For years, scientists have sought to better understand the relationship between cannabis and mental health. Of particular interest has been a small but noteworthy correlation between cannabis consumption and psychosis. Though the connection remains poorly understood, it has fueled concerns about whether today’s cannabis products are safe.
“It’s our responsibility as researchers to get our house in order and up our game.”
Amid the confusion, some pundits have seized on the relationship itself as proof that cannabis in fact causes psychosis. In a recent book on the subject, writer Alex Berenson argues that cannabis legalization has spurred psychosis-related violence, causing “sharp increases in murders and aggravated assaults” in states that have legalized. Berenson’s work has been widely discredited by scientists—including some of the authors whose studies Berenson used to draw his conclusions—but the question of what role, if any, cannabis plays in psychosis still looms large.
A new report out of the UK attempts to tackle those questions head-on: When it comes to cannabis and psychosis, it asks, what do we actually know? And how, after years of research, is the relationship still so uncertain?
The study, “Cannabis and Psychosis: Are We Any Closer to Understanding the Relationship?” appears in this month’s issue of Current Psychiatry Reports and contains worrying takeaways about the current state of cannabis research. Overall, one of its authors told Leafly, research data is “a terrible mess” due to procedural inconsistency and a lack of demographic diversity. That can make it difficult or impossible to draw meaningful conclusions from the data, complicating evidence-based public policy efforts.
“It’s looking sloppy, and that really matters: It affects people’s lives and informs public health messages,” said one of the study’s co-authors, Ian Hamilton, a health sciences researcher at the University of York. “In some ways, it’s our responsibility as researchers to get our house in order and up our game.”
Despite Decades of Research, Knowledge Gaps Persist
Hamilton, whose work focuses on the scientific side of drug policy, emphasized that it’s still unknown how cannabis and psychosis interact. While it may be possible for cannabis consumption to bring out symptoms of psychosis, it might instead be the case that people predisposed to psychosis are simply more likely to seek out the drug. Teasing out causality from correlation is a complicated affair. Problems of messy or missing data just make it harder.
While reviewing available research on the connection, Hamilton and his co-author, Mark Monaghan, a social policy researcher at the University of Birmingham, noticed that the body of research on cannabis and psychosis overlooked some key demographics. Women, for example, were largely forgotten.
“When we reviewed the literature on this, it was really apparent that most of the studies either only include men, or, even when they do collect data on both men and women, they don’t report the difference,” Hamilton said. “That is really bizarre and almost strikes me as criminal. If you collect data on men and women but don’t report demographic detail, that’s a waste of your time.”
Why such a disproportionate focus on men? For one thing, Hamilton explained, “Researchers tend to go into substance abuse treatment centers to recruit participants, but the problem is that there are more men in treatment centers than women.”
The imbalance is hardly new, going back to what Hamilton described as a “seminal” study launched in the early 1980s that followed 50,000 Swedish participants—all of them men. “Even from the start,” Hamilton said, “there’s been a bias toward looking at men, and that has continued.”
“It seems remarkable to be having this conversation in 2019,” he added, “but that’s the state of things.”
Hamilton and Monaghan also found that, of the available studies on cannabis and psychosis, nearly all focused on North American or European populations.
“Regions like Africa or Asia simply don’t have the data,” Hamilton said. “If researchers are even interested in knowing what’s happening in Africa or Asia, they have no way of getting the data, because it’s just not collected.”
These kinds of demographic gaps pose a particular problem when it comes to drawing conclusions about cannabis and psychosis, Hamilton said, because available evidence suggests that issues such as cultural norms or an individual’s physiology may be significant factors in the development of psychosis.
“There are social influences on cannabis use and the development of psychosis, and cultural ones,” Hamilton explained. “If we understood and knew more about what was happening in different cultures, it could actually help us know more about the situation in the US and Europe.”
Similarly, “knowing more about women and psychosis might actually help us help men,” he said. “There’s some hints [in the data] that perhaps estrogen plays quite an important role in the development of psychosis.”
Problems With Research—and Researchers
Hamilton and other drug policy researchers believe many of the data’s shortcomings can be chalked up to structural issues.
Geographical gaps, for example, are likely a result of the US funding a bulk of cannabis research historically. “The largest funder in this area worldwide is the National Institute on Drug Abuse (NIDA),” Hamilton said, “which could explain why we know more about Americans than about Africans. If an organization in your home country is funding research, they want to know what’s happening to your home population.”
The outsized role played by NIDA in funding cannabis research has also skewed scientific findings. Sheila P. Vakharia, a researcher in the Office of Academic Engagement at the Drug Policy Alliance, noted that NIDA has traditionally limited funding to research looking for potential harms caused by cannabis, ignoring potential benefits.
“They’re a government organization, and they’re invested in only funding certain kinds of studies,” Vakharia said. “They don’t necessarily always fund studies to look at potential benefits of cannabis, but more often the potential harms.”
On top of that, the United States’ classification of cannabis as a controlled substance means any federally approved studies are generally forced to use government-issued cannabis, which is decidedly low-quality stuff. Tests run on a batch provided to researchers in 2017 revealed that government cannabis contained less THC than labeled. Other samples showed mold and yeast levels that would’ve failed safety tests in state-legal cannabis markets.
When study participants use their own cannabis, Vakharia said, it can also be nearly impossible to ensure that research procedures are standardized.
“If you perform tests with Wellbutrin or insulin, you’re using a well-regulated supply and can do experimental kinds of studies,” she said. “But a lot of research being done on cannabis is observational, not experimental—it’s looking at people who are already out there in the world using it—and we don’t know what kind of cannabis they’re using, what the THC–CBD ratios are, or anything else.”
Such messy data isn’t good enough to allow researchers to draw detailed conclusions—the kind that might shed light on how, exactly, cannabis influences psychosis. “Without any kind of standardized product or system for quantifying how much someone has consumed and when, there are no controls, which makes all of this [research] really challenging,” Vakharia said. “Without controls, you can’t determine causality, but can maybe start to determine relationships.”
And that’s precisely the problem. While research suggests there is, in fact, some sort of relationship between cannabis and psychosis, research problems have prevented scientists from drawing causal conclusions. They’ve also routinely ignored large demographics, including global consumers, women, and people of color.
“We know that Black folks use cannabis at similar rates to white folks but are more likely to be arrested and criminalized for their use,” Vakharia noted. “Beyond that arrest information, what do we know about different folks’ risks, if they’re biological or social?”
In some cases, Hamilton noted, researchers themselves may be part of the problem.
With regard to the gender imbalance in cannabis data, Hamilton said, “The other thing that’s contributed to this, I think, is that men dominate all the senior positions at institutions and journals. It’s men who control the research and publication of articles. If more women had been involved, they likely would have spotted this [gap] earlier.” He added, “Maybe some did, but if so, their voices weren’t heard—or were quashed.”
What Do We Know About Cannabis and Psychosis?
When it comes to cannabis and psychosis, experts agreed that understanding the risks of any drug is important. It’s also crucial that consumers be careful and pay attention to how they feel.
“The vast majority of people who use cannabis, like most drugs, never develop a problem,” Vakharia said, “but we don’t have their data. We know very little about how regular users moderate their own use.”
Vakharia, who is trained as a social worker, noted that psychotic episodes are “actually quite multifaceted and complex,” and that for some people, such episodes emerge “with no clear preceding chemical that they took or [connecting] experience.”
“As a society, we need to get comfortable with understanding what risk really means.”
Psychosis—disruptions in a person’s thoughts that makes it difficult for them to tell what’s real and what isn’t—is a symptom, not an underlying illness, and can be triggered by various factors, including hormones, genetics, trauma, underlying mental health conditions, and others. “We’re talking about people who might have different kinds of delusion, seeing or hearing different things, and get really scared,” Vakharia said.
While psychosis can result from schizophrenia, she noted that psychotic experiences are far more common than schizophrenia and don’t mean a person will develop schizophrenia. There are many circumstances and situations where someone “might act differently than they usually do,” she said.
“For some people, those features can emerge spontaneously or due to stress, and genetic risk can play a role. We also know that alcohol-induced psychosis is a reality, as during long-term use, or withdrawal from alcohol,” Vakharia said. “It also can occur with stimulant drugs, such as cocaine and amphetamines.”
That’s not all. “We also know that it can occur with side effects from psychiatric medications, but also from medical conditions, and from long periods of sleep deprivation or isolation,” she continued. “And it can also happen after people consume cannabis.”
“Our best guess is that you’d need to stop 20,000 people from using cannabis to prevent one case of psychosis.”
Dr. Uma Dhanabalan, a Massachusetts-based physician whose practice includes cannabis medicine, noted that it’s not uncommon for first-time cannabis consumers to experience anxiety, and sometimes this may reach the level of psychosis, clinically speaking.
“Suddenly all of their [inner] fear might surface, and they don’t know how to handle this,” Dhanabalan said. “If you overdo it, it can be intense. You’re not going to die, though—it will pass, and there’s not another drug I can say that about.”
Dhanabalan said that preparation and type of cannabis product, the time and place, and a user’s mood or perceptions prior to using cannabis can also have a profound effect on their experience. “It can also be genetic,” she said. “We found this out with research in 1964: Some [test subjects] who used cannabis felt euphoric, some of them felt nothing, some thought they were going crazy, and some people went to sleep. It can also have different effects at different times.”
One thing that’s clear is that cannabis can have a significant impact on our emotional states and state of mind, Dhanabalan said—including for the better when it’s used thoughtfully.
“I’m originally from India, where it’s been listed in [pharmacopeia] as one of five sacred plants,” she added. Cannabis has traditionally been considered a divine gift from the gods “with a sacred angel that lives in it,” Dhanabalan said, to help us cope with fear and anxiety, and find joy.
Her advice to patients and the public on using cannabis is threefold: “Hydrate before you medicate; always start low or slow, with food, because low blood-sugar can make you anxious; and keep a journal, so you can remember and compare your experiences.”
Some medical professionals have also advised against use of high-THC cannabis concentrates, which can be too potent for many consumers, especially those with a low tolerance.
In the event of a psychotic episode, the next question that individuals and medical professionals need to consider is whether it requires intervention or will pass on its own. In the case of cannabis psychosis, data on the answer to that question is still pretty much unavailable.
That’s in part because much of the available data comes from hospitals, and specifically trips to the emergency room. “We have to understand that, when we use ER data, we’re still not capturing everything,” Vakharia said. For example, there’s no indication of how many people came to the ER but weren’t admitted, who went home with or without treatment, what their preexisting or concurrent health conditions might have been. It’s also wholly unknown how many people may have experienced an episode and chosen not to come in at all.
“As a society, we need to get comfortable with understanding what risk really means,” Vakharia said. “Just because an outcome is possible doesn’t mean it will happen for you, but it’s important to know that it is possible,” she said. “We also need to understand that the way we gather and disseminate information is imperfect, and nuanced.”
Hamilton, who co-authored the UK study on gaps in research around cannabis-related psychosis, agreed that it’s important to “get the risk in perspective.”
“The risk of ever developing an acute form of cannabis psychosis is relatively small,” he said. “We have no exact estimates, but our best guess is that you’d need to stop 20,000 people from using cannabis to prevent one case of psychosis.”
That doesn’t mean the risk doesn’t exist—only that it’s important to put in context. His advice for people looking to try cannabis? “I’m a parent, and what I say to my kids is that I’d prefer them to delay their use until they’re a little older, and be very careful about the amount they use.”
“It’s the same advice I’d give to patients or the wider public,” Hamilton said: “Start with a small dose and work your way up.”