New research published in the Journal of Sleep Research points to a possible link between cannabis use and patterns of excessive sleepiness on a widely used sleep diagnostic test. Although the findings were not statistically significant, the trends observed suggest that cannabis use near the time of testing may be associated with outcomes that resemble hypersomnia disorders such as narcolepsy. The authors caution that these patterns could potentially contribute to diagnostic errors and unnecessary prescriptions, but emphasize that more rigorous studies are needed to confirm any connection.
Cannabis is the most widely consumed illicit substance worldwide, and its use has increased significantly since being decriminalized in places like Canada and several U.S. states. Despite its prevalence, the relationship between cannabis and sleep regulation remains poorly understood. Previous studies have reported mixed findings—some link cannabis to shortened time to fall asleep, while others associate it with poorer sleep quality or reduced rapid eye movement (REM) sleep.
In clinical sleep medicine, the multiple sleep latency test, or MSLT, is used to assess how quickly someone falls asleep under quiet, standardized conditions. It helps diagnose hypersomnia disorders such as narcolepsy and idiopathic hypersomnia. However, the test can be affected by various external factors, including medication and psychoactive substance use. While current guidelines discourage cannabis use before undergoing the MSLT, there has been limited empirical data showing how cannabis might influence the test’s outcomes. The new study set out to fill that gap by evaluating how cannabis use might impact MSLT results and possibly affect diagnoses.
“Cannabis use has been rising in the general population, particularly since its decriminalization in several states in the United States and in Canada,” explained study author Nicholas Vozoris, an associate professor at the University of Toronto and staff respirologist and sleep medicine physician at St. Michael’s Hospital at Unity Health.
“Although cannabis has a long history of human consumption, with record of it be used by humans as early as the 5th century B.C., its effects on sleep and wakefulness are still not that well understood, with published studies thus far showing contradictory or paradoxical effects. ”
“In our recently published study, we sought to explore how cannabis use might impact the MSLT, which is a test sleep medicine doctors use to examine a person’s tendency to fall asleep under standardized conditions and it is used in the diagnosis of hypersomnia disorders, such as narcolepsy. If cannabis consumption near the time of an MSLT promoted sleepiness, this might then lead to erroneous hypersomnia diagnoses and subsequent unnecessary prescribing of wake-promoting pharmacotherapy with the potential for adverse drug effects.”
The study was a retrospective analysis of medical records from all adults who underwent an MSLT at St. Michael’s Hospital in Toronto between 2008 and 2018. Data was gathered from the hospital’s electronic medical records, including patient-reported information on smoking habits, comorbid health conditions, prescription drug use, and results from both the MSLT and a preceding overnight sleep study.
The researchers sorted the patients into three groups: active cannabis-only smokers, active tobacco-only smokers, and non-smokers of either substance. People who used both cannabis and tobacco were excluded to avoid confounding the results. The final sample included 139 individuals: 9 cannabis-only smokers, 14 tobacco-only smokers, and 116 non-smokers.
The team compared a wide range of MSLT outcomes across the groups. These included the average time it took to fall asleep during the test, the number of REM sleep onset episodes, and the percentage of individuals who met diagnostic criteria for narcolepsy or idiopathic hypersomnia. They also analyzed results from the overnight sleep studies that preceded the MSLT, such as REM sleep latency and overall sleep efficiency. The statistical analyses were descriptive and did not attempt to control for additional variables, due to the small number of cannabis users in the sample.
Across most of the variables measured, there were no statistically significant differences between the three groups. However, the researchers noted several non-significant trends suggesting that cannabis-only smokers may have been more likely to show signs of excessive sleepiness or characteristics associated with hypersomnia disorders.
Cannabis-only users tended to fall asleep more quickly during the MSLT compared to non-smokers and tobacco-only smokers. Their average sleep onset latency was 8.1 minutes, compared to 9.2 minutes for non-smokers and 10.5 minutes for tobacco-only users. A lower latency can be indicative of excessive sleepiness.
A third of the cannabis smokers (33.3%) met the threshold for severe sleepiness, defined as falling asleep in under 5 minutes on average. This compared to 22.4% of non-smokers and 14.3% of tobacco smokers. Additionally, 55.6% of cannabis smokers had at least one REM sleep onset period—considered a red flag for narcolepsy—versus 28.4% of non-smokers and 42.9% of tobacco smokers.
The data also showed that 22.2% of cannabis smokers received a narcolepsy diagnosis following the test, compared to 8.6% of non-smokers and 7.1% of tobacco-only smokers. Similarly, a greater percentage of cannabis users were diagnosed with idiopathic hypersomnia (33.3%) compared to tobacco users (14.3%) and non-smokers (30.2%).
“Our study showed that there were in fact trends towards multiple markers or measures of increased sleepiness on an MSLT in association with cannabis consumption near the time of testing,” Vozoris told PsyPost. “Therefore, patients and physicians need to keep in mind that cannabis use near the time of MSLT testing has the potential to contribute towards hypersomnia disorder misdiagnosis, and as a result, potential inappropriate long-term prescription of wake-enhancing drug therapy.”
The authors point out that these findings align with a small number of earlier studies. For example, one study in people under age 21 found that cannabis use was linked to an increased number of REM-onset episodes and more frequent narcolepsy diagnoses. Another study in the United Kingdom suggested that drug use might contribute to inaccurate MSLT results in a sizable portion of patients diagnosed with narcolepsy or idiopathic hypersomnia.
The study had a significant limitation. The sample size of active cannabis-only smokers was very small, making it hard to detect statistically meaningful differences. The data also relied on patient self-report for cannabis and tobacco use, which may not always be accurate. Only a few patients underwent objective drug testing before their MSLT.
The researchers suggest that future studies should involve larger sample sizes and more detailed information about cannabis consumption. Objective drug screening, timing of use relative to the sleep test, and control for additional confounders would help clarify whether cannabis truly contributes to false positives in hypersomnia testing.
According to the researchers, patients planning to undergo an MSLT should inform their healthcare providers about any cannabis use. Rather than stopping abruptly, they should work with their providers to determine how best to approach testing.
“If one is consuming cannabis on a regular basis, abruptly stopping it at the time of an MSLT may not be the right step to take, because that might contribute to unwanted test effects or false test results too,” Vozoris explained. “If a patient is regularly using cannabis and an MSLT in planned for him/her, there needs to be discussion between patient and physician about how to come off of cannabis for the purposes of conducting the test.”
“There are also many other factors that can potentially affect MSLT results, beyond recent cannabis use, including how much time the patient slept the night before the MSLT, the presence of other types of sleep disorders, and the use of certain specific types of prescription drugs, such as, serotonergic antidepressants.”
The study, “Impact of Cannabis Smoking on Multiple Sleep Latency Test Outcomes,” was authored by Kosta Tzanis, Jenna Sykes, Clodagh M. Ryan, and Nicholas T. Vozoris.
