Acute, subacute and long-term subjective effects of psilocybin in healthy humans: a pooled analysis of experimental studies

Psilocybin is a classical serotonergic hallucinogen found in Psilocybe mushrooms and has been used historically in ritual contexts and, more recently, in psychopharmacological research. Earlier clinical and experimental work showed that psilocybin produces an altered state of consciousness (ASC) characterised by marked changes in perception, mood, thought, and self-experience, and that its acute effects are generally shorter in duration than those of LSD. However, many early human studies lacked standardisation, adequate controls, follow-up, or sufficiently large and representative samples, leaving uncertainties about dose–response relations, the incidence of acute and subacute adverse reactions, and possible long-term psychological sequelae such as persisting perceptual disturbances or increased drug use. Studerus and colleagues set out to address these gaps by pooling raw data from eight double-blind, placebo-controlled experimental studies conducted in their laboratory between 1999 and 2008. Their aim was to describe acute, subacute (24 h), and long-term (8–16 months) subjective effects and side effects of psilocybin in healthy volunteers using validated psychometric instruments, and to evaluate dose–response relationships and the tolerability of the drug when administered in a controlled research setting.

The analysis pooled data from eight double-blind placebo-controlled within-subject studies performed in Zurich between 1999 and 2008. Across these studies 110 healthy subjects (59 males, 51 females; age range 20–47, mean 26.9 ± 5.5 years, all Caucasian) underwent a total of 227 psilocybin sessions. Psilocybin doses reported in the extracted text ranged from 45–315 mg/kg body weight, with absolute doses given as 2–28 mg; the extracted text does not clearly reconcile the apparent inconsistency between the per‑kilogram and absolute dose units. Sessions were generally randomised and counterbalanced; in one study pre-treatment with the 5-HT2A antagonist ketanserin was also tested. Experimental sessions took place in clinical research facilities, subjects received preparatory briefing, had physiological monitoring during sessions, and were discharged only after investigator assessment. Participant screening excluded individuals with personal or first-degree family histories of schizophrenia, major depression, bipolar disorder, borderline personality disorder, neurological disorders, or regular substance abuse. Screening procedures included a structured psychiatric interview, a computerized diagnostic system, physical examination with ECG and blood tests, and psychometric screening (Freiburg Personality Inventory and SCL-90). High scores on emotional lability (FPI) beyond two standard deviations were exclusionary. Subjective effects were assessed acutely and subacutely with the 5D-ASC (Altered States of Consciousness Rating Scale) and the short Adjective Mood Rating Scale (AMRS); subacute physical and psychological complaints 24 h after dosing were measured with the List of Complaints (LC). Long-term outcomes were captured with an investigator-constructed follow-up questionnaire mailed 8–16 months after the last session, covering retrospective ratings of the acute experience, changes in values and attitudes, changes in drug consumption, spontaneous ASC or flashbacks, and negative changes in psychological well‑being. The 5D-ASC assesses five primary dimensions (Oceanic Boundlessness, Anxious Ego Dissolution, Visionary Restructuralization, Auditory Alterations, and Reduction of Vigilance) and a global ASC score; AMRS comprises 15 mood subscales. Because the pooled data were hierarchical (repeated measures within subjects, subjects within studies) and unbalanced across dose conditions, the researchers used linear mixed-effects models (nlme package in R) to model dose effects with study and subject as random effects. Dose–response shapes were tested with orthogonal polynomial contrasts and, where significant effects were found, one-tailed Dunnett contrasts identified minimum effective doses. The AMRS analyses focused on placebo versus medium-dose conditions (215–260 mg) where available, with drug × time interactions used to distinguish acute from longer-lasting effects. LC total scores were square-root transformed to reduce skew and analysed by mixed models; item-level comparisons used Cochran Q and McNemar tests. Holm correction was applied across multiple hypothesis tests to control type I error. For long-term follow-up, only descriptive statistics could be applied because all participants had received psilocybin and no contemporaneous control group was available.

Sample and retention: The pooled dataset comprised 227 psilocybin sessions in 110 participants. Dose counts were eight very low doses (45 mg/kg), 74 low doses (115–125 mg/kg), 104 medium doses (215–260 mg/kg) and 41 high doses (315 mg/kg). Most participants were students or graduates, 60% were hallucinogen‑naïve, and 90% had used cannabis at least once. Seven subjects dropped out after at least one active dose; five drop-outs were related to acute adverse reactions (including two unusually intense low-dose reactions, one transient hypotension with fainting and vomiting, and two voluntary withdrawals after high-dose anxiety) and all acute problems resolved by the end of the experimental day. Long-term follow-up questionnaires were completed by 90/110 subjects (82%), with responders similar to non-responders on key baseline measures. Acute psychological effects (5D-ASC): Psilocybin produced significant, dose-dependent increases on all 5D-ASC scales and on the global ASC score. The scales ranked in effect size with Visionary Restructuralization (VR) showing the largest drug effect, followed by the global ASC, Oceanic Boundlessness (OB), Reduction of Vigilance (RV), Anxious Ego Dissolution (AED), and Auditory Alterations (AA); all reported F-values were statistically significant at p < 0.001. One‑tailed Dunnett contrasts showed that all active dose conditions differed from placebo except the very low (45 mg/kg) dose. Dose–response relations were approximately linear across the administered range. Considerable inter-individual variability was evident: some individuals reported strong effects at lower doses while others reported minimal effects at the highest dose. Using a threshold of >70% of the maximum possible score to indicate a strong subjective effect, 22% of subjects in the highest dose reached this cut-off on OB (versus 0–7.8% in lower doses), VR was high in 19.6% at the highest dose, and marked AED occurred in 7.3% (315 mg/kg) and 5.7% (250–260 mg/kg) conditions. Adjective Mood Rating Scale (AMRS): Mixed-effects analyses across 15 AMRS subscales revealed significant drug × time interactions for several subscales (emotional excitation, dreaminess, heightened mood, dazed state, sensitivity, concentration, and tiredness), with most interaction effects reflecting stronger drug-induced changes at earlier measurement times (acute effects). Main effects of drug that were independent of time—interpreted as longer-lasting effects—were observed for introversion, inactivation, efficiency‑activation, extroversion and apprehension‑anxiety. Short-term side effects (LC at 24 h): LC global scores 24 h after dosing differed by drug condition (F 5,159 = 9.64, p < 0.001). Compared with placebo, active doses were associated with increased reports of somatic and cognitive complaints, predominantly symptoms of tiredness, fatigue, headaches and reduced energy. Item‑level analyses found increased fatigue, headaches and lack of energy to survive correction for multiple comparisons. No serious complications (for example, fear of death, suffocation, persistent vomiting, or fainting) were reported in the 24‑h assessments. Long-term follow-up (8–16 months): Of the 90 respondents, the majority rated the acute experience positively in retrospect: over 60% described it as very enriching and more than 90% as at least moderately enriching. Many subjects reported positive changes in attitudes and values, most often increased appreciation of altered states, nature and aesthetic experiences. Most participants reported unchanged drug consumption habits; where change occurred it was more often a decrease than an increase in use. Only a small minority reported spontaneous ASC after the studies (8 subjects), and detailed descriptions did not meet DSM‑IV criteria for Hallucinogen Persisting Perception Disorder (HPPD) or ICD‑10 flashbacks. Five subjects (5.5% of follow-up responders) described experiences they labelled as flashbacks, but the study team interpreted these reports as broad, memory-like recollections rather than HPPD. Eleven respondents (12%) reported negative changes in psychological well‑being; four of these judged the changes unrelated to psilocybin. Among the seven who considered the changes drug-related, most symptoms were mild and transient. One case required psychotherapy for several weeks after a high-dose session; that subject stabilised after treatment and did not relapse.

Studerus and colleagues interpret their pooled analysis as confirming that psilocybin dose‑dependently induces a broad ASC in healthy, well‑screened volunteers, with prominent visual and perceptual changes (Visionary Restructuralization), followed by experiences captured by Oceanic Boundlessness (including phenomena reminiscent of mystical‑type experiences) and, to a lesser extent, Anxious Ego Dissolution. Reported hallucinations were typically recognised as unreal (pseudo‑hallucinations) and reality testing generally remained intact. The mixed‑model results and cumulative distributions highlight marked inter‑individual variability in subjective responses, leading the investigators to emphasise that dose alone poorly predicts the intensity of effects and that non‑pharmacological factors (expectations, personality, interpersonal support, environment) and pharmacokinetic variables (for example, psilocin plasma levels) likely play important roles. In comparing their findings to other recent high‑dose studies, the authors note differences in participant selection, preparation and the experimental setting that may explain variance in rates of profound mystical experiences and acute adverse reactions seen across studies. Acute adverse reactions characterised by intense anxiety, panic or derealisation occurred infrequently and were confined to higher doses; all such events in their samples resolved with interpersonal support and did not require psychopharmacological rescue. Subacute complaints were primarily fatigue and headaches and were typically resolved within 24 h. At long‑term follow‑up most subjects retrospectively rated the experience as enriching and influential, and the investigators observed no cases of sustained psychosis, persisting perception disorder meeting HPPD criteria, or subsequent drug dependence attributable to the experimental sessions. The authors explicitly acknowledge limitations that constrain generalisability: their sample was highly selected (exclusion of individuals with psychiatric vulnerability or substance‑use histories, and a likely self‑selection bias towards people favourably disposed to hallucinogens), the follow‑up questionnaire was investigator‑constructed with unknown psychometric properties, and some analytic choices (for example, cut‑off values for 'strong' experiences) were arbitrary. Given these caveats, the investigators stress that their findings pertain to carefully prepared, supported, and well‑screened participants in a controlled research setting and should not be generalised to recreational contexts or unsupervised use. They conclude that, within those controlled conditions and predominantly moderate dose ranges, psilocybin appears to have an acceptable tolerability profile and that further clinical and experimental research using psilocybin is justified, while emphasising the need for careful subject selection, preparation and monitoring.

Pooling data from 227 administrations in 110 healthy, well‑screened volunteers, the investigators conclude that psilocybin produced dose‑dependent altered states of consciousness and was generally well tolerated acutely, subacutely and at 8–16 month follow‑up when given in a carefully controlled research environment. Acute adverse reactions were infrequent, occurred mainly at the highest doses, were managed successfully with interpersonal support, and all resolved without pharmacological intervention. No cases of subsequent drug abuse, persistent perception disorders meeting diagnostic criteria, prolonged psychosis, or other long‑term functional impairments were observed. The authors therefore consider psilocybin administration to healthy, high‑functioning and well‑prepared subjects in responsible clinical or research settings to be generally tolerable and justified for future studies.