The impact of cannabidiol placebo on responses to an acute stressor: A replication and proof of concept study

Abstract

Background:

Our group has previously reported that cannabidiol (CBD) expectancy alone blunts markers of stress, particularly during anticipation, but it is not clear the extent to which such findings were specific to the methods utilized.

Aims:

To examine CBD-related placebo effects on stress reactivity and anticipation and to validate a protocol to be used in a neuroimaging study.

Methods:

Forty-eight healthy adults (24 female) were randomly assigned to be informed that they ingested a CBD-containing oil or a CBD-free oil despite receiving the same oil (CBD-free). Following oil administration, participants engaged in a laboratory stressor and were then incorrectly informed that they would engage in a second more difficult task following a waiting period. Subjective state (sedation, energy, stress, anxiety) and heart rate were assessed at baseline, post-oil administration, immediately following the first stressor, and while anticipating the second stressor.

Results:

Subjective stress and anxiety were significantly elevated immediately following the stressor (p-values < 0.001). CBD expectancy was associated with increased subjective sedation (p < 0.01) and tended to be associated with blunted subjective stress (p = 0.053). Post hoc within-condition pairwise compassions suggested a return to pre-stressor levels during the anticipation period in the CBD condition for subjective stress and anxiety (p = 0.784, 0.845), but not the CBD-free condition (p = 0.025, 0.045).

Conclusion:

Results replicate and extend previous findings that CBD expectancy alone can impact stress- and anxiety-relevant responses in the laboratory context.

Keywords

Cannabidiol placebo stress anxiety

In recent years, there has been considerable interest in the use of cannabidiol (CBD), a cannabinoid found in cannabis, as a therapeutic agent. In contrast to delta-9-tetrahydrocannabinol (THC), the main psychoactive constituent of cannabis, CBD is a nonintoxicating compound that is relatively free of toxicity, serious side effects, or reinforcing properties that may lead to misuse (Bergamaschi et al., 2011a; Blessing et al., 2015; Crippa et al., 2011; Stohs and Ray, 2020). CBD also has a high margin of safety across a range of doses (Blessing et al., 2015), and it does not appear to significantly alter critical physiology parameters such as heart rate, blood pressure, or body temperature (Arndt and de Wit, 2017; Bergamaschi et al., 2011a; Blessing et al., 2015), making it an appealing compound for various therapeutic applications (Khoury et al., 2017; Walsh et al., 2018. Although CBD has been claimed to produce a broad range of therapeutic effects including antiepileptic, sedative, anxiolytic, antipsychotic, antidepressant, and neuroprotective actions (Bergamaschi et al., 2011a; Campos et al., 2016; Khoury et al., 2017), to date, much of the evidence that has been used to justify the use of CBD in a therapeutic context has been derived from preclinical findings and anecdotal reports. As such, there is currently a dearth of well-controlled scientific studies to support CBD’s clinical use (Walsh et al., 2018).

One potential application for CBD that has generated considerable interest has been in the treatment of anxiety- and stress-related disorders (e.g., social anxiety disorder (SAD); post-traumatic stress disorder). In preclinical animal studies, CBD has been shown to have anxiolytic effects across a range of animal anxiety models, including the elevated plus maze, the Vogel-conflict test, and the elevated T maze (for a review, see Blessing et al., 2015). Similarly, CBD has also been shown to attenuate several stress-related responses, including dampening the increases in heart rate that are typically induced by restraint stress, as well as reducing the magnitude of anxiogenic responses that typically follow the chronic exposure to a predator or to unpredictable stress (Blessing et al., 2015).

Direct evidence for CBD reducing anxiety- and/or stress-related responses in humans is more limited. In an early double-blind study, Zuardi et al. (1982) reported that CBD acutely reduced the subjective anxiogenic effects of THC in eight healthy participants, relative to a placebo. A second double-blind, placebo-controlled study by the same group (Zuardi et al. 1993) reported that CBD reduced subjective anxiety following the induction of stress by a public speaking task in a sample of 10 healthy participants. Crippa et al. (2004) examined the impact of CBD on the subjective anxiety induced by undergoing neuroimaging procedures and reported that CBD significantly reduced feelings of anxiety and increased feelings of sedation during the neuroimaging challenge, compared to a placebo. A second double-blind study by the same group (Crippa et al., 2011) extended these findings to a sample of men with an anxiety disorder by demonstrating that neuroimaging-related subjective anxiety is also attenuated by CBD in clinically anxious participants. Bergamaschi et al. (2011a) compared the effects of CBD and placebo on an anxiety- and stress-provoking public speaking task in a sample of 24 participants diagnosed with SAD and found that CBD (vs placebo) significantly reduced subjective ratings of anxiety, alertness, cognitive impairment, and discomfort during the public speaking task. However, not all studies have reported an anxiolytic effect of CBD. For example, several double-blind, placebo-controlled studies with healthy participants found no effect of orally administered CBD on indices of anxiety (Arndt and de Wit, 2017; Hundal et al., 2018; Stanley et al., 2022), but because these reports all cited the possibility that their manipulations did not induce sufficient anxiety as a potential limitation, it is difficult to interpret these negative findings. Collectively, the existing scientific literature is suggestive of a potential beneficial effect of CBD on anxiety and stress. However, the precise mechanisms through which CBD diminishes anxiety and stress responses in humans are not known.

CBD has a complex pharmacological profile, with over 60 identified molecular targets, and some of these targets, such as serotonin 5HT-1A receptors, cannabinoid CB1 receptors, and transient receptor potential vanilloid type 1 receptors, are known to be involved in modulating anxiety- and stress-related responses (Blessing et al., 2015; García-Gutiérrez et al., 2020). This suggests plausible pharmacological mechanisms through which CBD might diminish anxiety- and stress-related responses. However, drug effects in humans are believed to result from a combination of both the pharmacological actions to the drug itself and nonpharmacological factors such as placebo or expectancy effects (Kirsch, 1985). Indeed, open-label trials among adults with diagnoses of anxiety disorders suggest that daily oral CBD administration can lead to clinically significant decreases in anxiety symptoms (Dahlgren et al., 2022), even among those with treatment-resistant anxiety (Berger et al., 2022). Given that participants in these open-label trials were aware that they were receiving CBD, it is possible that expectancy factors were implicated in the observed treatment effects. Even in the previously cited noninterventional studies that report anxiolytic effects of CBD using a randomized double-blind design (the current “gold standard” for assessing drug-related effects), experimenters did not directly assess participants’ perceived drug assignment to verify that blinding was successful. Thus, it is unclear to what extent participants in these studies were able to correctly guess whether they had received the active CBD or the placebo, and we cannot definitively rule out the potential contribution of expectancy-related factors. Nevertheless, the extent to which placebo effects contribute to CBD-related effects on stress and anxiety responses is currently not clear.

Although experimental research on CBD-related placebo responses is limited, a recent study by our group found evidence to suggest that CBD expectancy alone was associated with increased subjective sedation, decreased subjective anxiety among those with the strongest beliefs that CBD has anxiolytic effects, as well as changes in physiological and endocrine markers of stress that are consistent with enhanced anticipatory stress regulation, but not a reduction in stress responsivity (Spinella et al., 2021, 2023). However, there are potential limitations to this previous study. For example, the investigators used a 90-min mock absorption period following oil administration, prior to implementing the stress challenge, which may have diminished the salience of the expectancy manipulation and thus minimized the degree of placebo effects observed. Furthermore, this study utilized the Maastricht Acute Stress Test (MAST; Smeets et al., 2012), a protocol that involves implementing a combination of physical, mental, and social stressors, which led to near ceiling effects on indices of subjective stress and anxiety following its implementation. It is, therefore, not clear the extent to which MAST-related findings would generalize to milder stressors that are more typically encountered in a natural environment. In the present study, we attempted to address these limitations by (i) examining CBD-related placebo effects in response to a less intense stressor that was presented closer in proximity to stimulus administration and (ii) determining the relative impacts of CBD expectancy on acute stress responsivity versus stress anticipation. Moreover, because we planned to examine the impact of CBD placebo effects on neural responses using functional magnetic resonance imaging (fMRI), we also sought to pilot an fMRI-compatible stress induction protocol (Wang et al., 2005, 2007) as a proof of concept.

Method

Subject selection

Forty-eight (24 male, 24 female) participants were recruited for our study through posters and online advertisements (i.e., Instagram, Facebook) around the Halifax Regional Municipality area in Nova Scotia, Canada. To be eligible, individuals must have used cannabis at least once in their lifetime and must have been at least 19 years of age—the age of majority in Nova Scotia. Furthermore, participants had to be free from current psychiatric disorders (American Psychiatric Association, 2013), current substance dependence/substance use disorders, and current prescription medication use (except birth control in women). Lastly, to enhance believability of the CBD content information provided to participants during the study, all participants had never participated in a study by our group that involved explicit deception. Interested candidates that met the selection criteria for our study were invited to schedule a 2-h experimental session. Participants were informed that the study would investigate the impacts of CBD on cognitive performance.

Measures and apparatus

Carbon monoxide measurement

Prior to their session, participants were informed that a breath carbon monoxide (CO) analyzer (Vitalograph, UK) would be used to verify their abstinence from cigarette and cannabis smoking. Given that there is no reliable CO cut-off for cannabis abstinence, the CO analyzer was used as a bogus pipeline: a technique used to enhance compliance with the study’s abstinence requirements by informing participants that a reliable device (i.e., CO analyzer) will be used to verify their abstinence.

Demographics and cannabis use history, substance use calendar, and CBD belief ratings

Demographics (i.e., sex, gender identity, ethnicity, education) and cannabis use history were assessed using an experimenter-compiled questionnaire adapted from the Daily Sessions, Frequency, Age of Onset, and Quantity of Cannabis Use Inventory (Cuttler and Spradlin, 2017). To assess substance use over the previous week, participants were asked to complete a weeklong Timeline Follow-Back Calendar (Sobell et al., 1996). Participants’ a priori beliefs regarding CBD’s effects on stress and anxiety (i.e., reduces stress, reduces anxiety) were assessed using a numerical rating scale ranging from 1 “Not at all,” to 10 “Completely” and, for relative specificity, beliefs about THC’s effects on stress and anxiety were assessed using the same rating scale.

Trait anxiety and perceived stress

The well-validated 20-item Trait version of the State-Trait Anxiety Inventory (STAI-T; Spielberger, 1983) was used to measure trait-level anxiety. Participants were instructed to rate each statement (e.g., “I feel nervous and restless,” “I am cool, calm, and collected”) based on how they felt on a four-point scale from 1 “Not at all,” to 4 “Very much.” Additionally, the 10-item Perceived Stress Scale (PSS; Cohen et al., 1983) was used to measure perceptions of stress during the month before participants’ study session. Participants were instructed to rate how often various life situations were appraised as stressful on a five-point scale ranging from 0 “Never” to 4 “Very often.” Component items were summed to generate a total score, with higher scores indicating greater levels of trait anxiety and perceived life stress.

Acute subjective state

A numerical analogue scale was used to measure participants’ acute subjective stress, anxiety, sedation, and energy. Participants were asked to rate the extent to which they felt “stressed,” “anxious,” “sedated,” and “energized” on a horizontal line from 1 “Not at all” to 10 “Extremely.” Elevated scores on the “stressed” measure are thought to be indicative of nonspecific arousal that led to difficulties with relaxation, agitation, and irritability (DASS; Lovibond and Lovibond, 1995), while higher scores on the “anxiety” measure are believed to be indicative of apprehension, tension, nervousness, and worry (STAI-T; Spielberger, 1983). In contrast, measures of energy and sedation were derived from the Brief Biphasic Alcohol Effects Scale (B-BAES; Rueger and King, 2013), where “energized” represents stimulating drug effects (e.g., elated, excited, stimulated), while “sedated” represents possible sedative effects (e.g., slow thoughts, sluggish). Similar numerical analogue scales have been demonstrated as having sound psychometric properties in the assessment of stress (e.g., Karekla et al., 2017; Lesage et al., 2012) and anxiety (e.g., Davey et al., 2007; Rossi and Pourtois, 2012). Specifically, we opted to include these single-item descriptors as they closely resembled those used in prior research by Spinella et al. (2021), while also enabling us to conduct brief assessments that could be suitable for a neuroimaging context.

Hemp seed oil

CBD-free hemp seed oil (Manitoba Harvest: Manitoba, Canada) was administered sublingually to all participants at a dose of 0.3 mg/kg to mimic the CBD doses that reportedly produce anxiolytic effects in humans (MacCallum and Russo, 2018). Participants were instructed to hold the oil under their tongues for 60 s before swallowing. Notably, hemp seed oil is considered inactive and free of psychoactive properties (Hazekamp et al., 2010).

Stress induction

An adapted version of the standardized Trier Social Stress Test (Kirschbaum et al., 1993; Wang et al., 2005) was used to induce state-anxiety and stress in our sample. Participants were required to perform serial subtraction in increments of 13, starting from a four-digit number (i.e., 2043) as quickly and accurately as possible. During the task, participants were prompted for faster performance by the experimenter and were required to restart the task if they made an error. Participants were instructed to expect a second and more difficult trial following the 10-min resting period to induce anticipatory anxiety. This adaptation was incorporated to enable us to differentiate between acute versus anticipatory stress and anxiety in the present study, as well as to determine the extent to which stress- and anxiety-related responses induced by our stress task could be sustained for the duration of a 10-min post-stressor resting state MRI scan in the planned neuroimaging study. A similar stress-induction paradigm has been shown to induce reliable changes in subjective and physiological measures of stress and anxiety in human research (Wang et al., 2005, 2007).

Heart rate

An Equivital sensor belt (AD Instruments; ADI, Colorado Springs, CO, USA) was used to monitor heart rate continuously throughout the experiment. Electrocardiogram (ECG) data were recorded through an Equivital EQ02 sensor electronic module (SEM) that was attached to a complimentary sensor belt. The sensor belt was secured underneath participants’ clothing. All ECG data were transmitted to a data acquisition software LabChart (ADI, Colorado Springs, CO, USA) for processing. All ECG signals were passed through a high-pass filter (0.5 Hz) to reduce baseline wandering. Three-minute time intervals were selected in correspondence with subjective questionnaires (i.e., baseline, post-oil administration, post-stressor, anticipation). All segments were manually inspected for ectopic beats above/below three standard deviations of the segment’s mean. If ectopic beats exceeded 5% of total beats, the segments were excluded from the analysis.

Concluding questions

Concluding questions were intended to provide a manipulation check to verify whether participants believed the verbal information about the CBD content of the oil they received and to allow participants to provide feedback on their experience with the study.

Procedure

The 2-h experimental session took place in the Substance Use Research Laboratory at Dalhousie University. After providing informed consent, participants completed baseline measures (i.e., demographic information, cannabis use history, beliefs about the effects of CBD and THC) and were weighed to determine the quantity of oil they would receive. A minimum of 12 h of abstinence from smoking, alcohol, recreational drugs, and caffeine consumption was verified by self-report, and a CO analyzer was administered as a bogus pipeline to enhance compliance with study abstinence requirements.

At the start of the experimental session, participants were given privacy to attach to their chest, the Equivital sensor belt, which was used to monitor heart rate frequency continuously throughout the experiment. After the Equivital sensor belt had been attached, participants were asked to rate their current state of stress, anxiety, sedation, and energy using a numerical analogue scale. Initial scores on the numerical analogue scale served as a baseline for individual subjective state.

Following the first numerical analogue scale, all participants received a CBD-free hemp seed oil (0.3 mg/kg) with varying information about the CBD content of the oil, depending on their assigned condition (told CBD vs told CBD-free). To ensure that the experimenter’s observations were unbiased, an independent blinder was assigned to sublingually (i.e., under the tongue) administer the oil and to provide participants with information about the CBD content of their assigned oil (told CBD vs told CBD-free). To enhance the believability of the expectancy manipulation, the oil was delivered in a packaging that was consistent with the participants’ assigned condition (commercial CBD packaging in the told CBD condition, hemp seed oil packaging in the told CBD-free condition). Prior to oil administration, participants were instructed to keep the oil under their tongues for 1 min before swallowing. Oil administration was followed by a 10-min sham absorption period. In order to standardize participants’ beliefs about the timing of CBD effects and to further enhance believability, participants in the told CBD condition were informed that the effects of CBD can emerge 10-min after sublingual administration. At the end of the 10-min sham absorption period, participants were asked to complete the numerical analogue scale for a second time. Following this, participants were instructed to begin the 4-min stress-induction “counting task” by performing serial subtraction in units of 13 starting from a four-digit number (e.g., 2043). Scripted verbal instructions were provided to participants that made a mistake, made no mistakes, slowed down, and/or stopped counting (e.g., “Could you please count faster,” “That is incorrect, please start counting again from 2043”). Immediately following the stress induction, participants were asked to complete a single-item numerical analogue scale that assessed their perceived task difficulty on a scale from 1 to 10, followed by the same numerical analogue scale used previously. Participants were then instructed to wait for 10 min before starting a “second, more difficult trial of the counting task.” By misleading participants to expect a second trial, we hoped to distinguish between experienced, and anticipatory stress and anxiety effects (Spinella et al., 2021). Following the 10-min waiting period, participants completed the numerical analogue scale for a fourth time and were then informed that they are no longer required to complete the second trial of the stress task. At this point, participants were asked to complete a series of concluding questions that would assess whether they believed the verbal information provided by experimenters about the CBD content of the oil they received (told CBD vs. told CBD-free). At the end of the study, participants received a follow-up call to debrief them about the true aims of the study, including the use of deception and the rationale for its use. During debriefing, participants were informed of the actual CBD content of the oil they received during the study (CBD-free) and were given an opportunity to ask questions, and to withdraw their data, if desired. All participants were compensated $15 per hour for their participation. The study received ethical approval from the Nova Scotia Health Authority Research Ethics Board (File number 1027181).

Statistical analysis

We were unable to conduct a precise power analysis using Marginal Linear Models as this would require knowledge about the specific within-subject covariance structure of a given model. However, given that this information is empirically derived from the models themselves (Gueorguieva and Krystal, 2004), we based our sample size estimates on a power calculation for a similar, but less powerful analytic approach (repeated-measures analyis of variance) using G*Power. Assuming an effect size of f² = 0.35, a correlation of r = 0.5 among time points, and an alpha level of 0.05, we required a total of 44 participants to achieve a power of 0.8.

Data were analyzed using mixed models in the Statistical Package for the Social Sciences version 26 (SPSS Inc., Chicago, IL, USA). Model simplicity and likelihood ratio tests were used to select appropriate covariance structures. Main outcome measures included changes in subjective state (e.g., stress, anxiety) and heart rate frequency. Data for each main measure were analyzed using time (10-min post-oil consumption (PO), immediately post-stressor (PS), and following a 10-min anticipation period (ANT)) as a fixed and repeated factor and Expectancy condition (Told CBD, Told CBD-free) as a fixed factor, with baseline scores entered as time-varying covariates.

Results

Sample characteristics

Of the 48 participants recruited, one female (randomized to the told CBD-free condition) failed to attend her scheduled session and two participants (one male, one female; both randomized to the told CBD-free condition) were excluded from the analyses due to not believing the CBD content information provided by the blinder, resulting in a final sample of 45 (22 females, 23 males; 24 told CBD, 21 told CBD-free). The physiological data of two participants were excluded due to ectopic beats exceeding 5% of total beats during baseline, which prevented us from running the analysis where baseline serves as a covariate. Furthermore, the STAI-T score of one participant was excluded due to missing data. A summary of participant characteristics can be found in Table 1. Importantly, there were no significant differences between the two expectancy groups (told CBD vs told CBD-free) on measures of trait anxiety, perceived stress, or other individual difference factors that could systematically bias the present findings (e.g., a priori beliefs about CBD’s stress- and anxiety-relieving effects, task difficulty ratings, CO levels; see Table 1).

Table 1.

Sample characteristics and between-group comparisons.

Variable		Told CBD	Told CBD-free	t (df)	p
Variable		Mean (SD)	Mean (SD)	t (df)	p
Age		20.6 (1.95)	21.7 (3.32)	1.36 (43)	0.180
CO levels (ppm)		4.1 (3.11)	2.9 (1.55)	−1.58 (43)	0.123
STAI-T		41 (11.15)	38.8 (10.24)	−0.69 (42)	0.495
PSS		15.8 (6.28)	16.4 (7.03)	0.3 (43)	0.744
CBD beliefs
Reduces stress		7.8 (1.38)	7.3 (1.98)	−0.91 (43)	0.369
Reduces anxiety		7.3 (1.89)	7.2 (1.99)	−0.1 (43)	0.919
THC beliefs
Reduces stress		7.0 (2.10)	7.6 (1.53)	1.19 (43)	0.240
Reduces anxiety		6.2 (2.59)	6.0 (2.24)	−0.29 (43)	0.776
Past-week cannabis use		1.5 (2.04)	1.8 (2.14)	0.35 (43)	0.726
Task difficulty		7.1 (1.43)	7.4 (1.85)	0.48 (43)	0.633
Variable	Total (%)	Told CBD	Told CBD-free (% of condition)	X² (df)	p
Variable	Total (%)	(% of condition)	Told CBD-free (% of condition)	X² (df)	p
Sex (female)	48.9%	50.0%	47.6%	0.25 (1)	0.873
Ethnicity (White)	80.0%	83.3%	76.2%	0.36 (1)	0.550

STAI-T scores have possible values from 20 to 80; PSS scores have a possible values of 0–40; Past-week Cannabis use is defined as the number of days cannabis was used in the past week; CBD beliefs, THC beliefs, and Task Difficulty were rated on a scale from 1 (Not at All) to 10 (Completely or Extremely).

STAI-T: state-trait anxiety inventory; PSS: perceived stress scale.

Dependent measures

Marginal linear models were used to examine Time, Expectancy, and Time by Expectancy effects for subjective stress, anxiety, sedation, and energy. Because our a priori-determined analytic strategy involved including baseline measures as covariates, “raw” means and standard errors are presented in Table 2, while Figure 1 presents the marginal means and standard errors associated with our analyses.

Table 2.

Raw means and standard errors for all outcomes across time points.

Outcome measure	Time	Told CBD, Mean (SE)	Told CBD-free, Mean (SE)
Stress	Baseline	3.4 (0.41)	3.3 (0.49)
	Post-oil	2.6 (0.36)	2.8 (0.46)
	Post-stress	4.3 (0.43)	5.0 (0.50)
	Anticipation	2.7 (0.34)	3.3 (0.45)
Anxiety	Baseline	2.9 (0.44)	3.4 (0.50)
	Post-oil	2.4 (0.37)	2.7 (0.48)
	Post-stress	3.9 (0.40)	4.1 (0.57)
	Anticipation	2.3 (0.29)	3.2 (0.45)
Sedation	Baseline	2.2 (0.33)	2.2 (0.36)
	Post-oil	3.6 (0.41)	2.6 (0.46)
	Post-stress	3.3 (0.37)	2.8 (0.51)
	Anticipation	4.0 (0.42)	2.6 (0.44)
Energy	Baseline	5.1 (0.27)	5.4 (0.43)
	Post-oil	4.6 (0.35)	4.5 (0.45)
	Post-stress	4.9 (0.38)	5.0 (0.52)
	Anticipation	4.4 (0.31)	4.4 (0.42)
Heart rate	Baseline	79.5 (2.79)	76.1 (2.04)
	Post-oil	79.0 (2.52)	77.5 (2.00)
	Post-stress	76.6 (2.55)	77.1 (2.28)
	Anticipation	77.3 (2.62)	76.4 (2.01)

Subjective outcomes were rated on a scale from 1 (not at all) to 10 (extremely); Raw ECG signals were used to compute discrete heart rate values. Heart rate was assessed in 3-min segments at the time of subjective questionnaires.

Figure 1.

Marginal means and standard errors for subjective ratings across time points. Time varying covariates for each model are shown as baseline values (B), followed by post-oil (PO), post-stress (PS), and anticipation (ANT); *p < .05.

We identified significant main effects of expectancy for subjective sedation; those in the CBD expectancy condition reported increased levels of subjective sedation following oil administration, relative to those in the CBD-free expectancy condition (F(1, 41.99) = 7.47; mean difference (MD) = 1.00, p = 0.009). An Expectancy by Time trend was also evident for sedation (F(2, 43) = 2.50; p = 0.094). Examination of pairwise comparisons within each group further revealed increased sedation from PS to ANT in the CBD expectancy condition (MD = 0.69, p = 0.017); however, there were no significant differences for the CBD-free expectancy condition (MD = 0.19, p = 0.525). In terms of subjective energy, no significant main effects of expectancy (F(1, 42.39) = 0.23; MD = 0.20, p = 0.61) or Expectancy by Time interactions (F(2, 43) = 0.24; p = 0.78) were identified; however, a main effect of Time was observed (F(2, 43) = 3.79; p = 0.030), reflecting significantly elevated subjective energy PS relative to ANT (MD = 0.51, p = 0.010) collapsed across CBD expectancy conditions.

Significant main effects of Time were also identified for subjective stress (F(2, 43) = 32.46; p < 0.001) and anxiety (F(2, 43) = 19.58; p < 0.001). Specifically, pairwise comparisons revealed that both subjective stress and anxiety increased from PO to PS (MD = 1.94, p < 0.001; MD = 1.48, p < 0.001) and decreased from PS to ANT (MD = −1.66, p < 0.001; MD = −1.23, p < 0.001). We further identified a main effect trend of Expectancy for stress, in which CBD expectancy was associated with lower ratings of stress, overall, relative to the CBD-free expectancy condition (F(1, 41.94) = 4.00; MD = 0.58, p = 0.053). An examination of pairwise comparisons within each Expectancy condition revealed that in the CBD expectancy condition, subjective stress returned to PO levels at ANT (MD = 0.06, p = 0.78), but remained elevated relative to PO levels at ANT in the CBD-free expectancy condition (MD = 0.50, p = 0.045). However, the overall Expectancy by Time interaction was not significant (F(2, 43) = 0.99; p = 0.38). For subjective anxiety, the main effect of Expectancy was not significant (F(1, 42.3) = 0.208; MD = 0.13, p = 0.651), but an Expectancy by Time effect was identified at the trend level (F(2, 43) = 2.41; p = 0.102). Pairwise comparisons within each condition reveal that both CBD and CBD-free expectancy, increased in ratings of subjective anxiety from PO to PS (MD = 1.52, p < 0.001; MD = 1.43, p < 0.001) and decreased in ratings of anxiety from PS to ANT, although the robustness of these decreases varied between the CBD (MD = −1.56, p < 0.001) and CBD-free (MD = −0.91, p = 0.007) expectancy conditions. Notably, an examination of pairwise comparisons within each expectancy condition revealed that similar to stress, subjective anxiety returned to PO levels at ANT in the CBD expectancy condition (MD = 0.04, p = 0.85), but not the CBD-free condition (MD = 0.52, p = 0.025).

Finally, there were no significant main effects of Expectancy (F(1, 40.3) = 0.18; MD = 0.38, p = 0.68) or Expectancy by Time interactions (F(2, 40.1) = 0.41; p = 0.66) identified for heart rate; however, a main effect of Time was identified (F(2,40.10) = 6.58; p = 0.003). A further breakdown of this effect suggests that heart rate decreased from PO to PS (MD = −2.25, p = 0.001) and from PO to ANT (MD = −1.66, p = 0.007), though no significant changes were observed from PS to ANT.

Discussion

This research examined the effects of CBD expectancy on responses to a cognitive stressor. Consistent with our previous findings (Spinella et al., 2021, 2023), we found that CBD expectancy was associated with heightened levels of sedation and tended to be associated dampened subjective stress and anxiety in anticipation of a stressor. These findings suggest that placebo effects likely contribute to CBD’s purported stress- and anxiety-relieving properties.

Placebo effects occur when an individual perceives drug-related effects in the absence of the drug’s pharmacological actions. Such effects are thought to be derived from an individual’s beliefs and expectations regarding a given substance’s effects, which can be formed by prior experience, observational learning, and exposure to information regarding the content and supposed effects of a substance (Dar and Barrett, 2014; Kirsch 1997, 2018). In the case of CBD, there is growing evidence that many members of the general population tend to hold beliefs that CBD possesses anxiolytic and stress-relieving properties. For example, Tran and Kavuluru (2020) examined 64,099 social media posts and found that CBD was most discussed as a remedy for anxiety disorders. Similarly, Wheeler et al. (2020) surveyed 340 young adults and found that CBD was most likely to be associated with stress relief and relaxation. Moreover, in a recent study by our group (Spinella et al., 2023), both cannabis users and nonusers attributed stress- and anxiety-relieving effects of cannabis to CBD more so than to THC. Importantly, placebo effects tend to be particularly robust for conditions that have a subjective component and for treatments that cannot be easily identified based on their overt physiological actions (Wampold et al., 2005). Thus, one might expect a relatively strong placebo response for a nonintoxicating substance like CBD when used to manage stress or anxiety.

Participants in the CBD expectancy condition tended to report diminished subjective stress following oil administration and appeared to exhibit blunted subjective stress and anxiety when anticipating another stressor, but not in response to the stressor itself. In our previous research, CBD expectancy did not impact subjective stress and anxiety immediately following a more intense stressor, but physiological and endocrine markers of stress were suggestive of a blunted anticipatory stress response (Spinella et al. 2021, 2023). The present findings provide further evidence that stress anticipation may be more sensitive to CBD-related placebo effects relative to acute stress per se. In the present study, we intended to create anticipatory stress/anxiety by informing participants that they would need to complete a second, more difficult task, following the completion of the initial stress challenge. Although previous research that has used the same stress induction procedures as the present study (Wang et al., 2005, 2007) suggests that participants’ subjective state returns to baseline within 12-min following the stress challenge, because our subjective assessments were taken slightly earlier (10-min following the stress challenge), we cannot rule out a faster recovery from the stress task in the CBD expectancy condition as a contributor to these effects.

Our observation that the administration of oil in the CBD expectancy condition was associated with increased subjective sedation replicates and extends the findings reported by Spinella et al. (2021) and is consistent with effects of actual CBD administration that have been previously reported (e.g. Bergamaschi et al. 2011a; Crippa et al., 2004). The consistency of these findings across studies suggests that subjective sedation may be particularly amenable to CBD-related expectancy influences. Because increased sedation is a widely reported side effect associated with CBD administration, and this was noted on the consent forms for each study, it is not clear the extent to which such effects were primed during the consent process versus participants’ a priori beliefs about CBD-related effects. Future research could help clarify this by manipulating information provided to participants about potential CBD-related effects.

Our findings should be interpreted with the following considerations in mind. First, because a goal of our study was to pilot a stress induction procedure that could be used to assess CBD-related expectancy effects in the context of neuroimaging, our experimental protocol was required to be relatively brief; therefore, we were unable to include validated methods of physiological parameters for stress and anxiety such as heart rate variability, cortisol, as well as longer, psychometrically validated subjective measures, which could have helped in the interpretation of our findings (Spinella et al., 2021, 2023). Further, although the relatively lower levels of anticipatory stress and anxiety, as well as the significant increase in mental sedation observed in the told CBD group, compared to the told CBD-free group are consistent with CBD-specific effects reported in previous research (e.g., Bergamaschi et al., 2011a; Crippa et al., 2004; 2011), we did not directly manipulate expectancies about other drugs (e.g., THC) as part of this study. Consequently, we are unable to definitively rule out general drug-related expectancies as potential causal explanations for our findings. Considerations should also be made regarding statistical power. Our sample size was only sufficient to detect relatively large effects using a between-subject design, and it is possible that we were underpowered to detect subtle changes in subjective ratings. Indeed, although several within condition pairwise comparisons were statistically significant, interactions between condition and time tended to be only evident at the trend level and thus should be interpreted with caution. Moreover, the present study was underpowered to examine sex differences, which are important to consider in the evaluation of drug and placebo responses. Our sample was also relatively homogenous, with most participants being university students of primarily European descent. Additionally, our sample consisted of healthy adults, which did not allow us to examine whether our findings would extend to individuals with anxiety- or stress-related conditions. Lastly, our study did not employ a fully balanced-placebo design, which does not allow us to make further inferences regarding CBD expectancy (e.g., in combination with CBD pharmacology, in comparison to CBD pharmacology). Future research using a full balanced placebo design, with conditions involving actual CBD administration, is needed to determine potential CBD expectancy by pharmacology interaction effects on anxiety- and stress-related outcomes.

Conclusion

The present findings replicate and extend previous findings that CBD expectancy alone can impact stress- and anxiety-relevant responses in the laboratory context. Results highlight the importance of explicitly evaluating and controlling for placebo effects when considering potential therapeutic applications of CBD. Future research is necessary to determine the extent to which the pharmacological effects of CBD have independent effects on stress- and anxiety-related processes.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by a Catalyst Grant from the Canadian Institutes of Health Research awarded to SPB, TCS and SHS.

ORCID iDs

Toni C Spinella

Sean P Barrett

References

American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders, 5th edn. Arlington, VA: American Psychiatric Publishing.

Arndt

De Wit

(2017) Cannabidiol does not dampen responses to emotional stimuli in healthy adults. Cannabis Cannabinoid Res 2: 105–113. DOI: 10.1089/can.2017.0014.

Bergamaschi

Queiroz

RHC

Chagas

MHN

, et al (2011a) Cannabidiol reduces the anxiety induced by simulated public speaking in treatment-naive social phobia patients. Neuropsychopharmacology 36: 1219–1226.

Bergamaschi

Queiroz

RHC

Zuardi

, et al (2011) Safety and side effects of Cannabidiol, a Cannabis sativa constituent. Curr Drug Saf 6: 237–249. DOI: 10.2174/157488611798280924.

Berger

Rice

, et al (2022) Cannabidiol for treatment-resistant anxiety disorders in young people: An open-label trial. J Clin Psychiatry 83: 21m14130. DOI: 10.4088/JCP.21m14130.

Blessing

Steenkamp

Manzanares

, et al (2015) Cannabidiol as a potential treatment for anxiety disorders. Neurotherapeutics 12: 825–836. DOI: 10.1007/s13311-015-0387-1.

Campos

Fogaça

Sonego

, et al (2016) Cannabidiol, neuroprotection and neuropsychiatric disorders. Pharmacol Res 112: 119–127. DOI: 10.1016/j.phrs.2016.01.033.

Cohen

Kamarck

Mermelstein

. (1983) A global measure of perceived stress. Journal of Health and Social Behavior 24: 386–396.

Crippa

Derenusson

Ferrari

, et al (2011) Neural basis of anxiolytic effects of cannabidiol (CBD) in Generalized Social Anxiety Disorder: A preliminary report. J Psychopharmacol 25: 121–130. DOI: 10.1177/0269881110379283.

10.

Crippa

Zuardi

Garrido

, et al (2004) Effects of cannabidiol (CBD) on regional cerebral blood flow. Neuropsychopharmacology 29: 417–426. DOI: 10.1038/sj.npp.1300340

11.

Cuttler

Spradlin

(2017) Measuring cannabis consumption: Psychometric properties of the Daily Sessions, Frequency, Age of Onset, and Quantity of Cannabis Use Inventory (DFAQ-CU). PLoS ONE 12(5): e0178194. https://doi.org/10.1371/journal.pone.0178194.

12.

Dahlgren

Lambros

Smith

, et al. (2022) Clinical and cognitive improvement following full-spectrum, high-cannabidiol treatment for anxiety: Open-label data from a two-stage, phase 2 clinical trial. Commun Med 2: 139. DOI: 10.1038/s43856-022-00202-8.

13.

Dar

Barrett

(2014) The effects of beliefs regarding drug assignment in experimental and field studies of nicotine delivery devices: A review. J Psychopharmacol 28: 1071–1079. DOI: 10.1177/0269881114548295.

14.

Davey

Barratt

Butow

, et al (2007) A one-item question with a Likert or visual analog scale adequately measured current anxiety. J Clin Epidemiol 60: 356–360. DOI: 10.1016/j.jclinepi.2006.07.015.

15.

García-Gutiérrez

Navarrete

Gasparyan

, et al (2020) Cannabidiol: A potential new alternative for the treatment of anxiety, depression, and psychotic disorders. Biomolecules 10: 1575. DOI: 10.3390/biom10111575.

16.

Gueorguieva

Krystal

(2004) Move over ANOVA: Progress in analyzing repeated-measures data and its reflection in papers published in the Archives of General Psychiatry. Arch Gen Psychiatry 61: 310. DOI: 10.1001/archpsyc.61.3.310.

17.

Hazekamp

Fischedick

Diez

, et al (2010) Comprehensive Natural Products II: Chemistry and Biology, Essay, Amsterdam, The Netherlands: Elsevier, pp. 1034–1077.

18.

Hundal

Lister

Evans

, et al (2017) The effects of Cannabidiol on persecutory ideation and anxiety in a high trait paranoid group. J Psychopharmacol 32: 276–282. DOI: 10.1177/0269881117737400.

19.

Karekla

Panayiotou

Collins

(2017) Predictors of urge to smoke under stressful conditions: An experimental investigation utilizing the PASAT-C task to induce negative affect in smokers. Psychol Addict Behav 31: 735–743. DOI: 10.1037/adb0000309.

20.

Khoury

De Neves

MCL

Roque

MAV

, et al (2017) Is there a role for Cannabidiol in psychiatry? World J Biol Psychiatry 20: 101–116. DOI: 10.1080/15622975.2017.1285049.

21.

Kirsch

(1985) Response expectancy as a determinant of experience and behavior. Am Psychol 40: 1189–1202. DOI: 10.1037/0003-066x.40.11.1189.

22.

Kirsch

(1997) Response expectancy theory and application: A decennial review. Appl Prevent Psychol 6: 69–79. DOI: 10.1016/s0962-1849(05)80012-5.

23.

Kirsch

(2018) Response expectancy and the placebo effect. Int Rev Neurobiol 138: 81–93. DOI: 10.1016/bs.irn.2018.01.003.

24.

Kirschbaum

Pirke

Hellhammer

(1993) The ‘Trier Social Stress Test’—A tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology 28: 76–81. DOI: 10.1159/000119004.

25.

Lesage

Berjot

Deschamps

(2012) Clinical stress assessment using a visual analogue scale. Occup Med 62: 600–605. DOI: 10.1093/occmed/kqs140.

26.

Lovibond

(1995) Manual for the Depression Anxiety Stress Scales, 2nd edn. Sydney, NSW: Psychology Foundation.

27.

MacCallum

Russo

(2018) Practical considerations in Medical Cannabis Administration and dosing. Eur J Intern Med 49: 12–19. DOI: 10.1016/j.ejim.2018.01.004.

28.

Rossi

Pourtois

(2012) Transient state-dependent fluctuations in anxiety measured using STAI, POMS, PANAS, or VAS: A comparative review. Anxiety Stress Coping 25: 603–645. DOI: 10.1080/10615806.2011.582948.

29.

Rueger

King

(2013) Validation of the brief Biphasic Alcohol Effects Scale (B-BAES). Alcohol Clin Exp Res 37:470-6. DOI: 10.1111/j.1530-0277.2012.01941.x.

30.

Smeets

Cornelisse

Quaedflieg

CWEM

, et al (2012) Introducing the Maastricht Acute Stress Test (MAST): A quick and non-invasive approach to elicit robust autonomic and glucocorticoid stress responses. Psychoneuroendocrinol 37: 1998–2008. DOI: 10.1016/j.psyneuen.2012.04.012.

31.

Sobell

Brown

Leo

, et al (1996) The reliability of the alcohol timeline followback when administered by telephone and by computer. Drug Alcohol Depend 42: 49–54. DOI: 10.1016/0376-8716(96)01263-x.

32.

Spielberger

(1983) Manual for the State-Trait Anxiety Inventory STAI (form Y; “self-evaluation questionnaire”). Canada: Consulting Psychologists Press Inc.

33.

Spinella

Bartholomeusz

Stewart

, et al (2023) Perceptions about THC and CBD effects among adults with and without prior cannabis experience. Addict Behav 137: 107508. DOI: 10.1016/j.addbeh.2022.107508.

34.

Spinella

Burdeyny

Oprea

, et al (2023) The impact of cannabidiol (CBD) expectancy on cortisol responsivity in the context of acute stress: Associations with biological sex. Cannabis Cannabinoid Res. Online ahead of print. DOI: 10.1089/can.2022.0326.

35.

Spinella

Stewart

Naugler

, et al (2021) Evaluating cannabidiol (CBD) expectancy effects on acute stress and anxiety in healthy adults: A randomized crossover study. Psychopharmacol 238: 1965–1977. DOI: 10.1007/s00213-021-05823-w.

36.

Stanley

Ferretti

Bonn-Miller

, et al (2022) A double-blind, randomized, placebo-controlled test of the effects of Cannabidiol on experiences of test anxiety among college students. Cannabis Cannabinoid Res 8: 1090–1099. DOI: 10.1089/can.2022.0062.

37.

Stohs

Ray

(2020) Is cannabidiol hepatotoxic or hepato-protective: A review. Toxicol Res Appl 4. DOI: 10.1177/2397847320922944.

38.

Tran

Kavuluru

(2020) Social media surveillance for perceived therapeutic effects of cannabidiol (CBD) products. Int J Drug Policy 77: 102688. DOI: 10.1016/j.drugpo.2020.102688.

39.

Walsh

Babalonis

Rehm

(2018) Cannabidiol (CBD): Critical review report. World Health Organization. Available from: https://www.who.int/our-work/access-to-medicines-and-health-products/controlled-substances (accessed 27 April 2023).

40.

Wampold

Minami

Tierney

, et al (2005) The placebo is powerful: Estimating placebo effects in medicine and psychotherapy from randomized clinical trials. J Clin Psychol 61: 835–854. DOI: 10.1002/jclp.20129.

41.

Wang

Korczykowski

Rao

, et al (2007) Gender difference in neural response to psychological stress. Soc Cogn Affect Neurosci 2: 227–239. DOI: 10.1093/scan/nsm018.

42.

Wang

Rao

Wetmore

, et al (2005) Perfusion functional MRI reveals cerebral blood flow pattern under psychological stress. Proc Nat Acad Sci 102: 17804–17809. DOI: 10.1073/pnas.0503082102.

43.

Wheeler

Merten

Gordon

, et al (2020) CBD (Cannabidiol) product attitudes, knowledge, and use among young adults. Subst Use Misuse 55: 1138–1145. DOI: 10.1080/10826084.2020.1729201.

44.

Zuardi

Cosme

Graeff

, et al (1993) Effects of ipsapirone and Cannabidiol on human experimental anxiety. J Psychopharmacol 7: 82–88. DOI: 10.1177/026988119300700112.

45.

Zuardi

Shirakawa

Finkelfarb

, et al (1982) Action of cannabidiol on the anxiety and other effects produced by Delta-9-THC in normal subjects. Psychopharmacol 76: 245–250.