example, with cannabis, it doesn't look like that. It's not it's not that people you know, who, who overdo it on cannabis acutely aren't in for a rough time, they can be in for a very rough time. But we wouldn't be concerned about acute loss of life because of the shutdown of bodily functions.
Kirk: So help me out here, if you and I were going to change the world, how would we describe the cannabis overdose? What would be the best way to say to a clinician, they haven't cannabis overdosed, they have. what have they done.
Johannes Trul: Yeah, I would probably acute intoxication.
Kirk: From THC?
Johannes Trul: Hard to know. Obviously, now in the US, right, you could, it could be Delta 8, for example.
Kirk: Oh, gosh. Yeah, good point.
Johannes Trul: So it's gotten a bit more tricky. But in general, I will assume we're dealing with the consequences of an intoxicating cannabinoid. More likely than not, that's still going to be Delta 9 THC. But the marketplace has gotten complicated.
Kirk: Yeah, yeah, we spent some time in, I spent some time in Spain, and they're selling hemp products out the front door, but it's, it's Delta 8. And, and it's a different, like, I, to be honest with you, I'm almost, almost scared to try it, you know, because I don't I'm very familiar with THC Delta 9. But what's the Delta 8 high like, and there's no effect on the endocannabinoid system. There's no different effect is just a different cannabinoid from hemp.
Johannes Trul: Yeah, it tends to be, so it's synthesized from the hemp plant. Usually, if you're just taking a hemp plant, right, like you would have delta-8 in relatively small concentrations. My understanding is also I'm not a biochemist, so I take this with a grain of salt, but that you can synthesize it from maybe CBD, but don't quote me on that. But anyway, it's coming from the hemp plant in one way or another. It's derived from the hem plant. And the clinical studies that I'm familiar with where they did kind of the comparison to look at pharmacokinetics, pharmacodynamics, it looks like roughly maybe 50%, half as potent as Delta-9-THC with people are describing it. If you're looking at, you know, Reddit or survey data, what have you, people are describing it as a maybe a little bit more of a mellow hype with less of the... Maybe anxiety, paranoia, what people can experience with Delta 9 THC. Now, whether or not that's actually a property of Delta 8, other than that it's just not as potent and people, you know, if you doubled the dose of Delta-8 compared to Delta-9, you might be looking at fairly similar effects, but it's not as potent. So then if people are taking the same dose over delta 8 is compared to delta 9. That they would experience obviously a more mellow effect.
Kirk: Okay, very cool. All right, so let's go to something that you're more comfortable with. And I think you're extremely excited about from what I've heard on the other podcast is your, let's talk about the Cannabis Health Research Initiative. Now, from what understand, there's basically three components to this, right? There's the National Cannabis Survey, there's the Repository of Data, and then there's The Research Library. Where would you like to begin?
Johannes Trul: We can start with the National Cannabis Study.
Kirk: Sure, okay. Let's give me a give me an overview of what that is for our listeners
Johannes Trul: Yeah, so maybe I'll start even a bit broader if you don't mind.
Kirk: Sure.
Johannes Trul: That this is a large project that my colleague, Dr. Ryan Vandrea here in the Department of Psychiatry at Johns Hopkins in the Behavioral Pharmacology Research Unit and I are co-leading on the large team effort with folks here at Hopkins and then the Realm of Caring Foundation in Colorado. And This was, we wrote this project proposal in response to a request for applications from the National Institute on Drug Abuse that was asking for research to establish a medical cannabis research registry to investigate the health effects of medicinal cannabis. So this entire research program that, as you already mentioned, has multiple different components that we can get into. Is all in service of helping us build better evidence for the health effects of medicinal cannabis. May they be positive, absent, or negative? And we already, we started getting into this a little bit and talking about different cannabinoids, right? This, I agree with you that at this point, the term cannabis is almost a misnomer because we're really dealing with many, many different cannabis products and cannabinoid products under the umbrella that we expect can have very different effects. So one thing is the complexity of the products, of the active ingredients. The other thing is from behavioral pharmacology research, we know that route of administration matters. So whether or not you are. Smoking a product, you're inhaling a vaporized product, you're eating an edible, you using a tincture, something topical, suppositories, what have you, we would assume that those have different effects just based on the route of administration. And then under the medical medicinal cannabis, use kind of when you're looking at the population, right, of folks who are using medicinal cannabis, they can use for all kinds of different reasons, different conditions, maybe multiple conditions. And when you then put all of those things together, the product ingredients are also administration, the different conditions. You're dealing with an insane complexity. That is very, very tricky to get a handle on scientifically. And so that's where we start, all right? How can we start getting a handle on these things? And the goal is, and so the project includes approaches that collect new data of folks, that's the National Cannabis Study, of folks who are newly initiating medicinal cannabis use for all kinds of reasons would really take all comers. And we follow them for a year. So what we're trying to do in this study is to get folks in the door right before they initiate, to have a baseline that is pre-initiation because our assumption is that a lot of the changes that we might see will happen in that initial year. And if you're just taking a cross-sectional snapshot of medicinal cannabis users, you don't really know how long they've been using, they might have been using for already. They've established a pattern that works for them, you're not going to pick up on any changes over time because obviously they have arrived at a product, a dose, a pattern of use that works for them presumably. And so the change that you would be able to pick-up over time in their symptom burden, conditions, what have you, would be fairly minimal, I would argue. So that's why we're trying to get people in the door when they're interested in initiating use, but they have not initiated yet. So that the prospective part of the study. And a couple of sub-studies to this that I don't really have to go into much detail, but we have sub-Studies that use smartphone-based data collection on a subset of people around every assessment time point for. We're getting more detailed pre and post dosage data on pre and posts cannabis exposure. We have a subset of folks that we invite to get biomarkers, send them to a lab where they get blood draws. We can do clinical chemistry analysis on the bloods and get biomarks. And So that's the prospective data collection. And then we're also working with large scale electronic health records here at Hopkins to identify medicinal cannabis use in the electronic medical records. So the situation in the US is, and I'm not sure what it's like in Canada, maybe it's fairly similar, is that there's no checkbox to give folks a label of this is somebody who is using medicinal cannabis. It's not like cannabis use disorder, what we talked about before that gets coded in the electronic health record, medicine cannabis use does not. And so it's challenging to identify those types of patients. But if we could identify them, then we would all of a sudden have a whole wealth of data over time to look at health outcomes, right, you could look it. Healthcare utilization, you could look at changes in symptoms of disorders that they have just because of things that already get recorded in the electronic health record when you see a doctor and you know, a bunch of tests have been done and things like that. And so the approach that we're taking with that work on electronic health records is in collaboration with some of our good colleagues in engineering and computer science. To develop approaches that utilize AI large language models to go through those records and identify patients who are using medicinal cannabis based on provider notes. So that's, you can imagine messy free text data that are inconsistent across providers, across clinics, to pull out the relevant information and flag patients if they are meeting certain criteria, right? We might see language around where a model is then relatively confident that this is somebody who's using cannabis for medicinal reasons. That could be hemp, it could be THC dominant products, it could be different kinds of routes of administration. And once we can identify those folks, then we can start building these kind of studies where We compare and contrast folks who are matched based on say like social demographics and conditions that they have and are using medicinal cannabis versus not to then look at the trajectory and the outcomes over time. So that's the work with the existing data. And then as you already said, all of the data we have that we collect in the prospective survey will go into a repository that will be accessible for the research community for non-profit research. Once we have a decent number of data in there, then we'll open it up for folks to use. There will be a system in place where people can request the data. There'll be a pilot program to provide funding for junior investigators to get more people into the space. And we have research library that has studies that we have gone through and our team and... Extracted relevant information and categorize the studies based on the relevance for this area of research. And then I'll say the last thing is, broadly speaking, I don't think that observational data will ever replace clinical trials. Clinical trials will still be important, but what these large scale observational studies can show us is where are potential signals for the health effects of medicinal cannabis that we want to dive deeper into, to inform clinical trials moving forward. Because currently, even if you wanted to test a certain cannabinoid for a certain condition, you'd be hard pressed to find data to inform what kind of route of administration should you use, what kind product with which cannabanoid, terpene, what have you, concentration, At what dose should you start? And so the goal of these large-scale observational studies are to kind of find where the signals are to then prioritize potentially which products to move into clinical trials moving forward.
Rene: Well, this is where we're going to pause Kirk's discussion with Dr. Johannes True, and we're going to continue it in episode E73, part two. For now, it's Rene back here in the studio. Thanks for listening to Reefer Medness, the podcast.