Positive effects of psychedelics on depression and wellbeing scores in individuals reporting an eating disorder

Treatment options for eating disorders (EDs) such as bulimia nervosa, binge eating disorder and anorexia nervosa remain limited in efficacy across presentations, and existing approaches tend to be symptom-focused. Earlier research has identified abnormal serotonin functioning and high emotional avoidance as potential contributors to ED pathology, creating a mechanistic rationale for exploring classic psychedelics (LSD, psilocybin, DMT/ayahuasca) that act primarily via 5-HT2A receptor agonism. Recent clinical trials in other conditions have suggested that psychedelic-assisted interventions can produce meaningful mental health benefits and that the quality of the acute experience—for example, emotional breakthrough—predicts longer-term outcomes. Naturalistic reports and qualitative work have also suggested possible benefits for people with EDs, but no controlled clinical trials have yet reported results in this population. This study sought to provide the first quantitative, prospective assessment of psychological outcomes following a psychedelic experience in people reporting a lifetime ED diagnosis. Specifically, the investigators hypothesised that depressive symptom severity and psychological wellbeing would improve from baseline to two weeks after a planned psychedelic experience, and they explored whether self-reported emotional breakthrough during the acute experience was associated with these changes. Data were collected in a naturalistic, online cohort design to inform future controlled trials and mechanistic research relevant to ED treatment.

Data came from three anonymous, prospective online cohort studies conducted between March 2017 and December 2019 using the Psychedelic Survey platform. Participants completed baseline measures 1–2 weeks before an intended psychedelic experience, acute-experience ratings one day after the session, and endpoint measures two weeks after the experience. The design therefore provides pre/post naturalistic data rather than an experimental intervention with randomisation or control conditions. Eligible participants were adult English speakers who intended to take a psychedelic in the near future; eligible drugs included psilocybin (mushrooms/truffles), LSD/1P-LSD, ayahuasca, DMT/5‑MeO‑DMT, Salvia divinorum, mescaline, and iboga/ibogaine. Across the three studies, 50 people reported a lifetime ED diagnosis; of these, 28 completed the dependent variables at baseline and two-week follow-up, and 27 completed the acute-experience measure. Demographic data collected at baseline included age, gender, country, frequency of lifetime and recent (last 6 months) psychedelic use, lifetime psychiatric diagnoses and current medication. The extracted text does not include the full table of demographics. Psychological measures comprised the Warwick–Edinburgh Mental Wellbeing Scale (WEMWBS), a 14-item measure of hedonic and eudaimonic wellbeing scored over the past two weeks; the Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR16), a 16-item measure of depressive symptom severity with established score bands (0–5 none, 6–10 mild, 11–15 moderate, 16–20 severe, 21–27 very severe); and the Emotional Breakthrough Inventory (EBI), a recently validated 6-item visual analogue scale (0–100) assessing emotional release during the acute psychedelic experience. All analyses employed Bayesian hypothesis testing (Bayes Factors reported) with moderate JZS priors for t tests (r = 0.701) and Jeffreys–Beta priors for correlations (r = 0.3). Equivalent frequentist statistics (paired t tests, Pearson correlations) were also reported for readers preferring conventional inference. Normality tests (Shapiro–Wilk) were non-significant for QIDS-SR16 and WEMWBS, supporting the use of parametric tests. Analyses were conducted in R using BayesFactor, ggplot2 and dplyr. The extracted text does not report further details such as adjustments for multiple comparisons or specific effect-size metrics beyond means, standard deviations, correlations and test statistics.

Sample and preliminary checks: Of the 50 participants who reported a lifetime ED across the three studies, 28 (56%) completed baseline and two-week follow-up measures and 27 completed the acute-experience EBI. The extracted text indicates that the large majority had used a psychedelic at least once previously, with 20 participants reporting use within the last six months, but the detailed demographic table is not included in the text provided. Psychological wellbeing: A Bayesian paired t test provided very strong evidence for an increase in wellbeing following the psychedelic experience (BF10 = 126.24). The paired frequentist test was also significant (t(27) = 4.25, p = 0.0002). Group means show WEMWBS scores increased from a baseline mean of 45.6 (SD = 10.00) to a two‑week follow-up mean of 53.6 (SD = 8.29). Depressive symptoms: There was very strong Bayesian evidence for a reduction in depressive symptoms after the experience (QIDS-SR16 BF10 = 202.81), supported by a significant paired t test (t(27) = -4.45, p = 0.0001). At baseline nine participants (32%) fell in the moderate-to-very-severe depression range on QIDS-SR16; two weeks after the psychedelic experience no participants remained in the moderate–severe range and the group mean fell below the threshold for depression. Associations and subgroup checks: Change in QIDS-SR16 was strongly negatively correlated with change in WEMWBS (Bayesian r = -0.65, BF10 = 2270.84; frequentist r(26) = -0.746, p < 0.0001), indicating that reductions in depressive symptoms were closely tied to increases in wellbeing. Preliminary analyses comparing participants with and without a lifetime comorbid diagnosis of major depression yielded weak Bayesian evidence favouring no difference for both baseline QIDS-SR16 (BF10 = 0.453; t(28) = -0.78, p = 0.446) and change in QIDS-SR16 (BF10 = 0.45; t(16.12) = 0.715, p = 0.484). Emotional breakthrough (EBI): Bayesian correlations provided weak evidence for a positive relationship between EBI and change in WEMWBS (r = 0.26, BF10 = 1.34) and very moderate evidence for a negative relationship between EBI and change in QIDS-SR16 (r = -0.29, BF10 = 2.05). The corresponding frequentist correlations trended toward significance (WEMWBS r(25) = 0.328, p = 0.094; QIDS-SR16 r(25) = -0.380, p = 0.050). These results suggest a tentative association between acute emotional breakthrough and later psychological improvements, but effect sizes and Bayes Factors were modest. Missing data and attrition: The extracted text notes substantial attrition across the three studies (28/50 completed the core measures), which may have influenced results; no further imputation or sensitivity analyses are reported in the provided text.

Spriggs and colleagues describe this work as the first quantitative, prospective study of psychological aftereffects of a psychedelic experience in people reporting a lifetime ED diagnosis, and they emphasise the preliminary nature of the findings. Consistent with prior work in clinical and non-clinical samples, the study found very strong evidence for reductions in depressive symptoms and increases in psychological wellbeing two weeks after a naturalistic psychedelic experience. Improvements in depression and wellbeing were strongly correlated, suggesting linked changes in negative symptoms and positive mental health. The authors highlight clinical relevance in the context of EDs: psychological wellbeing and quality of life are often reduced even after symptomatic remission from an ED and are associated with relapse risk. They note that recovered individuals frequently cite positive wellbeing as central to sustained recovery, so simultaneous improvements in wellbeing and symptomatology may be meaningful for relapse prevention. The investigators also report tentative evidence that emotional breakthrough during the acute psychedelic experience may mediate positive outcomes, aligning with prior descriptions of psychedelic-facilitated emotional processing, insight and revision of entrenched mental schemata; however, the Bayes Factors and p values for these correlations were weak to moderate and the authors caution against strong inferences. Key limitations are acknowledged in detail. The naturalistic, anonymous online survey design precludes experimental control: drug dose, purity and the context of administration could not be verified, and no control group or placebo comparator was used. Reliance on self-report meant diagnostic status could not be confirmed and no ED-specific psychopathology measures or severity gradings were included. Sampling biases are probable because participants enrolled on the basis of intending to take a psychedelic, most had prior psychedelic experience and only 56% of those reporting an ED completed all measures, raising potential self-selection and attrition biases. The authors therefore call for replication in controlled clinical trials with larger, well-characterised samples to explore safety, efficacy and mechanisms—particularly the role of emotional breakthrough—in people with EDs. In conclusion, the investigators present these data as preliminary supportive evidence that a psychedelic experience can be followed by meaningful short-term improvements in depression and wellbeing among people reporting an ED, and they position the findings as a foundation for future clinical research rather than as definitive proof of efficacy.