This naturalistic longitudinal study (n=66) investigates the long-term effects of ayahuasca on mental health in adults with no prior exposure. Participants attending neo-shamanic ceremonies reported sustained improvements in depression, anxiety, stress, affect, personality traits, spirituality, and relationships up to 12 months. Individuals with depression or anxiety experienced lasting symptom reductions, while those without a diagnosis had short-term benefits. Alcohol and cannabis use decreased only at one month. Findings suggest ayahuasca's mental health benefits persist, with varying trajectories of change over time.
The durability of ayahuasca’s effects on mental health and the influence of clinical diagnoses on therapeutic response is unclear. Adults with no prior exposure to ayahuasca (n = 66) participating in neo-shamanic ayahuasca ceremonies completed questionnaires at baseline, 7 days, and 1, 6, and 12 months. Mixed models were used to characterize temporal trajectories in mental health, alcohol and cannabis use, affect, personality, spirituality, and relationships and examine the longevity of effects in individuals with and without a depressive or anxiety disorder. After multiple comparison correction, ayahuasca use was associated with decreases in depression, anxiety, stress, negative affect, negative emotionality, acceptance of external influence, and self-alienation at all time points. Improvements in mental health, self-efficacy, and spirituality were observed up to 12 months post-ceremony. Individuals with depression and anxiety diagnoses maintained significant symptom reductions, whereas those without a diagnosis experienced short-term benefits. Decreases in alcohol and cannabis use were only observed at month 1. Naturalistic ayahuasca use was associated with persisting improvements in mental health and wellbeing, with the largest magnitude of symptom reduction observed in those diagnosed with a depressive or anxiety disorder. Differing trajectories of change were identified across psychological constructs, suggestive of both enhancement and attenuation of gains over time.
Ayahuasca is a plant-based psychedelic brew containing N,N-dimethyltryptamine (DMT) and harmala alkaloids that enable oral activity. Earlier clinical trials, prospective observational studies, and large cross-sectional surveys have reported rapid and sometimes durable reductions in depression and anxiety symptoms following ayahuasca use, as well as improvements in life satisfaction, mindfulness, cognitive flexibility, empathy, and connectedness. Nevertheless, uncertainty remains about how long psychological changes persist after a single exposure, whether negative long-term consequences occur, and whether therapeutic benefits differ between people with and without formal psychiatric diagnoses. Pagni Bap and colleagues set out to characterise trajectories of mental health, substance use, wellbeing, personality, spirituality, body connection, and interpersonal relationships over 12 months in ayahuasca‑naive adults who participated in neo‑shamanic ceremonies. A further aim was to examine whether lifetime diagnoses of depressive or anxiety disorders influence the longevity of antidepressant or anxiolytic effects. The study therefore provides longitudinal naturalistic data on first‑time ayahuasca users to inform questions about durability and diagnostic moderators of effect.
Papers cited by this study that are also in Blossom
Bathje, G. J., Fenton, J., Pillersdorf, D. et al. · Journal of Humanistic Psychology (2021)
Bouso, J. C., González, D., Fondevila, S. et al. · PLOS ONE (2012)
Maia, L. O., Massarentti, C. M., Tófoli, L.F. · Psychopharmacology (2022)
Jiménez-Garrido, D. F., Gómez-Sousa, M., Ona, G. et al. · Scientific Reports (2020)
This prospective observational study recruited 66 ayahuasca‑naive adults by convenience sampling from two independent ayahuasca churches in North America between 2019 and 2021. Eligible participants were aged 18 or older, English speaking, and had never used ayahuasca. Church intake procedures (independent of the research team) excluded attendees with confirmed or suspected psychotic disorders and those on contraindicated medications. The research team did not run or oversee the ceremonies, which were 1‑ to 3‑day neo‑shamanic, mestizo‑styled events led by facilitators at each site. All participants provided written informed consent and the protocol received university ethics approval. Outcome measures were collected at baseline, 7 days, 1 month, 6 months, and 12 months for many instruments. Primary mental health instruments included the DASS‑21 (Depression, Anxiety, Stress Scale) and PANAS (Positive and Negative Affect Schedule). The SF‑12 measured general mental and physical health. Alcohol and cannabis use were assessed with the ASSIST and three AUDIT items. Wellbeing and personality measures administered at baseline, 1, 6 and 12 months included the General Self‑Efficacy Scale (GSE), Barratt Impulsivity Scale (BIS‑30), Big Five Inventory‑2 short form (BFI‑2‑XS), Relationship Satisfaction Scale (RS), Authenticity Scale (AS), Scale of Body Connection (SBC), Nature Relatedness Scale (NR), and the Intrinsic Spirituality Scale (ISS). Height/weight and fruit/vegetable intake were also collected. Statistical analysis was performed in R using Linear Mixed Models (LMM) for continuous outcomes and Cumulative Link Mixed Models (CLMM) for ordinal outcomes, with participant set as a random effect and time and post‑ceremony ayahuasca use as fixed effects. LMMs were chosen in part because they handle partially missing data and reduce bias from attrition. The impact of predictors was assessed using ANOVA from LMMs with Satterthwaite degrees of freedom. Effect sizes were reported as partial eta‑squared (ηp2), with thresholds defined in the paper (small 0.01–0.06; medium 0.06–0.14; large ≥0.14). Multiplicity across 42 measures was controlled using false discovery rate (FDR) correction (Benjamini–Hochberg), and significant time effects were followed by Holm‑Bonferroni adjusted post‑hoc pairwise comparisons reported as estimated marginal means and mean differences. Odds ratios were estimated for ordinal models where relevant.
Baseline characteristics and retention: The baseline sample comprised 66 participants (26 men, 40 women), mean age 37.56 years (SD 10.16, range 19–64). Most were US citizens (97%) and nearly half had a bachelor’s degree or higher. Twenty‑four participants reported a lifetime depressive disorder and 15 an anxiety disorder. During follow‑up, 27 participants reported additional ayahuasca use while 39 did not. Retention for primary mental health measures was 81.8% at 1 month and 63.6% at 12 months. Primary mental health outcomes: Significant main effects of time were observed on all DASS‑21 subscales and survived FDR correction (Depression ηp2 = 0.20; Anxiety ηp2 = 0.16; Stress ηp2 = 0.33; p‑FDR < 0.001). Across the full sample depressive symptoms were reduced by 75% at 7 days, 59% at 1 month, and 51% at 12 months relative to baseline. Anxiety and stress reductions across the sample were 57% and 63% at 7 days, 55% and 54% at 1 month, 50% and 36% at 6 months, and 47% and 46% at 12 months, respectively. Symptoms of depression and stress were higher at 6 months than at 7 days, indicating some attenuation of early gains. Effects by diagnostic history: Including lifetime diagnosis in models showed main effects of diagnosis for DASS‑21 depression (ηp2 = 0.07) and anxiety (ηp2 = 0.08), with those diagnosed having higher baseline symptoms. Importantly, participants with a lifetime depressive or anxiety disorder maintained significant symptom reductions at all follow‑up points; at 12 months reductions were approximately 60% for depression and 64% for anxiety. Those with diagnoses improved from a moderate to a normal range and closely approximated symptom levels of participants without a diagnosis by follow‑up. Affect and general mental health: PANAS Negative Affect showed a significant time effect (ηp2 = 0.27, p‑FDR < 0.01), with reductions of 33% at 7 days, 32% at 1 month, 21% at 6 months, and 27% at 12 months. Positive Affect increased significantly only at 7 days (ηp2 = 0.09). The SF‑12 Mental Component Score had a significant time effect (ηp2 = 0.21, p‑FDR < 0.001), with improvements at 1, 6 and 12 months; physical health scores did not change. Substance use: ASSIST alcohol and cannabis risk scores showed main effects of time (alcohol ηp2 = 0.16; cannabis ηp2 = 0.20) that survived correction, but reductions relative to baseline were observed only at 1 month and not at later time points. AUDIT frequency of alcohol consumption decreased at 1 and 6 months but not at 12 months; other AUDIT items did not show robust changes after correction. Personality and wellbeing: For BFI traits, Negative Emotionality showed a robust decrease over time (ηp2 = 0.17, p‑FDR < 0.001), with smaller or non‑surviving effects for Agreeableness (ηp2 = 0.06) and Open‑Mindedness (ηp2 = 0.06). Self‑efficacy (GSE) increased (ηp2 = 0.12) at 1 and 12 months. No significant time effects emerged for overall impulsivity (BIS). Relationship satisfaction with friends improved at 12 months (friends ηp2 = 0.08, p‑FDR = 0.042); partner’s parents showed improvement that did not survive correction. On the Authenticity Scale, External Influence (ηp2 = 0.09) and Self‑alienation (ηp2 = 0.17) decreased across follow‑up, with Self‑alienation reduced by about 50% at 12 months; Authentic Living increases did not survive correction. Body Dissociation decreased at 1 month (SBC subscale ηp2 = 0.10) while Body Awareness did not change. Nature Relatedness subscales showed selective effects: Experience increased at 12 months (ηp2 = 0.11, p‑FDR = 0.032) and Self increased at 6 months but did not survive correction. Spirituality (ISS) increased at 1, 6 and 12 months (ηp2 = 0.12, p‑FDR = 0.003). Lifestyle measures (BMI, fruit and vegetable intake) did not show significant time effects overall. Effects of further ayahuasca use: Models detected differences between those who did and did not report additional ayahuasca use for several outcomes (negative emotionality, relationship satisfaction, fruit/vegetable consumption). However, most between‑group differences reflected baseline differences. The only significant within‑group change from baseline attributed to non‑consumers was greater parental relationship satisfaction at 12 months. Participants who consumed more ayahuasca reported higher vegetable intake and higher parental relationship satisfaction at baseline and early follow‑up. Effect sizes: Large effect sizes were reported for depression, anxiety, stress (DASS) and SF‑12 MCS. Large effects were also observed for several negative mood related constructs (PANAS Negative Affect, BFI Negative Emotionality), body dissociation and self‑alienation, ASSIST alcohol and cannabis risk scores, self‑efficacy, and relationships with partner’s parents. Medium effect sizes were reported for positive affect, agreeableness and openness (BFI), aspects of authenticity, nature relatedness, spirituality, and relationships with friends.
Pagni Bap and colleagues interpret the findings as evidence that a single course of naturalistic ayahuasca consumption in first‑time users was associated with substantial improvements in mental health and multiple wellbeing domains that persisted up to 12 months. The authors highlight as novel the differential durability of symptom change by diagnostic history: individuals with a lifetime depressive or anxiety disorder maintained larger and longer reductions in symptoms than those without a diagnosis. They also note heterogeneous temporal patterns across measures, with some effects emerging early then attenuating (for example, positive affect and body dissociation), others appearing later (nature relatedness, relationship satisfaction), and several remaining stable across the year (negative emotionality, external influence, self‑alienation, spirituality), suggesting possible trait‑level change. The discussion situates the results within prior clinical and naturalistic literature reporting antidepressant and anxiolytic effects of ayahuasca from days to months after use, and acknowledges mixed long‑term findings from earlier observational work. The authors argue that concurrent improvements in constructs linked to adaptive functioning — lower negative emotionality, higher self‑efficacy, reduced self‑alienation, greater relationship satisfaction and nature relatedness — may underlie sustained mental health gains. Limitations acknowledged by the authors include the absence of a control group, which leaves open the contribution of expectancy, natural recovery, and time; selection and attrition bias that limit generalisability; reliance on self‑report measures and lack of expectancy or objective measures; and incomplete data on total frequency of ayahuasca use. They note that some null findings (for example in lifestyle and impulsivity) may reflect floor effects, small sample size, or limited power. To address causal mechanisms and mediators, the authors call for randomised clinical trials with objective outcomes and thorough measurement of moderators and mediators to disentangle temporal dynamics and confirm whether observed changes represent durable trait modifications.
The study identified long‑term improvements in mental health and wellbeing outcomes following naturalistic ayahuasca consumption among first‑time users who had relatively poor mental health at baseline.
Ayahuasca is a psychedelic plant decoction that has emerged as a promising candidate for the psychiatric treatment of mental health conditions. Ayahuasca contains the psychoactive compound N,N-dimethyltryptamine (DMT), and several harmala alkaloids which render DMT orally bioavailable. Clinical trials, prospective observational studies, and large cross-sectional surveys of ayahuasca users suggest that ayahuasca use can lead to rapid, robust, and enduring improvements in symptoms of depression and anxiety. In a double-blind clinical trial,randomized patients with major depressive disorder (MDD) to ayahuasca or an active placebo with purgative effects. Significant antidepressant response was observed 7 days after dosing, with remission rates of 64% in the ayahuasca group and 27% in the placebo group. Prospective observational studies have found reductions in anxiety and depression symptoms out to 6-months in mixed samples, and reductions in depression in patients with MDD out to 12-months after administration, but a separate study did not find therapeutic effects 4-7 years later. Cross-sectional research has found only slight, or nil associations between the number of years since ayahuasca was last drunk and better mental health and reduced alcohol/drug/tobacco use, suggesting effects may be highly durable or diminish at a very slow rate over longer time scales. Other clinically relevant changes have been observed, including increases in life satisfaction, mindfulness, cognitive flexibility, empathy, and connection to self, others, and nature, and alterations to personality traits, which are consistent with enhanced resiliency. Corroborating these relationships, crosssectional survey studies have found that regular ayahuasca users report better general well-being and cognitive performance, fewer diseases and symptoms of psychopathology, and healthier lifestyles relative to normative samples. Despite the health benefits reported in these studies, it is unclear how long psychological changes persist over time, whether negative long-term consequences arise after the purported short-term benefits, and whether effects are specific to individuals with clinical diagnoses. We previously published an analysis of data from a prospective observational study which identified increases in relationship satisfaction, nature relatedness, spirituality, authenticity, selfefficacy, and body connection 1 month after a neoshamanic ayahuasca ceremony in first-time users of ayahuasca. In this report, we characterize mental health and psychological outcomes and trajectories in this sample over a 12month period. We also evaluate whether clinical diagnosis of a depressive or anxiety disorder impacts the longevity of antidepressant or anxiolytic effects, respectively. This study contributes important longitudinal data to better understand the time course of changes in mental health and psychological wellbeing associated with naturalistic ayahuasca use.
Ayahuasca naïve participants (n = 66) were recruited using convenience sampling from two independent ayahuasca churches in North America from 2019 to 2021. Inclusion criteria for participating in the study were being 18 years of age or older, an English speaker, and having never used ayahuasca in the past. Seefor a detailed description of recruitment and enrollment procedures. The research team was not involved in recruitment, administration of ayahuasca, or conducting the ceremony -all of which were performed by the ayahuasca church sites. All participants provided written informed consent. The study protocol was approved by the University of Melbourne Human Research Ethics Committee (HREC number 1,852,071.1). As part of the churches' intake processes, independent of this study, all attendees with a confirmed or suspected diagnosis of a psychotic disorder or using a contraindicated medication use (e.g., antipsychotics, antidepressants, etc.) were excluded.
Ayahuasca ceremonies were conducted in a neoshamanic mestizo-styled setting at both church sites between November 2018 and December 2019. The churches were not organizationally affiliated. At each site, participants attended a 1-day, 2-day, or 3-day ceremony led by a facilitator. For more information on the ceremonial setting, see.
Participants reported lifetime mental health diagnoses, including depression and anxiety disorders. The DASS-21 (Depression, Anxiety, and Stress Scale) and Positive and Negative Affect Scale (PANAS) were administered at baseline (BL), day 7 (D7), 1 month (1 m), 6 months (6 m), and 12 months (12 m). The DASS-21 is a 21-item survey with subdomain scores for depression, anxiety, and stress. The PANAS is a 10-item survey with subscale scores for positive and negative affects. The Short-Form Health Survey (SF-12) was administered at BL, 1 m, 6 m, and 12 m, and features a mental health and physical health summary score.
The 8-item World Health Organization Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) was used to calculate a total risk score for alcohol and cannabis use at BL, 1 m, 6 m, and 12 m. Three categorical items were used in the Alcohol Use Disorder Identification Test (AUDIT) that evaluated the 1) frequency of alcohol use, 2) average amount of alcohol consumed on adrinking day, and 3) frequency of binge-related drinking episodes at BL, 1 m, 6 m, and 12 m.
All wellbeing measures were collected at BL, 1 m, 6 m, and 12 m. The 10-item General Self-efficacy Scale (GSE) total score was used to measure general perceived self-efficacy; the 30-item Barratt Impulsivity Scale (BIS-30) total scores was used to measure impulsivity; the 15-item short form Big Five Inventory-2 (BFI-2-XS) was used to measure the personality traits Extraversion, Agreeableness, Conscientiousness, Negative Emotion, and Open-mindedness. Quality of interpersonal relationships was measured using the 7-item Relationship Satisfaction Scale (RS). Interpersonal and intrapersonal influences were measured using the 12-item Authenticity Scale (AS), which contains the subscales Authentic Living, External Influence, and Self-alienation. Bodily awareness was measured using the 20-item Scale of Body Connection (SBC), which contains the subscales Body Awareness and Bodily Dissociation). Relationships with nature were measured using the 21-item Nature Relatedness Scale (NR), which contains the subscales Self, Perspective, and Experience. Spirituality was measured using the 6-item Intrinsic Spirituality Scale (ISS). Fruit and vegetable consumption was collected with the question "Approximately how many servings of vegetables/fruit do you eat on a usual day?." Body Mass Index was calculated using the formula: weight (kg)/ [height (m)] 2 .
Data were collected and managed using the REDCap data collection tool hosted at the University of Melbourne. All records were imported into R statistical software version 4.1.2 where data analysis was conducted. Linear Mixed Models (LMM) through the lmerTest package were used to test differences between groups and between time points within groups for all instruments unless stated otherwise. In all LMM models, an individual was set as a random effect. Fixed effects included time point and whether an individual had consumed ayahuasca after the initial treatment period. The impact of each predictor was assessed by generating ANOVA results from the LMM with Satterthwaite's method for estimating degrees of freedom. LMMs were utilized due to being well-suited to handling partially missing data, thus reducing potential bias due to differences between responders and non-responders. To analyze results involving ordinal scales, Cumulative Link Mixed Models (CLMM) were employed, utilizing functionalities from the ordinal and RVAideMemoire packages in R. The model implemented used the same fixed and random effects as LMM. Furthermore, odds ratios (ORs) were calculated using the ordinal package. Effect sizes were measured by using effect size R package and partial eta-squared (η p 2 ) with 95% confidence intervals (Ben-Shachar, Lüdecke, and Makowski 2020). Effect sizes are considered small from 0.01 to 0.06, medium from 0.06 to 0.14, and large 0.14 or above. Multiplicity across measures was controlled for with false discovery rate (FDR) correction using the Benjamani-Hochberg method. FDR correction was performed using the p-values from the main effects of time (the primary effect of interest) for all 42 measures. Post-hoc pairwise comparisons using Holm-Bonferroni's adjustment were conducted to examine significantly simple effects of time. Those comparisons were performed using R's emmeans package, and the results are reported with estimated marginal means (EMM), mean difference scores (M diff ), and their corresponding standard error (SE). P-values less than 0.05 after the FDR correction were considered significant.
The sample sizes for the linear mixed models for each respective instrument and time point are presented in Supplemental Table. The baseline sample size was 66 participants (26 men and 40 women) with a mean age of 37.56 (SD: 10.16, range 19-64). Among this sample, the vast majority of participants were US citizens (97%), with relatively high levels of education: almost half (49.9%) reported a bachelor's degree or above (bachelor's degree, 30.3%; a masters or other professional degree, 12.1%; doctoral degree, 6.1%). Twenty-four participants reported previously receiving a depressive disorder diagnosis (e.g., major depression) and 15 reported receiving an anxiety disorder diagnosis (e.g., generalized anxiety disorder) (Table). During the follow-up period, 27 participants reported additional instances of drinking ayahuasca, and 39 participants did not.
Significant main effects of time were detected on the DASS-21 subscales for Depression (η p 2 = 0.20), Anxiety (, η p 2 = 0.16), and Stress (η p 2 = 0.33), all of which survived correction (p-FDR <0.001). Relative to the baseline, depressive symptoms were reduced across the full sample by 75% at 7d, 59% at 1 m, and 51% at 12 m (Figure, Table). Relative to baseline, anxiety, and stress symptoms were reduced across the full sample by 57% and 63% at 7d, 55% and 54% at 1 m, 50% and 36% at 6 m, and 47% and 46% at 12 m, respectively (Figure(c,e) Table). Symptoms of depression and stress were statistically higher at 6 m compared to 7d, suggesting a temporary attenuation of therapeutic gains (Figure(a,e), Table). When the categorical variable for the lifetime history of a depressive disorder was included in the model for the DASS-21 depression subscale, a main effect of diagnosis was detected (η p 2 = 0.07). When the categorical variable for the lifetime history of an anxiety disorder was included in the model for the DASS-21 anxiety subscale, a main effect of diagnosis was detected (η p 2 = 0.08). In these two models, the main effects of time were retained for depression (η p 2 = 0.24) and anxiety (η p 2 = 0.25). At baseline, those with clinical diagnoses exhibited statistically greater values than those without diagnoses on the DASS-21 depression and anxiety subscales, suggesting that the diagnostic cutoff segregated those with lower and higher baseline symptomatology. In those with a diagnosis of depression or anxiety, reductions in depressive or anxiety symptoms were evident at all follow-up time points relative to baseline, with the 12 m follow-up point showing a 60% and 64% reduction in depression and anxiety, respectively (Figure(b,d), Table). Notably, individuals with either depression or anxiety diagnosis showed significant improvement, with symptoms improving from a moderate to a normal range. They maintained reductions within a normal range throughout the follow-up period, closely approximating the symptom levels of the group without the diagnosis.
Significant main effects of time were detected on the PANAS subscales Negative Affect (η p 2 = 0.27) and Positive Affect (η p 2 = 0.09; Table), which survived correction (p-FDR <0.01). Relative to baseline, Negative Affect was significantly diminished by 33% at 7d, 32% at 1 m, 21% at 6 m, and 27% at 12 m compared to baseline (Figure). Relative to baseline, Positive Affect was only statistically higher at 7d (Figure, Table).
A significant main effect of time was detected on the SF-12 Mental Health Component Score (MCS) (η p 2 = 0.21), but not Physical Health Component Score (p = .37). MCS survived correction (p-FDR <0.001). Relative to baseline, significant improvements in mental health were found at 1 m, 6 m, and 12 m. Additionally, mental health showed a significant drop in scores from 1 m to 6 m but remained significantly higher than baseline at 6 m (Figure, Table).
Significant main effects of time were detected on the ASSIST alcohol risk (η p 2 = 0.16) and cannabis risk scores (η p 2 = 0.20), which survived correction (p-FDR <0.01). Relative to baseline, reductions in alcohol and cannabis total risk scores were detected at 1 m, but not subsequent time points (p > .38) (Figure), Table). An effect of time was found on the AUDIT for frequency of alcohol consumption (p-FDR <0.01), which was significantly reduced at 1 m and 6 m, but not at 12 m (p > .05) (Table). An effect of time was also detected on the AUDIT for average alcohol beverages but is not interpreted since there were no significant changes between specific time points (p > .29) and it did not survive correction (p-FDR = 0.08). No effect of time was observed on the AUDIT for frequency of binge drinking episodes.
Main effects of time were identified for the BFI personality traits Agreeableness (η p 2 = 0.06), Negative Emotionality (η p 2 = 0.17), and Open-Mindedness (η p 2 = 0.06). Negative Emotionality survived correction (p-FDR <0.001), but Agreeableness (p-FDR = 0.066) and Open-Mindedness (p-FDR = 0.073) did not. Relative to baseline, trait Agreeableness was significantly higher at 12 m; trait Negative Emotionality was significantly lower at 1 m, 6 m, and 12 m; and Openmindedness was significantly higher at 1 m (Figures 2 (c-g), Table).
A main effect of time was detected on the GSES (η p 2 = 0.12) that survived correction (p-FDR <0.01). Relative to baseline, significant increases in self-efficacy were found at 1 m and 12 m, but not 6 m (Figure, Table).
No significant main effects of time were found on the BIS Total Score (p = .13) or the BIS subscales Cognitive Impulsivity and Behavioral Impulsivity (p > .05).
Main effects of time were detected for relationship satisfaction with: friends (η p 2 = 0.08) and partner's parents (η p 2 = 0.14). Relationship satisfaction with friends survived correction (p-FDR = 0.042), but satisfaction with partner's parents did not (p-FDR = 0.066). Relative to baseline, improvements in relationship satisfaction with friends and partner's parents were found at 12 m (Figure), Table).
Main effects of time were identified on the AS subscales Authentic Living (η p 2 = 0.06), External Influence (η p 2 = 0.09), and Self-alienation (η p 2 = 0.17). External Influence and Self-Alienation survived correction (p-FDR = 0.01 and p-FDR <0.001, respectively), but Authentic Living did not (p-FDR = 0.063). Relative to baseline, Authentic Living was increased at 12 m; External Influence was reduced at 1 m, 6 m, and 12 m; and Self-alienation was reduced at 1 m, 6 m, and 12 m, showing a 50% reduction at 12 m (Figure(h-j); Table).
A significant main effect of time was identified on the SBC subscale Body Dissociation (η p 2 = 0.10), but not Body Awareness (p = .06) (Figure)). Body Dissociation survived correction (p-FDR = 0.011). Relative to baseline, Body Dissociation was statistically lower at 1 m (Table).
Significant main effects of time were detected on the NRS subscales Self (η p 2 = 0.08) and Experience (η p 2 = 0.11) but not Perspective (p = .09; Figure)). Experience survived correction (p-FDR = 0.032), but Self did not (p-FDR = 0.074). Relative to baseline, Self was increased at 6 m and Experience was increased at 12 m (Table).
Significant main effects of time were detected on the ISS (η p 2 = 0.12), which survived correction (p-FDR = 0.003). Relative to baseline, spirituality was significantly increased at 1 m, 6 m, and 12 m (Figure, Table)
No significant main effects of time were found on BMI (p = .14) or vegetable (p = .07) or fruit (p = .32) consumption (Table).
Significant differences between those reporting/not reporting further ayahuasca consumption were identified in the models for negative emotionality (η p 2 = 0.09), relationship satisfaction, with an intimate partner [η p 2 = 0.13] and parents [η p 2 = 0.10] and fruit (η p 2 = 0.09) and vegetable consumption (η p 2 = 0.24). However, the only significant difference from baseline was in those with no further ayahuasca consumption, who reported greater relationship satisfaction with parents at 12 m. Only between-group differences were observed for vegetable consumption and relationship satisfaction. Those who consumed more ayahuasca reported higher vegetable intake at all time points, including baseline, and reported higher parental relationship satisfaction at baseline and the 1-month follow-up (p < .05).
Large effect sizes were observed for depression, anxiety, stress (DASS), and overall mental health (SF12-MCS). Of the well-being measures, constructs encompassing negative mood demonstrated large effect sizes, including subscales for negative affect (PANAS), negative emotionality trait (BFI), body dissociation (SBC), and self-alienation (AS). Additionally, large effect sizes were observed for alcohol and cannabis risk scores (ASSIST), self-efficacy (GSES), and relationships with partner's parents (RS). Medium effect sizes were observed for other measures of well-being, including positive affect (PANAS), agreeableness and openness (BFI), authentic living and external influence (AS), nature relatedness (NRS), spirituality, and relationships with friends (RS).
Despite growing observational evidence for the therapeutic potential of naturalistic ayahuasca consumption, the breadth and durability of improvements remain unclear. This study identified large improvements in mental health and wellbeing that were sustained over a 12-month follow-up period in ayahuasca naive participants. Particularly novel to this study, we found that reductions in depression and anxiety associated with ayahuasca consumption were maintained longer in individuals with a lifetime diagnosis of a depressive or anxiety disorder relative to those without. Ayahuasca use was also associated with a short-term enhancement of positive affect and a sustained reduction in negative affect. Trajectories of change across the psychological measures were varied, with some effects diminishing, some growing, and others remaining stable over the 1-year follow-up period. For example, Body Dissociation was only significantly altered from baseline at 1 month, while Nature Relatedness and Relationship Satisfaction were only significantly different from baseline at later assessments. Significant changes in Negative Emotionality, External Influence, Self-alienation, and Spirituality were maintained across the entire followup duration, raising the possibility that these are stable (trait) changes. The effect sizes for most of these wellbeing measures were moderate in magnitude, and generally not as robust as the large effect sizes observed on the mental health measures. Our findings contribute to the literature suggesting that structured use of ayahuasca may have therapeutic effects on psychopathology. Clinical and naturalistic studies have reported antidepressant and anxiolytic effects in healthy and psychiatric samples, ranging from 7 days to 6 months after consumption. One small study identified reductions in depression symptoms up to 12 months in clinically depressed patients. Another study at the Takiwasi Ayahuasca Center in Peru reported significant reductions in depression, anxiety, and alcohol use at 12 monthsbut was confounded by the multifactorial treatment program and long length of treatment (27.9 weeks). Cross-sectional research supports a negative relationship between the frequency of ayahuasca consumption and the level of alcohol and drug use, but longitudinal studies are scant. Although decreases in alcohol and cannabis use were detected 1 month after the ceremony, none of the subsequent time points were significant from baseline. Participants exhibited low-risk alcohol use and moderate-risk cannabis use at baseline, which may have presented floor effects. Combined with the small sample size, the analysis may be underpowered but the effect sizes hint toward large effects. Changes across wellbeing measures were in a direction associated with adaptive functioning and improved mental health. For instance, links have been established between mental health and lower Neuroticism and greater Agreeableness; wellbeing and Nature Relatedness (Grabowska-Chenczke, Wajchman-Świtalska, and Woźniak 2022); life satisfaction and Interpersonal Relationships; and lower prevalence of mental illness and Self-efficacy. Similarly, the Authenticity Scale subscales correlate strongly with mental health challenges, with External Influence and Self-alienation being positively correlated, and Authentic Living being negatively correlated. Thus, the observed decreases in Negative Emotionality, External Influence, and Self-Alienation, and contemporaneous increases in Relationship Satisfaction, Nature Relatedness, and Selfefficacy might reflect a pattern of improved adaptive functioning and mental health. For example, reduced self-alienation and greater relationship satisfaction may support meaningful social connections that foster resiliency toward changing environmental, social, and economic conditions. Further, reduced external influence and greater self-efficacy and connection to nature may improve one's ability to handle stress, recover from adversity, and maintain overall mental health. Multiple studies have reported alterations in personality structure after naturalistic ayahuasca use characterized by increases in Agreeableness, Open-mindedness, Conscientiousness, and Extraversion, and decreases in Neuroticism (analogous to Negative Emotionality). Our findings are partially consistent with these reports, although we did not observe significant changes in Conscientiousness or Extraversion, and increases in Open-mindedness and Agreeableness did not survive correction. Here, we found that personality traits undergo dynamic shifts across time, with numerical increases in Openmindedness only at the 1 month assessment, increases in Agreeableness only at the 12 month assessment, and significant decreases in Negative Emotionality across the entire follow-up duration. Personality traits qualitatively displayed distinct linear, quadratic, and cubic relationships with time (see Figure). Ayahuasca has also been associated with changes in other psychological constructs and health-related behaviors. Studies point toward enhanced spirituality, nature relatedness, self-efficacy, body connection, and interpersonal relationships, and positive lifestyle and health-related behaviors, including diet. The present study extends this work by demonstrating persistent increases in self-efficacy, nature relatedness, and relationship satisfaction with friends that lasted up to 12 months. Nonsignificant changes in body awareness and impulsiveness suggest some constructs may not be affected by ayahuasca treatment. Further, the absence of changes in some health-related behaviors (i.e., body mass index, vegetable and fruit consumption, and physical activity) is in contrast to several cross-sectional studies.
This study has notable limitations. First, a comparison group was not utilized, and therefore, the potential influence of demand characteristics and elapsed time on treatment effectiveness cannot be ruled out., likely contribute to the effects observed in this study. Selection and attrition biases are inherent to research and restrict the generalizability of results. Partially mitigating this concern, 81.8% and 63.6% of the original sample were retained for primary mental health measures at months 1 and 12, respectively. Further, linear mixed models were fitted for missing outcome data and have been shown to provide correct inferences. Another limitation is the use of self-report measures and the lack of expectancy data and the total frequency of ayahuasca use, which preclude conclusive evidence. However, broad coverage and converging results across mental health measures support the validity of these findings.
Randomized clinical trials using objective measures are needed to elucidate psychological mediators and moderators of therapeutic action, and to disentangle the temporal dynamics and causal mechanisms of ayahuasca-induced trait changes.
This study identified long-term improvements in mental health and wellbeing outcomes following naturalistic ayahuasca consumption among first-time users with relatively poor mental health at baseline.
Create a free account to open full-text PDFs.
Kiraga, M. K., Mason, N. L., Uthaug, M. V. et al. · Frontiers in Pharmacology (2021)
Kohek, M., Ona, G., Dos Santos, R. G. et al. · Journal of Psychoactive Drugs (2022)
Lawn, W., Hallak, J. E., Crippa, J. A. et al. · Scientific Reports (2017)
Murphy-Beiner, A., Soar, K. · Psychopharmacology (2020)
Netzband, N., Ruffell, S., Linton, &. S. et al. · Psychopharmacology (2020)
Palhano-Fontes, F., Barreto, D., Onias, H. et al. · Psychological Medicine (2018)
Perkins, D., Opaleye, E. S., Simonová, H. et al. · Drug and Alcohol Review (2021)
Perkins, D., Pagni, B. A., Sarris, J. et al. · Frontiers in Pharmacology (2022)
Perkins, D., Ruffell, S. G. D., day, K. et al. · Frontiers in Neuroscience (2023)
Perkins, D., Schubert, V., Simonová, H. et al. · Frontiers in Pharmacology (2021)
Ruffell, S. G. D., Netzband, N., Tsang, W. et al. · Frontiers in Psychiatry (2021)
Rush, B., Marcus, O., García, S. et al. · Journal of Studies on Alcohol and Drugs (2024)
Dos Santos, R. G., Osório, F. L., Hallak, J. E. · Archives of Clinical Psychiatry (São Paulo) (2018)
Uthaug, M. V., Mason, N. L., Toennes, S. W. et al. · Psychopharmacology (2021)
Ramaekers, J. G. · Psychopharmacology (2022)
Weiss, B., Miller, J. D., Carter, N. T. et al. · Scientific Reports (2021)