Home / For practitioners / Mental health history: what studies measured

Mental health history: what studies measured

The question here is whether a history of psychosis, mania or trauma is linked to what happens after psychedelic use. This card sets out what surveys, hospital-record studies and a few trials measured, in which groups, and what was not located.

Published Information current to 7 November 2026Review due by 5 February 2027

Read this first. This page is education, not instructions to use. Metanoia does not supply any substance. In the United States, most of the substances discussed here are Schedule I under federal law (21 CFR 1308.11). State programs such as Oregon's and Colorado's operate under state law only, and federal law still applies. Legal status differs between countries, and none of this is legal advice. Nothing here is medical advice, and nothing here gives or recommends a dose. People who host, supply or take part in sessions involving controlled substances can face criminal liability under federal and state law. This page does not advise on that. “Healthy adults” means people with no diagnosed psychiatric or major medical condition, the population Metanoia studies. This page reports what published sources say about work done outside supervised programs, and it takes no position on the legality of any setting, practice or person.
Not an emergency resource. Metanoia is not an emergency service. In a medical or mental health emergency in the United States the number is 911, and the 988 Suicide and Crisis Lifeline is available by call or text. Elsewhere, local emergency numbers apply. Nothing on this card is a reason to delay or avoid care. The studies below include people with and without psychiatric histories, which is a different group from the healthy adults Metanoia studies. This card describes what sources report. It is not a screening tool, it is not diagnostic guidance, it does not say who should take part in a session, and it is not medical guidance.

The short answer

Is a history of psychosis, mania or trauma linked to what happens after psychedelic use?

Studies point in different directions, and none can show cause. Some online follow-up surveys and hospital-record studies found more psychotic or manic symptoms or diagnoses in certain groups, and one insurance-claims study found that a link with psychosis disappeared once earlier psychiatric history was counted. A US household survey found no association between lifetime psychedelic use and past-year screening indicators of mania, psychosis or PTSD. Trials mostly exclude people with these histories, so supervised outcomes in them are largely unmeasured. Direct data on complex trauma were thin in the sources read. This card is not a screening tool or medical guidance.

What do studies report?

Reported findings, not recommendations. This marking applies to everything in this section. Each finding applies only to the group named, and an association is not a cause.

Surveys

  • US household survey, 2001 to 2004. Among 130,152 adults, lifetime use of LSD, psilocybin, mescaline or peyote was not significantly associated, after adjustment, with past-year screening indicators of mania, psychosis or PTSD. The design was cross-sectional, the measures were screening questions and not diagnostic interviews, and family history was not available. The authors write that they cannot exclude a negative effect in some individuals or groups. Source 1, level 1
  • Two online surveys with a two-month follow-up. In samples chosen to match US and UK adults (9,732 at the start, 100 reporting use during the study), psychotic symptoms did not change overall. A personal history of bipolar disorder went with more symptoms after use, a family history of bipolar disorder with a weaker rise, and a personal history of psychotic disorder (3 users) with fewer. None of these differences survived correction for multiple comparisons, and histories were self-reported (source 2). In a purposive US sample aged 18 to 50 (12,345 at follow-up, 505 reporting use), psychotic and manic symptom scores rose among people who used, and only among those who used in an illegal context, which described 97% of the users. Manic scores rose more with a personal history of schizophrenia (4 users) or bipolar I disorder (22 users). That study did not correct for multiple comparisons. Its authors write that causal conclusions cannot be drawn, and that they did not examine whether a rise in scores is harmful (source 3). Sources 2 and 3, level 1
  • Survey of people with and without a history. In a cross-sectional online survey of 548 adults recruited purposively, 267 with a personal or family history of psychotic or bipolar disorder, the number of lifetime use occasions was not associated, after adjustment, with any of four measures of psychotic symptoms. A personal history of psychotic disorder was associated with higher scores on three of them. Diagnoses were self-reported, and the authors note that people with more severe symptoms were probably less able to complete the survey. Source 4, level 1
  • Survey of hard experiences with mushrooms. Among 1,993 people who described their worst experience with psilocybin mushrooms, recruited mainly through websites and email, the authors saw three cases that appeared linked to enduring psychotic symptoms. Only people with a hard experience took part, so the survey cannot estimate how common this is. Source 10, level 1

A review and a small trial in bipolar disorder

  • Systematic review, 2026. Of 23 studies of psychedelics or MDMA, most at serious risk of bias, the review reports higher risk in observational studies among people with bipolar I disorder, familial vulnerability, and unsupervised or illegal use. The review also reports that manic symptoms, when they occurred, were typically acute and self-limited. The card does not report how severe they were. The authors write that long-term outcomes and repeated exposure remain insufficiently studied. Source 8, level 1
  • Open-label trial in bipolar II depression. In 15 adults with treatment-resistant bipolar II depression given a single dose of psilocybin with psychotherapy, mania ratings and a suicidality scale did not change significantly over 12 weeks. There was no control group. A trial of 15 people cannot show that an uncommon event does not happen, and each dosing day had a lead therapist and an assistant therapist present, with a study physician on site. The trial excluded people with a history of bipolar I disorder, schizophrenia, psychosis (including substance-induced psychosis), delusions, paranoid, schizoaffective or borderline personality disorder, and any substance use disorder in the past 12 months. The authors write that the group was selected and highly motivated, that this prevents extrapolation to people with bipolar illness in general, that the sample was small, and that follow-up lasted only 12 weeks. They write that the findings cannot be extrapolated to bipolar I disorder, or to a mixed or hypomanic phase. They add that it is premature to extend them to people with bipolar I disorder who carry a higher risk of mania and psychosis, and that caution is warranted in expanding this approach to bipolar I disorder given the higher potential risk. Source 9, level 1

Trauma

  • Childhood adversity and later difficulties. Among 3,168 US adults aged 18 to 50 with lifetime psychedelic use, recruited online, 87.9% reported no difficulties, 6.4% difficulties lasting more than a day, and 1.3% difficulties lasting more than a year. Four or more adverse childhood experiences went with higher odds of reporting difficulties than none (adjusted odds ratio 2.84, 95% interval 1.72 to 5.03). The design was cross-sectional, both measures were retrospective self-report, and the authors write that causality cannot be established. It measured adverse childhood experiences, not PTSD or complex PTSD. Source 11, level 1
  • People with difficulties who linked them to early trauma. In an online survey of 608 adults who reported difficulties lasting more than a day after psychedelic use, 41.8% said a childhood or youth trauma may have played a role, 31.0% said no and 27.2% were unsure. Compared with those who said no, those who said yes more often reported a mental illness diagnosis before the experience (73.8% against 50.2%) and emotional difficulties afterwards (81.9% against 68.3%), and less often perceptual difficulties (18.5% against 32.2%). “Trauma” meant whatever participants themselves defined as trauma, not a clinical criterion. In 18 interviews with people chosen for that link (counts overlap), 7 described re-living an event (3 of them with no earlier memory of it), 4 described bodily or symbolic experiences, and 9 described confusion or fragments, including doubt about whether a remembered event was real. The authors write that the interview percentages describe only that group and cannot be generalized, that they did not check whether remembered events happened, and that the findings concern self-reported memory and meaning. Source 12, level 1
  • Complex PTSD and trials of MDMA-assisted therapy. A 2026 systematic review of 24 studies of MDMA-assisted therapy in people with diagnosed PTSD (335 participants, most studies at high risk of bias, 23 sponsored by one organization) found one case report of a person with complex PTSD and no study of borderline personality disorder. Six of the studies did not exclude borderline personality disorder (source 13). One phase 3 trial in that review, in 90 adults with severe diagnosed PTSD, recorded developmental trauma in 84% and excluded primary psychotic disorder, bipolar I disorder and personality disorders (source 14). MDMA is not a classic psychedelic, and both describe trial settings with a therapy team. Sources 13 and 14, level 1

How do the record studies compare?

Reported findings, not recommendations. Two cohorts from Ontario and one from US insurance claims asked related questions in different groups. This card sets them side by side and does not rank them.

  • Ontario, psychosis. In Ontario residents aged 14 to 65 with no earlier psychosis (9,244,292 people, 2008 to 2021), 5,217 had a first emergency department visit involving hallucinogens. Within three years, 3.99% received a schizophrenia-spectrum diagnosis, against 0.15% in the general population. The association remained after adjustment for earlier mental health and substance-use care, and in an analysis that left out anyone with such care in the previous five years. The exposure was the visit and not use, and the code covered several drug types, including ketamine and MDMA. According to the authors, the study cannot show cause, the finding may not apply to use that does not lead to an emergency visit, and childhood trauma and genetic risk were not measured. Source 5, level 1
  • Ontario, mania. In a second Ontario cohort (9,311,844 people aged 14 to 65 with no bipolar diagnosis, 2008 to 2022), 7,285 had acute care, meaning an emergency visit or hospital stay, involving hallucinogens. Within three years 1.43% had acute care for mania, against 0.06% in the age- and sex-matched general population. The authors say the finding may not apply to people who use hallucinogens and do not need acute care. Part of the association, they add, could reflect reverse causation, such as use to manage undiagnosed symptoms, or closer detection after the visit. Source 6, level 1
  • United States, insurance claims. In US commercial and Medicaid claims from 2015 to 2019, 6,184 people with a hallucinogen-related admission or emergency visit were compared with 267,282 people admitted for other substances (opioids, stimulants, alcohol or cannabis). A psychotic disorder was diagnosed between 30 days and six months later in 16.4% and 6.6%. The unadjusted hazard ratio was 1.22. After adjustment for demographics and earlier psychiatric diagnoses it was 0.97 (95% interval 0.95 to 1.00). The authors read this as an association largely explained by earlier psychiatric conditions. Drug types were not distinguished, follow-up was six months, the sample was limited to people with substance-related admissions, and the authors note that the data limit causal inference. Source 7, level 1

What differs: the Ontario studies compare with the general population over three years, and the US study compares with people admitted for other substances over six months. Countries, health systems, drug supply and outcome definitions also differ. The studies are not like-for-like, and this card does not decide between them. All three describe association. None can show whether hallucinogens, earlier vulnerability or something else explains it.

What is not known?

“Not located” means that the searches Metanoia ran on 7 November 2026 did not return it. For this card they covered PubMed and ClinicalTrials.gov. Embase, PsycINFO, Web of Science, Google Scholar and preprint servers were not searched. A 2026 systematic review of psychedelics and psychosis (52 papers, by its abstract) was seen in abstract only and not read, and it may bear closely on this question. A 2015 follow-up analysis of the US national survey and several registered trials in bipolar II disorder were not read. More searching could change the picture.

  • Whether a history changes outcomes once other substances, age and setting are accounted for. The sources read disagree, and most are observational.
  • Outcomes after supervised sessions in people with psychotic or bipolar I histories. The trial read here excluded them, and exclusion lists of other trials are on the card about what programs ask and why.
  • Complex PTSD. No source read in full measured it directly in people using psychedelics outside trials. The one case report sits inside a review of MDMA trials in diagnosed PTSD. Childhood trauma was measured as self-reported history in two surveys, and trauma was self-defined in one. The cause of any link was not established.
  • Long-term outcomes beyond follow-up of weeks to three years, and repeated use. The 2026 review says both remain insufficiently studied.

What this card does not say

  • It does not say who should or should not take part. It is not a screening tool, not diagnostic guidance and not a standard of care. A higher rate in a studied group is not a statement about any person. It cannot tell any reader about their own history. A clinician who knows that history is the only one who can speak to it.
  • Screening in general has its own card, about what programs ask and why. Lasting difficulties and getting medical help are separate entries, not yet written. Medicines are on the card about antidepressants and other psychiatric medicines, and nothing here concerns starting, continuing or stopping any medicine.
  • This is not an emergency resource. In the United States, 911 and the 988 line are available.
  • It does not rate how risky, safe or worthwhile any substance or practice is.
  • The kinds of harm sources report, and how each was counted, are on the card when things go wrong.

Send a correction

If this card reports something wrongly, tell us. Write to main@metanoia-ca.org and keep the card name in the subject line. Accounts of what happened in a session belong in the field reports channel, not in email. Please leave out names and anything that identifies a person or a place. We read every message.

Where does this come from?

Grade key: 1 = peer-reviewed publication or official document. A grade names the type of source, and the design and limits are given in the text. “Read in full” means Metanoia read every sentence on the date given, and any part not read is named.

Review due by 5 February 2027.

  1. Krebs TS, Johansen PØ. Psychedelics and mental health: a population study. PLoS One 2013. DOI 10.1371/journal.pone.0063972 PMID 23976938. Grade 1. Cross-sectional analysis of a national survey. Text and tables read in full 7 November 2026 (supplements not read).
  2. Honk L, et al. Longitudinal associations between psychedelic use and psychotic symptoms in the United States and United Kingdom. J Affect Disord 2024. DOI 10.1016/j.jad.2024.01.197 PMID 38280572. Grade 1. Online longitudinal survey. Text read in full 7 November 2026 (supplement not read).
  3. Simonsson O, et al. Longitudinal associations of naturalistic psychedelic use with psychotic and manic symptoms. Psychol Med 2025. DOI 10.1017/S0033291725000716 PMID 40162511. Grade 1. Preregistered online longitudinal survey. Text and tables read in full 7 November 2026 (supplement not read).
  4. Dourron HM, Bradley M, Simonsson O, et al. Naturalistic psychedelic use and psychotic symptoms: a cross-sectional study of individuals with a personal or family history of psychotic or bipolar disorders. Psychedelic Med 2025. DOI 10.1089/psymed.2024.0036 PMID 40530408. Grade 1. Cross-sectional online survey, purposive sample. Text and tables read in full 7 November 2026.
  5. Myran DT, et al. Emergency department visits involving hallucinogen use and risk of schizophrenia spectrum disorder. JAMA Psychiatry 2025. DOI 10.1001/jamapsychiatry.2024.3532 PMID 39535804. Grade 1. Population-based retrospective cohort. Text and tables read in full 7 November 2026 from the PMC page (supplement not read).
  6. Myran DT, et al. Hospital-based care for hallucinogens and risk of mania and bipolar disorder: a population-based cohort study. PLoS Med 2025. DOI 10.1371/journal.pmed.1004805 PMID 41329646. Grade 1. Population-based retrospective cohort. Text and tables read in full 7 November 2026 (supplement not read).
  7. Steinle JT, et al. Hallucinogen-psychosis associations are confounded by baseline psychiatric history. J Clin Psychiatry 2026. DOI 10.4088/JCP.25m16034 PMID 42283542. Grade 1. Retrospective insurance-claims cohort. Text read in full 7 November 2026 (tables, figures and supplement not read).
  8. Eskinazi M, et al. Psychedelic-induced hypomania and mania: a systematic review and meta-analysis. Mol Psychiatry 2026. DOI 10.1038/s41380-026-03657-6 PMID 42215638. Grade 1. Systematic review, so the studies it covers were not read. Text and tables read in full 7 November 2026 (supplement not read). A table line disagrees with the abstract: Table 1 reports about 32.2% for one naturalistic survey of people with bipolar disorder, while the abstract gives 30% as the upper bound for such studies. This card therefore cites only statements found in both the abstract and the body. Table 1 and the abstract re-checked 9 November 2026.
  9. Aaronson ST, et al. Single-dose synthetic psilocybin with psychotherapy for treatment-resistant bipolar type II major depressive episodes: a nonrandomized open-label trial. JAMA Psychiatry 2024. DOI 10.1001/jamapsychiatry.2023.4685 PMID 38055270. Grade 1. Open-label nonrandomized trial. Text and table read in full 7 November 2026 (supplements not read).
  10. Carbonaro TM, et al. Survey study of challenging experiences after ingesting psilocybin mushrooms: acute and enduring positive and negative consequences. J Psychopharmacol 2016. DOI 10.1177/0269881116662634 PMID 27578767. Grade 1. Online retrospective survey, self-selected sample. Text and tables read in full 7 November 2026 (figures not read).
  11. Olofsson M, et al. Difficulties following naturalistic psychedelic use and associations with adverse childhood experiences. Int J Drug Policy 2026. DOI 10.1016/j.drugpo.2025.105105 PMID 41389560. Grade 1. Cross-sectional mixed-methods survey. Text read in full 7 November 2026 (supplement not read).
  12. Simon G, Tadmor N, Skragge M, Evans J, Robinson O. Recalled childhood trauma and post-psychedelic trajectories of change in a mixed-methods study. Sci Rep 2025. DOI 10.1038/s41598-025-26198-4 PMID 41309700. Grade 1. Online survey and interviews. Text and tables read in full 7 November 2026 (supplementary material and reference list not read).
  13. Kasiviswanathan K, Nilaweera D, Morando M, et al. Making a case for using MDMA-assisted psychotherapy for borderline personality disorder and complex PTSD: a descriptive systematic review of the literature. Ther Adv Psychopharmacol 2026. DOI 10.1177/20451253251408626 PMID 42422592. Grade 1. Systematic review, so the studies it covers were not read. Text and tables read in full 7 November 2026 (supplement and reference list not read).
  14. Mitchell JM, Bogenschutz M, Lilienstein A, et al. MDMA-assisted therapy for severe PTSD: a randomized, double-blind, placebo-controlled phase 3 study. Nat Med 2021. DOI 10.1038/s41591-021-01336-3 PMID 33972795. Grade 1. Randomized trial. Text and tables read in full 7 November 2026 (supplement not read).