Psychedelic use and psychiatric risks

Earlier research has shown therapeutic promise for psychedelics in treating several psychiatric disorders, but the safety evidence comes mostly from clinical trials that apply strict exclusion criteria. As a result, relatively little is known about psychiatric risks associated with naturalistic (non‑clinical) psychedelic use, particularly among groups typically excluded from trials such as people with personal or family histories of psychotic or bipolar disorders. Potential risks discussed in the literature include persistent visual phenomena (hallucinogen persisting perception disorder, HPPD) and, more rarely, onset or exacerbation of psychosis or mania, but prevalence estimates and predictors remain uncertain and prior epidemiological findings are mixed. Simonsson and colleagues set out to examine associations between lifetime naturalistic psychedelic use, unusual visual experiences, and recent psychotic symptoms in a sample intended to be representative of the US adult population by sex, age, and ethnicity. The aim was to assess whether lifetime psychedelic use relates to lifetime unusual visual phenomena and to psychotic symptoms in the past 2 weeks, and to test whether personal or family histories of psychotic or bipolar disorders modify these associations.

The study used an online cross‑sectional survey administered via Prolific Academic in October 2021. A priori power calculations (G*Power 3.1) indicated a need for 395 psychedelic users to detect a small effect (Cohen's f2 = 0.02) with 80% power and alpha = 0.05; assuming ~14% lifetime psychedelic use in US adults, the investigators aimed to recruit ~2,800 participants and collected data from N = 2,822 US residents aged 18 or older. The sample was stratified by sex, age, and ethnicity to mirror US adult distributions. Participants received US$2.20 for completion and the University of Wisconsin–Madison Institutional Review Board determined the procedures to be exempt. Measures included self‑reported demographics and lifetime use of several substances. Lifetime psychedelic use was coded dichotomously (ever used any of ayahuasca, DMT, LSD, mescaline, peyote or San Pedro, or psilocybin = 1; never = 0). Unusual visual experiences were assessed with a 9‑item scale asking whether respondents had ever experienced listed visual phenomena outside of acute intoxication or sleep‑related states; internal consistency was alpha = 0.80 and item scores were summed. Participants were also asked whether a medical professional had ever told them they had HPPD (yes/no). Recent psychotic symptoms were measured with the 6‑item psychotic ideation subscale of the Psychiatric Diagnostic Screening Questionnaire (past 2 weeks), summed to form a total score (alpha = 0.69). Respondents reported current or past personal history of any psychotic disorder or bipolar I/II disorder and whether a first‑ or second‑degree relative had such disorders; both personal and family histories were coded as binary variables. For analyses, the investigators used Pearson chi‑squared tests and t‑tests to explore unadjusted group differences (psychedelic users versus non‑users). Primary inferential analyses comprised four multiple linear regression models: model 1 tested lifetime psychedelic use predicting lifetime unusual visual experiences; model 2 tested lifetime psychedelic use predicting past 2‑week psychotic symptoms; model 3 tested the interaction between lifetime psychedelic use and personal history on psychotic symptoms; and model 4 tested the interaction between lifetime psychedelic use and family history on psychotic symptoms. All models adjusted for covariates used in prior comparable work: age, gender, ethnoracial identity, educational attainment, annual household income, marital status, engagement in risky behaviour, and lifetime use of cocaine, sedatives, pain relievers, marijuana, PCP, MDMA, and inhalants (each drug entered separately). Sensitivity analyses repeated the four models with dichotomised outcomes using logistic regression. The authors note a data collection error resulted in omission of some covariates (for example, lifetime exposure to an extremely stressful event). P‑values were reported to three decimal places.

The final sample comprised N = 2,822 respondents, of whom 613 reported lifetime psychedelic use. Descriptive comparisons showed that individuals reporting lifetime psychedelic use had significantly higher scores on the lifetime unusual visual experiences scale than non‑users, while no unadjusted difference was observed for past 2‑week psychotic symptoms. Personal and family histories of psychotic or bipolar disorders were more commonly reported among psychedelic users than non‑users. A doctor or other medical professional‑diagnosed HPPD was reported by 1.3% of respondents who had used psychedelics versus 0.3% of non‑users. The extracted text reports no group differences in reported frequency or treatment seeking for unusual visual experiences. In covariate‑adjusted multiple linear regression models, lifetime psychedelic use was associated with higher lifetime unusual visual experiences. No direct association was observed between lifetime psychedelic use and past 2‑week psychotic symptoms in adjusted models. However, the interaction between lifetime psychedelic use and family history of psychotic or bipolar disorders was statistically significant: psychotic symptom scores were highest among respondents who both reported lifetime psychedelic use and a family history, and lowest among respondents who reported lifetime psychedelic use but no family history. There was no significant interaction between lifetime psychedelic use and personal history of psychotic or bipolar disorders on recent psychotic symptoms. Sensitivity analyses using dichotomised outcomes and logistic regression produced broadly consistent results according to the authors.

Simonsson and colleagues interpret the results as suggesting that lifetime naturalistic psychedelic use may be associated with a greater prevalence of unusual visual experiences across the lifetime, and that psychedelic use may be associated with increased recent psychotic symptoms specifically among people with a family history of psychotic or bipolar disorders. The authors note the observed 1.3% HPPD diagnosis rate among lifetime psychedelic users aligns broadly with prior community estimates of HPPD‑like onset linked to drug use, while acknowledging uncertainty about diagnostic accuracy and HPPD subtypes in their data. The investigators situate the family history interaction within a potential genetic predisposition framework, consistent with current clinical guidelines that recommend caution or exclusion for individuals with a family history of psychotic or bipolar illness. They also discuss how psychotic and bipolar disorders are genetically related and raise the possibility that psychedelics could in some cases precipitate manic switches, which would warrant separate examination of psychotic and manic outcomes. The authors acknowledge multiple limitations that constrain inference. Chief among these is the cross‑sectional design, which precludes causal conclusions and allows reverse causation (for example, pre‑existing unusual visual experiences or family history could influence likelihood of psychedelic use). The sample was stratified only on sex, age, and ethnicity and may not be representative on income, education, or other relevant factors. Unmeasured confounding (for example, amphetamine use) could have influenced associations. The single HPPD item did not distinguish type I from type II HPPD and may reflect diagnostic or reporting inaccuracies. Measurement time frames differed across key outcomes (lifetime for unusual visual experiences versus past 2 weeks for psychotic symptoms), limiting direct comparability. Finally, personal and family history items did not disaggregate psychotic disorders and bipolar disorders or their subtypes, reducing specificity of interaction findings. Based on these limitations, the authors recommend longitudinal designs, more precise psychiatric history measurement, disaggregation of psychotic and bipolar conditions (and mania versus psychosis), and incorporation of genetic risk indicators such as polygenic risk scores in future research to clarify for whom and under what circumstances psychedelic use might be associated with psychiatric harms.

The study's findings indicate that lifetime naturalistic psychedelic use may be linked to a higher prevalence of unusual visual experiences across the lifetime, and that psychedelic use is associated with elevated recent psychotic symptoms among individuals reporting a family history of psychotic or bipolar disorders but not among those without such a family history. The authors conclude that future studies should distinguish between different psychotic and bipolar disorders, specify subtypes, employ longitudinal designs, and incorporate variables such as polygenic risk scores to better elucidate possible cause‑and‑effect relationships.