P.329 State anxiety induced by THC and CBD does not change after correcting for baseline anxiety levels

N. Hutten*, T. Arkell, F. Vinckenbosch, E. Theunissen, K. Kuypers, I. McGregor, J. Ramaekers

*Corresponding author for this work

Research output: Contribution to journalConference Abstract/Poster in journalAcademic


Background: In the last decade, the psychoactive substance delta-9-tetrahydrocannabinol (THC) in cannabis products has risen considerably, which has been associated with increased anxiety [1]. The non-psychoactive cannabis compound cannabidiol (CBD) is suggested to counteract THC-induced anxiety to some extent [2,3], however, these effects are mild [3]. Trait anxiety has been suggested to moderate cannabis-induced anxiety [4]. The present study aimed to examine whether the CBD effects on THC-induced anxiety can be explained by baseline state anxiety levels and/or trait anxiety levels in healthy volunteers. Methods: A placebo-controlled, randomized within-subjects study including 26 healthy recreational cannabis users was set up to test the effects of vaporized cannabis containing THC (11% THC, <1% CBD), CBD (<1% THC, 11% CBD), THC/CBD (11% THC, 11% CBD), and placebo cannabis (<1% THC, <1% CBD) on anxiety measures. Participant ratings of subjective state anxiety using Visual Analogue Scales (VAS) and State-Trait Anxiety Inventory (STAI)-state served as a measure of explicit anxiety, and a computer-based emotional Stroop task was used to assess implicit anxiety. Subjective anxiety state was scored repeatedly over 6 hours, while the STAI-state and emotional Stroop task were administered directly after drug inhalation. Trait anxiety levels (STAI) were measured once prior to the test days and baseline state anxiety level were measured once prior to each drug inhalation. All data were analysed with Linear Mixed Models (LMMs); the VAS ratings of state anxiety were analysed with unstructured covariance structure, and the STAI-state and emotional Stoop data were analysed with compound symmetry as covariance structure. The STAI-trait and anxiety state baseline scores were added as covariates. When a covariate was non-significant, it was removed from the model. In the case of a main Drug effect, planned Bonferonni-corrected pairwise comparisons were performed. Results: There was a main effect of the covariate baseline state anxiety (p< 0.01), a drug by trait anxiety interaction (p= 0.02) and a main Drug effect (p< 0.01) on the subjective state rating of anxiety (VAS). Participants felt more anxious during THC (p< 0.01; 95% CI [-2.02, -0.80]) and THC/CBD (p< 0.05; 95% CI [-1.46, -0.23]) compared to placebo. Both covariates did not show a significant effect on the STAI-state and emotional Stroop and thereby removed from further analyses. There was a main Drug effect on the STAI-state; (p< 0.01) both THC (p< 0.01; 95% CI [-17.20, -9.30]) and THC/CBD (p= 0.03; 95% CI [-10.89, -2.76]) increased state anxiety compared to placebo, however anxiety induced by THC/CBD was less compared to the THC condition (p= 0.02; 95% CI [2.30, 10.54]). There were no significant Drug effect on the emotional Stroop task. Conclusion: The present study shows that trait anxiety partly predicts cannabis induced increments in state anxiety. However, cannabis effects on anxiety are robust as they do not change after correcting for baseline differences in state anxiery. While baseline levels and trait anxiety are both subjective measures, future studies might also include pharmacokinetics and pharmacodynamics modelling in order to include individual differences in metabolism which might explain the cannabis-induced anxiety.
Original languageEnglish
Pages (from-to)S60-S61
Number of pages2
JournalEuropean Neuropsychopharmacology
Early online date25 Feb 2021
Publication statusPublished - Mar 2021

Cite this