Lifetime use of psychedelics is associated with better mental health indicators during the COVID-19 pandemic

Psychedelic compounds such as mescaline, psilocybin and N,N-dimethyltryptamine act principally as 5-HT2A receptor agonists and have long histories of ceremonial and medicinal use. Interest in their psychological effects revived after early clinical work in the mid-20th century, but research was largely curtailed following classification of many psychedelics as Schedule 1 substances amid concerns about long-term harms. Subsequent epidemiological and experimental studies have not supported a clear association between lifetime psychedelic use and increased rates of mental illness, and recent clinical research has renewed interest in their potential to produce durable psychological change and therapeutic benefit, particularly when the acute experience has mystical-type features. Cavanna and colleagues set out to examine whether lifetime psychedelic use was associated with mental health indicators during the COVID-19 pandemic, a broadly shared environmental stressor. Specifically, the investigators compared anxiety, positive and negative affect, well-being and resilience between people reporting past psychedelic use and those reporting other psychoactive drug use, and explored whether lifetime psychedelic exposure related to personality traits and higher-order personality factors that might confer resilience. The study was conducted in Argentina during the early months of the pandemic and lockdown measures, a context expected to depress population-level measures of well-being.

The researchers ran an anonymous, internet-based survey in two parts between April and June 2020. Recruitment used social media; both parts were presented separately (one focusing on drug use and mental health scales, the other on personality) and were completed in Spanish. Inclusion criteria for analysis were Argentinian residency, age over 18, male or female gender identity, and provision of online informed consent. Ethical approval was obtained from the ethics committee of Centro de Educación Médica e Investigaciones Clínicas Norberto Quirno. The first questionnaire collected sociodemographic information and lifetime use of a range of psychoactive compounds, together with the State-Trait Anxiety Inventory (STAI), the Positive and Negative Affect Schedule (PANAS), the Psychological Well-being Scale (BIEPS) and a local adaptation of the Resilience Scale (RS). The second questionnaire administered the Big Five Inventory (BFI). Psychometric scales were scored using participants’ responses, gender and age relative to regional normative values; scores were then transformed to a uniform scale (described in the paper as z-scores) to facilitate comparison. Data from the two parts were merged via a unique identifier. Descriptive statistics and Pearson correlations between scales/subscales were computed to characterise relationships among measures. To address poly-drug use and reduce the number of statistical tests, the investigators used principal component analysis (PCA; via singular value decomposition) on a binary matrix of lifetime use for each drug. PCA yielded interpretable components grouping drugs by co-occurrence patterns. Participants were allocated to mutually exclusive groups based on component scores: "recreational" (e.g., LSD, MDMA, psilocybin and dissociatives), "entheogen" (serotonergic psychedelics often used ceremonially such as DMT, ayahuasca, San Pedro, bufo variants and 5-MeO-DMT), "prescription" (antidepressants, antipsychotics, sedatives), and a "legal/non-user" group (only legal drugs such as caffeine, alcohol, tobacco). Participants whose PCA scores did not clearly place them in a group (11.36% of the sample) were excluded from group comparisons. Group differences were tested with ANOVA followed by pairwise Student’s t-tests, with Bonferroni correction applied where stated.

Of 11,365 respondents to the first survey and 157,101 to the second, 10,722 completed both parts. After excluding 5,104 participants who did not meet inclusion criteria, the final analysed sample comprised 5,618 individuals (mean age 29.15 ± 0.12 years; 71.97% female). A total of 32.43% reported at least one lifetime use of a psychedelic drug. When psychometric scores were transformed relative to regional norms, the sample as a whole showed reduced positive indicators compared to the normative average: resilience notably below the regional mean (reported as −2.74), well-being −0.74, and PANAS positive affect −0.54. Conversely, STAI state anxiety (0.30), STAI trait anxiety (0.28) and PANAS negative affect (0.06) were above the regional averages, consistent with an adverse effect of the pandemic and lockdown on mental health. Pairwise correlations among scales behaved as expected: resilience and well-being correlated positively and both correlated negatively with state/trait anxiety, negative affect and neuroticism. The alpha factor (a composite of agreeableness, conscientiousness and inverse neuroticism) correlated positively with healthy indicators, while the beta factor (a composite of extraversion and openness) correlated with well-being and resilience. Openness and agreeableness showed limited correlations with other scales. Initial drug-by-drug comparisons (unadjusted for poly-drug use) suggested a double dissociation: certain psychedelics (principally psilocybin, LSD and, to a lesser degree, ayahuasca) were associated with lower state/trait anxiety and negative affect and higher well-being, positive affect and resilience-related subscales. By contrast, non-psychedelic drugs (including prescription medications, caffeine, alcohol, tobacco and cannabis) were associated with higher scores on measures linked to mental health impairment and lower well-being indicators; MDMA was an exception within this group, being linked to higher scores on the social ties subscale of well-being. PCA-derived groups (recreational, entheogen, prescription, legal/non-user) produced significant group effects (ANOVA P < 0.05, Bonferroni corrected for four comparisons) across most BFI scores and psychometric scales, with the exception of agreeableness (P = 0.121). Key group differences included higher extraversion and lower conscientiousness in the recreational group compared with non-users; increased openness in both recreational and entheogen groups, with entheogen users showing higher openness than recreational users; and elevated neuroticism and lower mental-health-related measures in the prescription group relative to others. Positive affect was higher in the entheogen group compared with all other groups. The alpha factor was lower in the prescription group compared with the other groups, whereas the beta factor was higher in both recreational and entheogen groups. Importantly, the reported number of lifetime psychedelic experiences correlated positively with beta factor scores, indicating a dose‑related association between lifetime psychedelic exposure and this higher-order personality dimension.

Cavanna and colleagues interpret their findings as failing to support an association between lifetime psychedelic use and impaired mental health; rather, they report relationships consistent with enhanced traits and states linked to resilience among psychedelic users during the COVID-19 crisis. The study replicated expected associations among personality traits, affect and measures of well-being and resilience, and used PCA to disentangle overlapping patterns of poly-drug use. Users classified in recreational and entheogen groups exhibited higher openness and, in some cases, higher extraversion; both groups also showed elevated beta factor scores, which the authors discuss as indexing "plasticity" or a tendency toward self‑expansion. The entheogen group in particular showed higher positive affect, whereas the prescription group displayed greater neuroticism and lower scores on several healthy‑mental‑state indicators. The investigators highlight two main interpretations for the link between lifetime psychedelic use and personality differences: first, repeated pharmacological activation of 5-HT2A receptors could induce lasting neurobiological or gene-expression changes that alter personality; second, psychedelic use might be leveraged psychotherapeutically to promote adaptive personality change, either as a direct therapeutic goal or as an intermediate step to treat disorders such as depression. The authors also outline a hierarchical conceptual model in which stable factors (personality, trait anxiety) shape cognitive styles that in turn influence transient emotional states, with cumulative effects on perceived well-being and resilience. Several limitations acknowledged by the authors temper causal inference. Data were self-reported and could not be independently verified, including the identity and dose of substances consumed. Unmeasured confounders may influence associations despite the PCA approach used to mitigate poly-drug effects. Most importantly, the cross-sectional survey design precludes conclusions about causality or temporal direction: it remains possible that pre-existing personality differences predispose individuals to both psychedelic use and better coping during stress. The authors recommend longitudinal or experimental studies to determine whether observed associations reflect long-term effects of psychedelics and to clarify implications for therapeutic applications.