Extended difficulties following the use of psychedelic drugs: A mixed methods study

Evans and colleagues situate their study within the recent ‘‘psychedelic renaissance’’, noting renewed clinical and public interest in substances such as psilocybin, LSD, ayahuasca and MDMA. While much contemporary research focuses on therapeutic benefits and on challenging acute ‘‘trip’’ experiences, the authors identify a gap concerning adverse experiences that persist well beyond the pharmacological effects. Earlier studies and surveys suggest such extended difficulties can include anxiety, perceptual changes, derealisation/depersonalisation and, in some cases, psychotic symptoms or long-lasting distress, but prevalence estimates, phenomenology and predictors remain poorly characterised. This study aims to describe the types, phenomenology and perceived causes of difficulties that last more than 24 hours after a psychedelic experience, and to test hypotheses about predictors of the range and duration of those difficulties. Using a convergent mixed-methods approach, the investigators collected quantitative checklist data and brief written narratives from people who reported enduring problems, with the goal of combining epidemiological description and inductive qualitative analysis to inform harm-reduction and future research priorities.

The study used a convergent mixed-methods design in which quantitative and qualitative data were collected simultaneously via an anonymous online survey. Eligibility criteria were: having experienced difficulties after using a psychedelic that negatively affected functioning for more than one day, being aged 18 or over, and sufficient proficiency in written English. Recruitment occurred through social media, newsletters, student email lists and a newspaper advert; 608 individuals completed the survey between October 2022 and January 2023, and all provided written consent. The researchers report the sampling strategy as randomized sampling and state the sample provided satisfactory power for planned analyses. Ethical approval was obtained from the University of Greenwich. Quantitative data included closed-ended items on substance(s) used, setting (later grouped into guided versus unguided), whether dose was known, perceived duration and types of extended difficulties (emotional, self-perception, cognitive, social, ontological, spiritual, perceptual, other), and whether participants had prior or subsequent mental-health diagnoses or childhood trauma they believed relevant. Acute trip challengingness was measured by a single 4-point Likert item (Not at all to Extremely challenging). The ‘‘range of difficulties’’ variable was computed as the sum of selected difficulty categories. Statistical procedures were performed in IBM SPSS v.28, with significance adjusted to p = 0.02 using the Hochberg step-up method. Two linear regressions and one One-Way ANOVA were used to test predictors of range and duration of difficulties and differences by drug type; parametric assumptions are reported as met. For the qualitative component, the authors applied Structured Tabular Thematic Analysis (ST-TA), an inductive thematic method developed for brief-text datasets. Analysts proceeded through immersion, coding, tabulation, inter-analyst agreement checks, frequency exploration and thematic mapping. Two analysts independently coded subsamples until an agreement threshold of 80% was reached after three iterations. Three analysts (with backgrounds in philosophy/psychology, qualitative methods, and psychedelic research) conducted analysis within a dialogical, reflexive process including eight scheduled discussions to address potential biases and improve clarity of theme definitions. Participants were invited to write for about 3–5 minutes describing their lasting difficulties; 590 respondents supplied such narratives.

Sample and context: Among 608 respondents, the most commonly reported substances associated with enduring difficulties were psilocybin (27%) and LSD (25%), followed by ayahuasca (10%), cannabis (10%), MDMA (7%), DMT (5%), ketamine (4%), mescaline (2%) and salvia divinorum (1%); 8% selected Other and specified compounds including iboga, 5‑MeO‑DMT and bufo. Regarding dose knowledge, 449 (73.8%) said they or someone present knew the dose, 156 (25.7%) said no. The most frequent locations for the index experience were ‘‘with a friend, partner or group of friends’’, then ‘‘on my own’’, and ‘‘in a group ceremony’’. Settings were later grouped into guided versus unguided for analyses. Duration and types of extended difficulties (closed-ended): Responses to duration categories showed prominent peaks at 1–3 years and >3 years, indicating many participants reported very long-lasting problems. On the closed-ended checklist, emotional difficulties were the most endorsed type (76%), followed by self-perception difficulties (58%) and cognitive difficulties (52%); 50% reported ontological difficulties and 34% reported spiritual difficulties. Mental-health and trauma correlates, and attitudes: Prior to the experience, 173 (28.5%) reported a mental-health diagnosis; among these, 79 (45.9%) thought that prior diagnosis may be linked to the post-psychedelic difficulties. Childhood or youth trauma that participants regarded as potentially contributory was reported by 243 (40%). After the psychedelic experience, 114 (18.8%) reported receiving a subsequent mental-health diagnosis; of those, 59 (53.6%) believed the psychedelic experience had contributed. Regarding ongoing use and attitudes, 334 (54.9%) reported they still take psychedelics, 246 (40.5%) said they do not; 89.7% agreed or strongly agreed that insights gained from psychedelics in supportive settings are worth the risks. Predictors of range and duration: A linear regression predicting range of difficulties (N around 524) yielded a significant overall model (R2 = .04, F(4,519) = 4.97, p < .001). Two predictors reached significance: greater acute challengingness of the trip (β = .17, p < .001) and being in an unguided setting (β = .11, p = 0.008). A separate regression predicting duration produced a significant model (R2 = .06, F(4,518) = 7.73, p < .001) in which only trip challengingness was a significant predictor (β = .21, p < .001); being in an uncontrolled setting and not knowing dose trended in the predicted direction but were not significant. Drug differences: After excluding polysubstance reports and very small groups, One-Way ANOVAs tested drug effects. Drug type significantly affected duration (F(9,477) = 2.89, p = 0.002) but not range (F(9,480) = 1.33, p = 0.06). Post-hoc LSD comparisons indicated longer durations principally associated with ayahuasca, DMT and LSD, whereas psilocybin and MDMA were associated with shorter durations. Qualitative thematic findings: From 590 written narratives, the inductive ST-TA produced a large thematic scheme. Principal meta-themes and selected prevalence figures reported in the extracted text include Emotional Difficulties (67% of narratives), Existential and Ontological Difficulties (42%), Self‑Perception Difficulties (including depersonalisation at 16%), Somatic Difficulties (19% with sleep problems at 9%), Cognitive Difficulties (18% overall, with confusion in 9%), Behavioural Difficulties (substance-use problems 2%, work/study/financial impairment 5%) and discrete reports of psychotic episodes. Within emotional themes, anxiety/fear was the most common sub-theme (26%), followed by paranoia (11%), panic attacks (9%), depression (12%) and suicidality (6%). Perceptual disturbances and visual distortions (including reports coded as flashbacks and HPPD-like symptoms) were also described; the analysts distinguished flashbacks from prolonged perceptual symptoms when participants used those terms. Many narratives described resurfaced trauma, existential struggle, derealisation and fear of permanent damage. The authors present exemplar quotations to illustrate each sub-theme and note heterogeneity in phenomenology, intensity and duration of reported difficulties.

Evans and colleagues interpret their convergent findings as indicating that extended difficulties after psychedelic use are heterogenous but commonly emotional in nature, with anxiety, fear and panic prominent across both checklist and narrative data. The quantitative checklist found emotional problems most commonly (76%), and the inductive narratives corroborated this with emotional themes appearing in 67% of written accounts; other frequent problem domains included self‑perception changes, cognitive impairment, existential/ontological disturbance and perceptual anomalies. The authors emphasise that extended difficulties vary substantially in duration, with many participants reporting problems that last over a year. In relation to prior literature, the study’s results align with earlier surveys that highlight anxiety and social disconnection as common post-psychedelic difficulties, and they complement case reports and smaller qualitative studies documenting derealisation, depersonalisation and persistent perceptual changes. The finding that acute trip challengingness predicts both a broader range and longer duration of subsequent difficulties supports previous recommendations to limit the duration of acute distress. Likewise, identification of unguided settings as associated with a greater range of difficulties accords with literature emphasising preparation, dose knowledge and guided support as protective factors. Drug-specific differences were observed, with ayahuasca, DMT and LSD linked to longer-lasting problems than psilocybin or MDMA in this sample. The authors acknowledge several limitations that constrain inference. The sample was self-selected, mainly Western, English-speaking and predominantly white, limiting generalisability across cultures or contexts where psychedelics are ritualised. Data are cross‑sectional and based on retrospective self-report, so recall bias and subjective attributions affect the findings and causal claims cannot be made. The written narratives were brief by design, precluding in-depth probing; the authors note that follow-up interview studies are planned to explore themes such as childhood trauma and existential struggle more thoroughly. Methodological challenges in thematic coding are also acknowledged: the team conducted three iterations to reach an 80% inter-analyst agreement and suggest future work might consider a more streamlined coding frame for replicability. Practically, the investigators argue their findings have implications for harm reduction, informed consent and integration support. Given the prominence of anxiety and fear among enduring difficulties, they recommend that legal providers and clinical trial teams supply clear guidance on self-soothing and coping strategies, inform participants about the range and possible duration of adverse outcomes, and offer integration resources over months where needed. The authors also highlight the complexity that some difficult experiences are later appraised as meaningful or growth-promoting by participants, which complicates binary ‘‘adverse versus beneficial’’ classifications and suggests the importance of individual appraisal and integration support. Finally, they call for further research into specific phenomena identified here—such as derealisation, long-term perceptual disturbances, social disconnection and harms arising within guided settings—to better characterise risks and protective practices.