Outcomes from this pilot trial suggested nightly supplementation with 150 mg of CBD does not benefit daytime cognitive performance in people with primary insomnia when compared to placebo. Subjective states remained unimpaired as noted with sustained baseline ratings of calmness, clear headedness and co-ordination throughout the trial. There were negligible and transient side effects reported throughout the trial period. Overall, CBD may have some benefits for subjective mood over placebo; however, additional, larger, randomized controlled studies are required to more definitely determine these effects in populations with moderate-severe primary insomnia.

In studies assessing CBD effects on cognitive performance, limited improvements in tasks involving attentional switching, verbal learning, and memory in people who use cannabis (Solowij et al. 2018) and selective attention capacity, processing speed (Hallak et al. 2010), motor speed and executive functioning in people with schizophrenia had been attributed to its neuroprotective effects (Solowij et al. 2018), in addition to suggested linkages with its anxiolytic effects via the regulation of hypothalamic-pituitary-adrenal axis hyperarousal (Narayan et al. 2022). These effects were observed after both long-term administration (Solowij et al. 2018) and single, acute doses (Hallak et al. 2010). It is therefore unknown if these mechanisms were affected based on present cognitive performance outcomes under current study conditions or what effect the current dose size and frequency might have had. The absence of statistical differences between CBD and placebo groups throughout the trial period, including the start of the placebo run-in week suggest no discernible treatment effects of CBD on performance under this specific study design, methodology and timeframe. The timepoints at which improvements occurred for limited outcomes (simple reaction time and numeric working memory accuracy for new stimuli and reaction time) suggest the presence of a learning effect, as speed and accuracy of task completion have previously been observed to increase with repetition (Tao et al. 2019). Comprehensive systematic reviews and meta-analyses reported small to moderate cognitive impairments in the general domains of attention, memory, concentration and executive function (Ballesio et al. 2019; Brownlow et al. 2020; Fortier-Brochu et al. 2012; Wardle-Pinkston et al. 2019). Yet, key limitations emerged from the utilization of cognitive tasks established and validated for neurological conditions known to present major cognitive deficits such as traumatic brain injuries (Fortier-Brochu et al. 2012; Brownlow et al. 2020). This highlighted issues regarding task sensitivity and the detection of subtle cognitive impairments commonly reported in insomnia (Fortier-Brochu et al. 2012). Resultantly, the sensitivity of current validated cognitive tasks to detect subtle impairments is uncertain and warranted necessary validation in samples with insomnia. Moreso, the comparisons of different tasks measuring the same domain across studies further presented a significant limitation in accurately characterizing cognitive performance in insomnia (Fortier-Brochu et al. 2012). Comparable demographic characteristics in each group suggest that other, otherwise unmeasured factors, may have played a role in the overall sample improvements seen in these domains. Importantly, while we did not observe an improvement in performance, a lack of definitively impairing effects produced by continuous CBD dosing demonstrates a decided advantage over frequently prescribed drugs for insomnia, which are known to impair performance over numerous cognitive domains with both acute and chronic administration (Stewart 2005), given past observations of CBD’s potential sleep and mood effects (Narayan et al. 2022; Ranum et al. 2022; Suraev et al. 2020). It must be noted that the time to the maximum measured plasma concentration (T_max) for sublingual CBD is anywhere between 0 and 4 h (Millar et al. 2018). As cognitive tests were often completed over 8 h following the previous night’s dose, we acknowledge that this is unlikely to align to the expected the T_max of CBD (Millar et al. 2018). Nonetheless, we argue that these timelines appropriately align with typical schedules for consumption of this treatment as a sleep remedy, and for assessing potential impacts on daytime functional performance. As the half-life of CBD can vary considerably depending on dose and administration route, any potential treatment hangover effects that could potentially diminish performance were not observed under present conditions (Millar et al. 2018; Nicholson et al. 2004).

Subjective states outcomes demonstrated that at baseline, the CBD group reported greater energy, happiness and lower appetite and therefore, changes to these scales described over the course of the trial indicated that these were not a result of receiving the active treatment. Similarly, despite increased feelings of alertness after being administered CBD, without interaction effects, it is most likely that other factors including changes in daily routines caused by trial participation may have contributed to this effect over time (Perlis et al. 2005). It also noted that feelings of lethargy, sadness, drowsiness or daily appetite did not increase for participants after receiving CBD as measured. Contrastingly, statistically significant interaction effects demonstrating differences between groups during the 2-weeks of active treatment for feelings of calm, clear-headedness and coordination suggested possible treatment effects of CBD. Previous neuroimaging assessments in clinical populations have shown acute CBD doses produced significant changes in the modulation of functional networks and resting limbic activity in conditions including people with psychosis (O’Neill et al. 2021) and generalized social anxiety disorder (Crippa et al. 2011), potentially reflecting its mechanisms for therapeutic benefits in emotional processing (Batalla et al. 2021) over cognitive performance. In contrast, unchanged mood was noted in a sample with schizophrenia when dosing with 600 mg CBD (Boggs et al. 2018), further suggesting these mood effects were dose-dependent and highlight the importance of dose size and frequency for therapeutic effects (Linares et al. 2019; Zuardi et al. 2017; Narayan et al. 2022). Though no consensus exists on beneficial therapeutic doses for mood, higher acute doses (300 mg-400 mg) than those used in this study (Zuardi et al. 2017; Linares et al. 2019) have previously shown anxiolytic effects suggesting the 150 mg was too small for any significant mood improvements. In addition to this, the elevated baseline feelings of energy and happiness in the CBD group suggested the presence of a positive assessment bias as a result of various group differences including lifestyle changes and environmental factors, that may have influenced observed outcomes at the end of the trial. This further signalled the need for results concerning the treatment effects of CBD to be interpreted cautiously.

To our knowledge, this trial was the first to explore effect of CBD on cognitive performance in a population with primary insomnia, further adding to the narrow breadth of research available on cannabinoid medicine, primary insomnia, and cognitive performance. Due to the pilot nature of this trial and the small sample size utilized, there are some notable limitations in the generalizability of our findings. Specifically, primary insomnia was analysed as a homogenous condition and may have resulted in phenotype-specific differences in cognitive performance outcomes between treatments being undetected (Ballesio et al. 2019; Gencarelli et al. 2020). The present results should therefore be considered in the context of these limitations when being generalized to wider groups of individuals with primary insomnia, and more specifically, those using CBD medicinally. The criteria for identifying placebo responders were also purposefully stringent to identify only those participants with the strongest improvements to baseline sleep over the placebo run-in period. Including placebo responders may unintentionally distort results, interfering with the potential to identify treatment effects (Dumitrescu et al. 2019)., and this approach may also inadvertently reduce the generalizability of these findings to the broader insomnia population (Kärppä et al. 2020). We also acknowledge that these exclusions may have resulted in an overestimation of treatment efficacy (Trivedi and Rush 1994). Taking the limitations discussed into consideration, it is stressed that the outcomes of this trial must be interpreted cautiously, with future trials necessary to corroborate outcomes. Despite these limitations, this trial effectively managed confounding factors through robust methodological processes of randomization, double-blinding and strict screening, and exclusion procedures. Nonetheless, we acknowledge that unmeasured factors inherent in the design and implementation of therapeutic trials designed to address sleep disturbance, including changes to sleep hygiene (Stepanski and Wyatt 2003) and overall the influence of trial conditions (Perlis et al. 2005) may have affected outcomes. Several measures were implemented to uphold participant adherence and consistency. These included verbal checks at the beginning of each visit for any cannabis use, excluding people with high caffeine consumption during screening and daily tracking of deviations from participants’ routines including exercise, alcohol consumption and hangovers logged via self-report in sleep diaries. Beyond these measures, adherence monitoring was limited with testing scheduling consistency challenged by changing participant availabilities falling outside the expected start times (9:30am − 12pm). Additionally, it is uncertain how effective blinding was without its ongoing explicit assessment throughout the 3-week trial period. Therefore, high quality crossover studies are required to confirm results and are recommended to analyse differences in primary insomnia based on objective and subjective phenotypes to determine differences between treatment groups when testing the efficacy of CBD. Future studies are thus recommended to explore and compare higher and lower acute CBD doses taken daily for varying treatment periods in the context of these phenotypes whilst utilizing a larger sample size. The pharmacokinetic profile of CBD is also recommended to be explored within these contexts to highlight patterns of metabolism, distribution, and elimination in relation to potential therapeutic benefits (Britch et al. 2021). Use of a crossover design may aid in mitigating participant variations in cognitive ability (Helle et al. 2019) and clarifying neurocognitive differences within those with insomnia whilst assessing the efficacy of CBD treatment effects for this group; however, the potential for introducing further confounding learning effects should be considered when using this design.

With growing interest in the clinical and therapeutic profile of cannabinoid products, these results support the preservation of daily cognitive functioning, calmness, clear-headedness and co-ordination over two weeks of nightly consumption of 150 mg of CBD when used to manage symptoms of self-reported primary insomnia. It additionally corroborates the minimal and transient negative side effects that past studies have reported with short term use and recommends an increased need to explore side effects of its longitudinal use and drug interactions (Huestis et al. 2019). To better clarify the magnitude of these effects, larger trials are urgently needed to corroborate present findings and differentiate between treatment effects and biases perpetuated by limitations in trial design, measures and participants to help establish robust therapeutic guidelines as over the counter CBD availability increases.