This cross-sectional study (n=3,094) of US young adults aged 19–30 found that around 1 in 15 reported past-year hallucinogen microdosing, with microdosers showing substantially higher rates of alcohol, cannabis, and other substance use compared to non-microdosers, and Black respondents being significantly less likely to microdose than White respondents.
Introduction
This paper aimed to determine the prevalence, demographic correlates and co-occurring substance use patterns of hallucinogen microdosing among young adults in the United States.
Methods
Design: Cross-sectional analysis of data from a nationally-representative cohort study collected in 2022–2023. Setting: United States. Participants: Sample of 3094 young adults aged 19–30 years in the Monitoring the Future panel study. Measurements: Self-reported past-year hallucinogen use and at least one past-year incident of microdosing, demographic characteristics (age, sex, race/ethnicity, college attendance, parental education) and other substance use (alcohol, cannabis, nicotine, other drugs).
Results
Past-year hallucinogen use was reported by 9.5% (SE = 0.68) of young adults, with microdosing reported by 6.8% (SE = 0.61). Among those who used hallucinogens, 73.1% (SE = 3.6) engaged in microdosing. Individuals who reported microdosing demonstrated substantially higher rates of other substance use, with odds ratios ranging from 2.53 (95% CI 1.43–4.47) for past-month cigarette use to 37.73 (95% CI 19.72–72.21) for 3+ occasions of past-year cannabis use. Among those who microdosed, 72.4% reported 10+ occasions of past-year alcohol use and 85.8% reported 3+ occasions of past-year cannabis use. There were few significant demographic differences in microdosing, though Black respondents were less likely (OR = 0.43, 95% CI 0.21–0.90) to microdose compared with White respondents.
Discussion and Conclusions
Approximately 1 in 15 US young adults reported microdosing hallucinogens, with strong associations between microdosing and other substance use. Despite growing interest in potential therapeutic applications of microdosing, the context in which microdosing typically occurs, including patterns of other drug and alcohol use, raises concerns and warrants focused prevention efforts.
Papers cited by this study that are also in Blossom
Polito, V., Stevenson, R. J. · PLOS ONE (2019)
Kuypers, K. P. C. · Therapeutic Advances in Psychopharmacology (2020)
Rosenbaum, D., Weissman, C. R., Anderson, T. et al. · Journal of Psychopharmacology (2020)
Rootman, J. M., Kryskow, P., Harvey, K. et al. · Scientific Reports (2021)
The paper situates microdosing—regular consumption of very small amounts of psychedelics such as LSD or psilocybin—as a growing practice purported to produce subtle changes in mood and behaviour and to be used by some for self-management of depression, anxiety or substance-use problems. The authors note that despite survey evidence suggesting high lifetime incidence of microdosing among people who have tried psychedelics internationally, population-level prevalence and the epidemiological profile of microdosing in the United States remain poorly characterised. Young adults (aged 19–30) are identified as a key group because hallucinogen use concentrates in this developmental period and prior Monitoring the Future (MTF) data show increases in non-LSD hallucinogen use in recent years. The study sets out to estimate the national prevalence of past-year hallucinogen microdosing among US young adults, describe demographic correlates, and document co-occurring patterns of other substance use. The researchers focus on MTF panel respondents aged 19–30 surveyed in 2022–2023 and aim to compare characteristics of those who report any past-year microdosing with those who do not, and to contrast microdosing with any past-year hallucinogen use more broadly.
The study used cross-sectional analyses of data from the Monitoring the Future (MTF) panel, a nationally representative cohort initially sampled in 12th grade and followed biennially at modal ages 19/20, 21/22, 23/24, 25/26, 27/28 and 29/30. Participants included respondents who were in 12th grade from 2010–2022, selected for the panel, and who completed survey versions containing the relevant items at ages 19–30 in 2022 or 2023. The initial analytic frame comprised N = 3177 respondents; item-level missingness reduced sample sizes for specific analyses (maximum analytic N = 3094 for hallucinogen use after 83 missing, and maximum analytic N = 2860 for microdosing after 234 missing). Response rates for the longitudinal follow-ups averaged 61.5% across modal ages. Measures: Past-year hallucinogen use was assessed for LSD and other hallucinogens (examples given included psilocybin mushrooms) and respondents were asked whether they had "micro-dosed (took a very small amount of) a hallucinogen." Microdosing and hallucinogen use variables were dichotomised as any occasions in the past 12 months versus none. Other substance measures captured number of occasions in the past 12 months for alcohol and various modes of cannabis use (smoking, vaping, edibles/concentrates), nicotine vaping, and a combined category for any past-year amphetamine, cocaine or tranquilliser use. Past-30-day cigarette use and past-two-week binge drinking (5+ drinks) were also recorded and dichotomised. Demographic variables were taken from the 12th-grade survey (sex, race/ethnicity, parental education) with sex restricted to male/female analyses due to small counts in other categories. Race/ethnicity was recoded into mutually exclusive groups (Hispanic; non-Hispanic White; non-Hispanic Black; multi-race/other). Parental education was coded as at least one parent with a college degree versus not; college attendance at ages 19–30 was also queried. Statistical analysis: The researchers estimated weighted prevalence and demographic distributions for past-year microdosing versus not, and for any hallucinogen use, using analysis weights that accounted for selection probabilities, the complex survey design and predictors of participation. Logistic regression models produced odds ratios (ORs) and 95% confidence intervals (CIs) for demographic correlates. Prevalence of other substance use among microdosers versus non-microdosers (and among any hallucinogen users versus non-users) was examined, with multinomial logistic regressions reported as ORs and 95% CIs. Analyses were conducted using SAS survey procedures. The analyses were not pre-registered and item-missing data were excluded from analyses pertaining to those items.
Weighted prevalence estimates indicated that 9.5% (SE = 0.68) of the analytic sample aged 19–30 reported past-year hallucinogen use and 6.8% (SE = 0.61) reported past-year microdosing. Among respondents reporting any past-year hallucinogen use, 73.1% (SE = 3.6) reported at least one instance of microdosing in the past year. Demographic correlates of microdosing were broadly similar in direction and magnitude to those for any hallucinogen use. For microdosing, there were no statistically significant differences by sex, age, 4-year college attendance or parental education. Non-Hispanic Black respondents were less likely than non-Hispanic White respondents to report microdosing (OR 0.43, 95% CI 0.21–0.90). For any hallucinogen use, females were less likely than males to report use (OR 0.62, 95% CI 0.46–0.85), and Black respondents were less likely than White respondents (OR 0.35, 95% CI 0.19–0.66). Prevalence of microdosing and any hallucinogen use did not differ significantly between 2022 and 2023. Substance-use correlates showed substantially higher prevalence of other substance use among those who reported microdosing compared with those who did not. Nearly all analysed substances were more common among microdosers, with the exception of a single occasion of past-two-week binge drinking. Reported associations (ORs) ranged from 2.53 (95% CI 1.43–4.47) for any past-30-day cigarette use versus none, to 37.73 (95% CI 19.72–72.21) for reporting 3+ occasions of cannabis use in the past year versus none among microdosers compared with non-microdosers. As examples provided in the text: 72.4% of those who reported microdosing had 10+ occasions of past-year alcohol use (compared with 2.4% with 0 occasions within that group), while among those who did not report microdosing 44.9% had 10+ occasions and 17.3% had 0 occasions; the corresponding OR for 10+ versus 0 occasions of alcohol use was 11.82 (95% CI 3.73–37.44). For cannabis, 85.8% of microdosers reported 3+ occasions in the past year versus 5.3% reporting 0 occasions; among non-microdosers 27.0% reported 3+ occasions and 63.2% reported 0 occasions; the OR for 3+ versus 0 occasions of cannabis use was 37.73 (95% CI 19.72–72.21). The pattern and magnitude of associations for any hallucinogen use were similar to those for microdosing, with ORs ranging up to 26.26 (95% CI 14.56–47.35) for high-frequency cannabis use.
The authors interpret their findings as evidence that hallucinogen microdosing is relatively common among US young adults: roughly 1 in 15 reported past-year microdosing, and nearly three-quarters of past-year hallucinogen users reported at least one microdosing incident. They emphasise that people who report microdosing also report high levels of other substance use, particularly frequent alcohol and cannabis use and nicotine vaping, indicating that microdosing typically occurs in the context of polysubstance use rather than in isolation. The researchers position these results against limited clinical evidence for microdosing and broader hallucinogen research. They note that although some trials and earlier studies have reported modest mood benefits from microdosing, trial evidence is limited (small samples, inconsistent controls) and some trials have observed increases in anxiety. They also point out that many clinical protocols for therapeutic hallucinogen use employ substantially higher doses than those described as microdosing. Given these uncertainties, the high prevalence of co-occurring substance use among microdosers raises concerns that any putative benefits could be offset by harms associated with concurrent use of substances with established health risks. The authors acknowledge several key limitations: reliance on self-reported data which may be affected by recall bias or misunderstanding (respondents may not accurately identify a dose as a "microdose"); absence of information on frequency, timing, duration, motivations for microdosing, or simultaneous use with other substances, precluding causal or temporal inference about relationships between microdosing and other substance use; and limited generalisability because the MTF panel includes individuals originally sampled in high school (excluding those who dropped out) and is subject to differential attrition. They also note that subgroup sample sizes were small for some comparisons, producing wide confidence intervals. At the same time, the researchers highlight that the inclusion of a microdosing item in a nationally representative survey is a unique strength and argue that continued surveillance is important, particularly amid policy shifts in some US jurisdictions that decriminalise hallucinogens. Finally, the authors call for further research—particularly robust randomised trials focused specifically on microdosing regimens and more detailed epidemiological work on patterns, motivations and co-use—to better characterise potential benefits and harms. They suggest prevention and education efforts should treat microdosing within the broader context of polysubstance use given the strong co-occurrence observed.
MTF sampled participants in the United States annually and assessed adolescent and adult substance use and other health behaviours. Initially, surveys were completed among students in the 12th grade by nationally-representative samples in school; a subset of respondents were then selected for longitudinal biennial follow-up in the MTF panel. Participants were probabilistically selected for longitudinal follow-up proportional to 12th grade sampling probabilities and enriched for 12th grade substance use. MTF panel participants who were selected for the longitudinal study were randomly assigned to begin follow-up at either age 19 or age 20; they were interviewed biennially thereafter (i.e., follow-ups are at modal ages 19/20, 21/22, 23/24, 25/26, 27/28, 29/30). Response rates beginning in the longitudinal follow-up at 19/20 were mean 61.5% (range 57.7% at age 27 to 76.8% at age 21). Additional information on the MTF panel design can be found elsewhere. The present study was based on respondents who participated in 12th grade from 2010 to 2022, were selected for the MTF panel, and responded to survey versions that included relevant variables when surveyed between ages 19 and 30 in 2022 or 2023 (N = 3177). Among those who were followed-up, item missing data was minimal; for example, 83 respondents were missing data on hallucinogen use (maximum analytic sample size of 3094) and of those, 234 were missing on microdosing (maximum analytic sample size of 2860). Demographic distributions and other item-level missingness are described in Table. Those with item-level missing data were removed from analysis pertaining to those items. Analyses were not pre-registered and thus should be considered exploratory.
Respondents were queried regarding the frequency of hallucinogen use in the past 12 months, including LSD; hallucinogen or psychedelic drugs ('like PCP, mescaline, peyote, "shrooms" or psilocybin'); and were asked if they 'micro-dosed (took a very small amount of) a hallucinogen (like LSD, psilocybin mushrooms, "shrooms", peyote, etc.)'. Responses were dichotomised into any occasions in the past 12 months versus none.
Other substances were queried in various time frames. The number of occasions of past 12-month use was reported for alcohol use, cannabis smoking, cannabis vaping and other cannabis use (eating in food, drink in a beverage, use a concentrate [such as 'wax', 'budder' or 'shatter'] and other methods), and nicotine vaping. The number of days of past 30-day cigarette use was queried (dichotomized as any days versus no days). Past two-week binge drinking (5+ drinks in a row) was dichotomised as any occasions versus none. A combined category of other drug use included any past-12-month amphetamine, cocaine or tranquilliser use.
Data on sex, race and ethnicity and parental education were provided at 12th grade. Sex was asked as, 'What is your sex?' (male, female or [starting in 2021] other or prefer not to answer). Due to the small number of respondents who selected other or prefer not to answer, sex analyses included only those selecting male or female. Respondents were allowed to check multiple options for race and ethnicity, and we recoded into mutually exclusive groups based on available sample size: Hispanic, non-Hispanic White, non-Hispanic Black and multi-race/other race. Respondents were asked about each parent's highest level of education; responses were combined based on whether at least one parent had a college degree or higher versus not. At ages 19-30, respondents were queried about whether they attend or attended a 4-year college. Age was coded based on the modal age at time of survey (19-30).
Demographic distributions among those who reported past-year microdosing versus not were estimated, and logistic regressions were estimated generating odds ratios and 95% confidence intervals. Similar models were estimated for any past-year hallucinogen use. The prevalence of other substance use was estimated among those who reported past-year microdosing versus not, and among those who reported any past-year hallucinogen use versus not, and multinomial logistic regressions were estimated generating odds ratios and 95% confidence intervals. Analysis weightsaccounting for selection probabilities, complex survey design and a wide range of individual characteristics associated with the probability of participation at each modal age (e.g., 12th grade demographics, substance use, grades, truancy;) were incorporated into all analyses using SAS survey procedures.
Past-year hallucinogen use was reported by 9.5% (SE = 0.68) of the analytic sample age 19-30, and past-year microdosing was reported by 6.8% (SE = 0.61). Of those who reported pastyear hallucinogen use, the majority, 73.1% (SE = 3.6), reported microdosing. Tableshows the demographic correlates of past-year microdosing and any hallucinogen use. Correlates were largely similar in direction and magnitude between microdosing and any hallucinogen use. For microdosing, no significant differences in microdosing emerged by sex, age, college status or parental education. Black respondents were significantly less likely (odds ratio [OR] 0.43, 95% confidence interval [CI] 0.21-0.90) to microdose compared with White respondents. For any hallucinogen use, females (OR 0.62, 95% CI 0.46-0.85) and Black respondents (OR 0.35, 95% CI 0.19-0.66) were significantly less likely to use than males and White respondents, respectively. Prevalence of use did not significantly differ between 2022 and 2023 for microdosing or any hallucinogen use. Tableshows the substance use correlates of microdosing and any hallucinogen use. All substances analysed were significantly more prevalent among those who reported microdosing compared with those who did not, with the exception of 1 occasion of past twoweek binge drinking compared with no occasions. Significant odds ratios ranged from 2.53 (95% CI 1.43-4.47) for past 30-day cigarette use vs. none, to 37.73 (95% CI 19.72-72.21) for 3+ occasions of cannabis use in the past year vs. none, among those who reported pastyear microdosing compared with those who did not. For example, the likelihood of 10+ occasions (vs. 0 occasions) of past-year alcohol use was significantly higher among those who reported microdosing (72.4% vs. 2.4%) than those who did not (44.9% vs. 17.3%); OR 11.82, 95% CI 3.73-37.44. Cannabis use was also highly concentrated in those who microdosed; the likelihood of 3+ occasions of cannabis use in the past year (vs. 0 occasions) was substantially higher among those who microdosed (85.8% vs. 5.3%) than those who did not (27.0% vs. 63.2%), OR 37.73, 95% CI 19.72-72.21. Also shown in Tableare the substance use correlates of any hallucinogen use. The magnitude and range of prevalence and odds ratios were similar to those for microdosing, with odds ratios ranging from 2.28 (95% CI 1.44-3.59) for 1 occasion of past two-week binge drinking vs. none to 26.26 (95% CI 14.56-47.35) for 3+ occasions of cannabis use in the past year vs. none among those who reported past-year hallucinogen use compared with those who did not.
The present study demonstrates in a national sample of young adults that instances of microdosing hallucinogens are prevalent, with approximately 1 in 15 young adults, and 2 in 3 young adults who used hallucinogens reporting at least one past-year incident of microdosing. Those who report microdosing represent a group in which other substance use is also concentrated, with a large majority reporting 10+ occasions of alcohol and cannabis use as well as nicotine vaping, as well as higher odds of other drug use. Both demographic and other substance use correlates were similar between those who used any hallucinogens and those who microdosed, indicating that microdosing has a similar profile as has been previously demonstrated for hallucinogen use more generally. Further, these results underscore that while the safety and health consequences of microdosing remain understudied, those who microdose are at higher risk for adverse health outcomes given that substance use with known adverse health consequences frequently co-occur. Despite growing interest in microdosing for purported benefits, including enhanced creativity and improved mood, rigorous scientific evidence supporting these claims remains limited. Hallucinogens commonly used for microdosing such as psilocybin remain Schedule I controlled substances, indicating no accepted medical use and high potential for abuse. Research on potential therapeutic uses of hallucinogens for conditions such as post-traumatic stress disorder, depression and anxiety, and addiction has broadly demonstrated initial efficacy, though there may be variation in doses and protocols often include higher doses than the amounts consumed in microdosing. However, small sample sizes, inconsistent control conditions, inability to isolate pharmacological effects in the context of intensive psychological support and limited longterm data remain challenges to understanding potential benefits. Fourteen randomised blinded controlled trials of microdosing protocols have been conducted with small sample sizes, and results indicate modest increases in mood enhancement, although elevation in anxiety as well. Thus, the clinical utility of microdosing remains relatively unknown, and future research with robust randomised designs specifically focused on microdosing regimens would be a valuable addition to the field. Our findings raise concerns about the substantially higher rates of polysubstance use among those who microdose. While some research suggests microdosing may have therapeutic potential for psychiatric and substance use disorders, high levels of other substance use may reduce any potential health benefits. It has long been demonstrated that use of different psychoactive substances is correlated (including that individuals who use psychedelics frequently use other substances as well), given that individuals who use substances may be more open to experimenting with a variety of products and may be more likely to be in situations where multiple products are available. We demonstrate here that these well-documented correlations are also apparent for those who microdose hallucinogens, although it is important to note that we did not have data on simultaneous use of microdosing with other substances. Nevertheless, in the context of recent policy changes in select US jurisdictions to decriminalise hallucinogen possession and use, continued surveillance of hallucinogen use and correlations with other drug use remains pertinent as availability potentially increases. Several limitations of this study should be noted. First, data are selfreported, which may be subject to recall bias and misunderstandings of questions, including that respondents may not accurately report or know whether their dosage was a 'microdose' of hallucinogens. Second, the MTF survey did not query frequency, pattern, timing, duration or motivation for microdosing. Associations were based on concurrent reports of recent drug use and there were no data on the extent to which substances were used simultaneously, precluding assessment of temporality regarding the relationship between microdosing and other substance use patterns. While this is a significant limitation, MTF is also the only nationally representative survey of drug use that queries microdosing, thus the unique addition to the survey is a strength. Nevertheless, concentrated epidemiological research on microdosing patterns is needed. Third, these data are based on young adult longitudinal follow-ups of participants initially surveyed in high school; thus, those who dropped out of high school are not included and differential attrition presents a challenge to generalisability. Further, while our sample size, close to 3000 respondents, is among the largest for longitudinal studies of drug use ongoing in the United States, sample sizes were small for some subgroups (e.g., those who both report microdosing and high levels of other drug use) resulting in wide confidence intervals for some comparisons; additional research, including with MTF data as we accumulate more cohorts, is needed to confirm patterns observed here. In summary, this national study provides the first epidemiological evidence on microdosing hallucinogen prevalence among young adults in the United States, with approximately 1 in 15 reporting past-year use, and demonstrates that microdosing is strongly associated with polysubstance use patterns that warrants further investigation. Microdosing is an understudied practice, and polysubstance associations highlight the need for comprehensive approaches to substance use prevention and education that address microdosing of hallucinogens within broader patterns of polysubstance use.
Create a free account to open full-text PDFs.
Keyes, K. M., Patrick, M. E. · Addiction (2023)
van der Meer, P. B., Fuentes, J. J., Kaptein, A. A. et al. · Frontiers in Psychiatry (2023)
Johnson, M. W., Griffiths, R. R., Hendricks, P. S. et al. · Neuropharmacology (2018)
Aday, J. S., Mitzkovitz, C. M., Bloesch, E. K. et al. · Neuroscience and Biobehavioral Reviews (2020)
Reiff, C. M., Richman, E. E., Nemeroff, C. B. et al. · American Journal of Psychiatry (2020)
Murphy, R., Muthukumaraswamy, S., De Wit, H. · Biological Psychiatry (2024)