Posttraumatic stress disorder, sleep and medical cannabis treatment: A daily diary study, Sharon R. Sznitman et al., 2022

Posttraumatic stress disorder, sleep and medical cannabis treatment: A daily diary study


Despite increasing use of Medical Cannabis (MC) among posttraumatic stress disorder(PTSD) patients, research is lacking on how MC treatment relates to PTSD symptomatology, in particular sleep disturbances. This study examines the time gap between MC use and sleep onset and its association with (1) number of awakenings throughout the night, (2) early awakenings, (3) nightmares. Each morning over a two week period, 77 licensed MC patients suffering from PTSD reported on the timing of previous night MC use and sleep disturbances. Within-person analyses found that shorter time gaps between previous night MC use and sleep start time was associated with lower likelihood of experiencing nightmares throughout the night, but it was not associated with nightly awakenings or waking up too early. Between-person analyses showed that individuals who used MC products with higher CBD concentrations reported fewer early awakenings. These preliminary results indicate that future research should test causal relations between MC use and sleep problems in PTSD patients. Future research is warranted in order to explore causal relationships between MC use, nightmares and insomnia in PTSD patients.


Sleep disturbance is considered by some investigators to be a “hallmark” feature of Posttraumatic Stress Disorder (PTSD) (Germain, 2013). Nightmares and other sleep disturbances are included in the Diagnostic and Statistical Manual of Mental (DSM–5) PTSD diagnostic criteria (APA, 2013). It is estimated that as many as 80-90% of PTSD patients report sleep impairment (Richards et al., 2020). The most common sleep disturbances among PTSD patients are insomnia and nightmares (Krakow et al., 2002, Ohayon and Shapiro, 2000). Research indicates that PTSD patients with higher levels of sleep disturbances have more severe PTSD symptomatology (Short et al., 2014), worse functional disability (Giosan et al., 2015), and a heightened risk of both suicidal ideation (Betts et al., 2013) and self-injury (Short et al., 2015).

It stands to reason, therefore, that treatments that reduce sleep problems, may have beneficial impacts on PTSD. Various medications have been used in attempts to alleviate PTSD sleep disturbances, including nightmares. While some (e.g., Prazosin and Olanzapine) have shown promising effectiveness in sleep disturbances related to PTSD (Lipinska et al., 2016), the management of sleep disturbances in PTSD patients remain challenging, partly due to a lack of standardized delivery and a high dropout rate (van Liempt et al., 2006).

In addition to standard pharmaceutical treatment options, a growing body of research shows that cannabis users, both recreational and medical, including PTSD patients, report that cannabis use improves sleep (Bonn-Miller et al., 2014, Walsh et al., 2013). Research has found that individuals with more severe PTSD are more likely to endorse cannabis to improve their sleep (Bonn-Miller et al., 2014, Metrik et al., 2016). Yet, the evidence base for the effects of cannabis on sleep in general, and in PTSD patients in particular, is limited.

While there are a multitude of cannabinoids and chemical compounds in the cannabis plant, Δ9-Tetrahydrocannabinol (THC) and Cannabidiol (CBD) have received the most scientific attention. Within the limited literature on sleep and cannabis, early phase clinical research shows that THC may be effective for improvement in self-reported sleep quality and reduction in nightmare frequency in PTSD patients (Fraser, 2009, Jetly et al., 2015, Roitman et al., 2014). In terms of CBD, research so far has mainly been based on pre-clinical animal models (Elms et al., 2018, Loflin et al., 2017). In one particularly relevant study, CBD administration in a rat model of PTSD showed improvements in sleep (Hsiao et al., 2012). Yet, animal model research has also shown that CBD has alert-promoting effects (Murillo-Rodríguez et al., 2006, Murillo-Rodríguez et al., 2011).

Despite potentially promising results, the preliminary evidence base has some critical limitations. Firstly, the studies examined short-term effects, whereas medical cannabis (MC) for PTSD is typically used for prolonged periods. Research suggests that long-term use of cannabis may lead to tolerance to potential sleep-inducing effects and other PTSD related symptoms (Cuttler et al., 2020, Gorelick et al., 2013, LaFrance et al., 2020, Schierenbeck et al., 2008). Secondly, the experimental studies are limited as they typically test effects of low-dose single (e.g. THC or CBD) molecule extracts or synthetic pharmaceutical cannabinoids administrated sublingually and by standardized dosing. This is very different from real-world everyday use of MC use for PTSD symptoms. Indeed, most MC patients, including PTSD patients, choose to administrate herbal cannabinoids through smoking (Loflin et al., 2019). Herbal cannabis products differ greatly in THC and CBD concentrations and patients tend to use ad libitum dosing regimes and self-titrate to much larger dosages than those typically used in clinical studies (MacCallum & Russo, 2018). Large THC dosages used by healthy individuals have been shown to be detrimental to sleep (Nicholson et al., 2004).

As far as we are aware, only one experimental study has examined the effects of smoked herbal cannabis on PTSD symptoms. The study was a randomized, double-blind, placebo-controlled, crossover trial of ad libitum use of smoked cannabis containing three different concentrations of THC and CBD, and placebo (Bonn-Miller et al., 2021). None of the treatment conditions were associated with reductions in sleep problems. This could be related to the fact that the sample included substantial proportions of participants with previous experience of cannabis use. Thus the results may have been influenced by prior expectations and biased responding. Many participants in this study reported significant cannabis withdrawal symptoms at the time of randomization and early phases of the study. The study illuminates some of the hitherto unresolved challenges related to studying whole plant smoked cannabis in the framework of experimental clinical trials with between-group study designs.

Given that many PTSD patients use cannabis to treat sleep disturbances and other symptoms (Sznitman, 2020), continued efforts to examine the relations between MC use and sleep disturbances in this population are critical. Israel has a well-developed MC program in which specialist physicians request MC licenses on behalf of their patients. Requests are sent to the MC Unit within the Israeli Ministry of Health (MoH) which approves or rejects the applications. The proportion of MC licenses for PTSD has increased drastically in Israel in recent years and the most recent estimates show that 10% of all MC licenses in Israel are granted for treatment of PTSD symptoms (Sznitman, 2020).

Considering the problems of conducting experimentally controlled studies of the effects of MC use and considering the importance of studying MC as it is used in the everyday life of PTSD patients, the current study used an electronic daily diary method to examine the associations between ad libitum MC use and sleep. The study was conducted in a sample of Israeli licensed MC patients who reported that they use MC to treat PTSD symptoms. More specifically, we examined within-person variation in time elapsed between MC use and sleep start time and its association with three main indicators of sleep disturbances common in PTSD patients (APA, 2013): (a) number of awakenings throughout the night, (b) early awakening, and (c) nightmares.

Electronic diary studies involve repeated measurements over smartphones or computers on a daily basis for multiple days and thus enable reporting of sleep and MC use in near real-time and in naturalistic settings (Carney et al., 2012, Tournier et al., 2003). Such daily diary studies attempt to capture real-life experiences without any kind of experimental manipulation, and thus cannot test causality. However, the observational nature of the diary study design enables the study of patient-directed MC use, rather than research-mandated protocols.

Despite the high external validity that can be achieved through daily diary studies, no study has so far used the method to examine the association between sleep and MC use in PTSD patients. Furthermore, and to the best of our knowledge, only a few studies have used daily data to capture within-person associations between cannabis and sleep in other populations. In one study that recruited 54 weekly cannabis users, cannabis use before sleep was associated with shorter sleep onset latency but not with number of awakenings (Sznitman et al., 2020). Another study with 83 college students found that using cannabis as a sleep aid was associated with longer same night sleep duration, less wake time after sleep onset, but greater next-day daytime fatigue within individuals (Goodhines et al., 2019). Another college student sample (n = 80) found that compared to no-use evenings, cannabis was associated with shorter sleep-onset latency and longer total sleep time (Sznitman et al., 2022). A study with middle-aged and older adults (with and without HIV, = 17) found evidence for longer sleep duration but not sleep efficiency or fragmentation after cannabis use days compared to no use days (Campbell et al., 2020).

Building on the literature, we hypothesized that MC use would be associated with fewer sleep disturbances. Specifically, we expected that respondents would report fewer nightly awakenings, nightmares and early awakenings when MC was used right before sleep time compared to when more time elapsed between MC use and sleep time. Beyond testing these within-person main hypotheses, exploratory models tested between-person differences in THC and CBD concentration and PTSD severity and their relation with sleep outcomes and whether or not these factors moderated the within-person associations between MC use and sleep.


The study was approved by the institutional review board (IRB) of the Faculty of Social Welfare & Health Sciences, University of Haifa [certificate number 161/20]. Written informed consent was obtained from all study participants. Recruitment occurred between April 2020 and April 2021, through multiple channels. Firstly, an email invitation to participate in the study was sent to individuals in the database at the Israel Institute of Technology Cancer Biology and Cannabinoid Research Laboratory

Sample description

Participants’ average age was 40 years (S.D. = 11.23); 56% were male. All participants reported having been diagnosed with PTSD although not all had been granted a MC license for their PTSD condition. Specifically, 82% reported that they had obtained a MC license for PTSD, 14% for chronic pain, 2.6% for gastrointestinal diseases and 1.3% reported other conditions. Mean PTSD severity was 47.57 (S.D. = 16.37) and 19% of participants scored less than 33 on the PCL-5 (e.g. the cut off for probable


In this study we aimed to explore the within-person associations between MC use and sleep outcomes in a sample of individuals with PTSD. We further explored whether CBD or THC concentration and PTSD severity moderated the associations. Results revealed that a shorter time gap between MC use and sleep start times was associated with lower likelihood of experiencing nightmares but it was not associated with waking up too early or nightly awakenings. This is of particular significance as


Policy makers, physicians, scientists, and the public currently have little scientific evidence to draw from in order to understand the effects of MC on PTSD symptoms. Cannabis and its constituent parts, THC and CBD, interact with the endocannabinoid system and sleep regulation in important ways. This may be the mechanism by which MC is related to improvements in both nighttime and daytime PTSD symptoms. Yet, the knowledge base is still being built to understand these nuanced relationships.


This work was supported by the Evelyn Lipper Foundation [Grant number 2027093]. The grant was received by D. M. The funding agencies were not involved in the design, analyses or interpretation of results.