American Psychiatrists' Opinions About Classic Hallucinogens and Their Potential Therapeutic Applications: A 7-Year Follow-Up Survey

Classic hallucinogens (also called classic psychedelics), including psilocybin, lysergic acid diethylamide (LSD), and mescaline, act primarily via 5-HT2A receptor agonism and have seen accelerating clinical research into therapeutic applications. Barnett and colleagues note that Phase III trials of psilocybin-assisted therapy for treatment-resistant depression began in 2023 and that psilocybin could plausibly receive US Food and Drug Administration (FDA) approval in the near future. Despite this, there is limited contemporary data on how American psychiatrists view these drugs and their therapeutic potential. To address this gap, the investigators repeated a national survey they had conducted in 2016 using virtually the same instrument, with the primary hypothesis that psychiatrists in 2022–2023 would be more optimistic about the therapeutic promise of classic hallucinogens and less concerned about their risks than respondents in 2016. The repeat survey also sought to measure support for research and federal funding, views on medical legalisation, and intentions to incorporate hallucinogen-assisted therapy into clinical practice if regulatory approval were granted.

The study was an anonymous, Internet-based survey using a 24-item instrument closely based on the authors' 2016 questionnaire. Although the term "psychedelic" is increasingly used, the survey used the term "hallucinogen" for consistency with the prior survey and instructed respondents to consider classic hallucinogens only (examples provided) and to exclude non-classic agents such as ketamine, MDMA, and phencyclidine. Responses to attitudinal items used a five-point Likert scale from "strongly disagree" to "strongly agree." The present instrument retained the 14 items from 2016 and added questions on medical legalisation, therapeutic potential for substance use disorders (SUDs), federal funding for clinical trials, and an item asking whether the respondent would incorporate hallucinogen-assisted therapy into their own practice if federally approved. Demographic items included age, gender, training level, year of training completion, primary daily activity, treatment orientation, publication history and, newly added for this administration, race/ethnicity, state/territory of residence, and subspecialty training. For sampling, the team used a random number generator to select 1,000 American Psychiatric Association members from the APA online directory, comprising 250 resident-fellows (from 5,044 resident-fellow members) and 750 other psychiatrist members (from 22,163 eligible members). Non-US members were excluded. Invitations were emailed between 30 August 2022 and 21 March 2023, with reminders at 2 and 4 weeks. The authors planned power calculations assuming the 2016 response rate; with that assumed rate (324/1,000) they estimated >80% power at alpha = 0.05 to detect a difference of more than 0.07 between an observed and true proportion. Data collection used REDCap and analyses were performed in Stata v15.1. The authors computed summary statistics for each item, used Fisher's exact test for gender comparisons between surveys, and applied Cuzick's nonparametric trend test to compare Likert-scale responses within and between survey years. They also constructed a multivariate logistic regression model with the dependent variable defined as moderate/strong agreement to "I will incorporate hallucinogen-assisted therapy into my own practice if it gains federal regulatory approval as a medical treatment." Candidate independent variables included age, gender, attending status, race (White vs other), primary clinical activity, West-region practice, and selected opinion items (concern about psychiatric disorder risk, concern about cognitive impairment, and belief that hallucinogens show therapeutic promise). The investigators reported that independent variables showed no significant collinearity. No imputation was performed for missing data and statistical significance was set at p < 0.05 two-tailed; the authors explicitly noted multiple comparisons and chose not to apply corrections such as Bonferroni, cautioning readers about increased type I error risk.

The survey received 131 responses for an effective response rate of 13.1%. Respondents were described as demographically similar to the 2016 sample; the extracted text indicates that most respondents were attending psychiatrists, the mean year of residency completion was 2007.1 (SD 16.5), 80.3% performed primarily clinical work, and 71.4% reported a biological treatment orientation, but the extract does not provide a full demographic table in prose. On primary attitudinal items, 80.9% (106/131) of respondents moderately or strongly agreed that hallucinogens show promise in treating psychiatric conditions, while 60.8% (79/130) moderately or strongly agreed they show promise for treating substance use disorders (SUDs); the difference between these two items was statistically significant (p < 0.001 by nonparametric trend test). Large majorities moderately/strongly supported further research into therapeutic applications for psychiatric conditions (93.9%) and for SUDs (88.6%), and favoured federal funding of clinical trials for psychiatric conditions (84.7%) and for SUDs (80.9%) as reported in the extracted summary. A substantial majority believed that use of hallucinogens during psychotherapy may improve outcomes. Only a minority moderately/strongly believed that hallucinogen use increases the risk for subsequent psychiatric disorders, increases the risk of long-term cognitive impairment, is unsafe even under medical supervision, or should be illegal for recreational/non-medical use. About half of respondents reported moderate/strong intentions to incorporate hallucinogen-assisted therapy into their own practice if federal regulatory approval were granted (approximately 50.4% reported such intentions in the abstract). Analyses by demographic subgroups found few significant differences: women were more likely than men to endorse the belief that hallucinogens increase the risk for long-term cognitive impairment (specific values are presented in a supplementary table referenced in the extract). No significant differences were observed by training status. Age-group analyses showed older respondents tended to express greater reservations and less optimism about hallucinogens on several items. Comparisons between the 2016 and 2022–2023 surveys on the seven items common to both administrations revealed highly significant changes in attitude across all comparisons: respondents in 2022–2023 reported less concern about potential risks of hallucinogens, stronger belief in therapeutic potential, reduced support for criminalising non-medical use, and stronger support for further research into therapeutic applications. The 2022–2023 sample had a non-significantly larger proportion of men (61.2% vs 52.5% in 2016; p = 0.095), and gender-stratified analyses showed that the year-to-year differences remained statistically significant within gender subgroups. The extracted text describes that a multivariate logistic regression was constructed to predict intention to incorporate hallucinogen-assisted therapy, but the extract does not include the regression results or parameter estimates.

Barnett and colleagues interpret the main finding as a marked shift toward more positive attitudes among responding American psychiatrists between 2016 and 2022–2023: lower concern about risks and greater enthusiasm for the therapeutic promise of classic hallucinogens. The study team emphasised strong support for further research and for federal funding of clinical trials, and noted that about half of respondents would intend to incorporate hallucinogen-assisted therapy into their practice if regulatory approval were obtained, a noteworthy level of enthusiasm given the logistical and reimbursement challenges associated with delivering these therapies. The authors comment on the difference in optimism for psychiatric disorders versus SUDs, suggesting this may reflect the smaller contemporary clinical-trial evidence base for SUDs and possible concerns about addictive potential. They note prior work showing that worries about addictive potential were associated with reduced support for hallucinogens as SUD treatments, and that research engagement and greater knowledge were associated with more favourable views. In discussing safety perceptions, the investigators state that only a minority of respondents believed hallucinogens increase risk for subsequent psychiatric disorders or long-term cognitive impairment and that these beliefs have limited support in population and clinical studies cited by the authors. They also observe that fewer than 10% of respondents believed hallucinogens cannot be safely administered under medical supervision and that respondents generally favoured medical legalisation despite current Schedule I classification. The authors acknowledge several limitations that constrain interpretation. The low response rate (13.1%) raises risk of selection bias; respondents with favourable views may have been more likely to participate, which limits generalisability to all American psychiatrists. The anonymous design prevented definitive exclusion of overlap with the 2016 respondents, though the investigators argue this is unlikely to be a substantial problem. They also note potential variability in respondents' familiarity with what constitutes a ‘‘classic’’ hallucinogen despite example lists in the instructions, and that the 2022–2023 sample (N = 131) was smaller than the 2016 sample (N = 324), reducing statistical power for subgroup comparisons and widening confidence intervals. The authors present a specific example showing that a previously observed gender difference in 2016 was not detected in 2022–2023 and that non-equivalence testing indicates substantial uncertainty remains. Finally, they stress the multiple-comparisons issue and their decision not to apply formal correction, advising caution when interpreting p values. In terms of implications, the investigators suggest younger psychiatrists may become the principal providers of hallucinogen-assisted therapy if regulatory approval occurs, given that older respondents expressed more reservations. They recommend that future surveys track changing attitudes domestically and internationally, and explore clinicians' concerns about addictive potential, patient safety, and financial and logistical aspects of delivering hallucinogen-based treatments.

In this repeat national survey of American psychiatrists, respondents in 2022–2023 reported significantly more positive attitudes toward the therapeutic potential of classic hallucinogens and significantly lower concern about their risks compared with 2016. The sample more strongly favoured medical legalisation, further research and federal funding for clinical trials, and showed less support for criminalising non-medical use. Approximately half of respondents indicated an intention to incorporate hallucinogen-assisted therapy into their practice if federal regulatory approval were granted, suggesting readiness among many psychiatrists to adopt these treatments pending approval.