This prospective survey study (n=247) finds that those who microdose psychedelics to manage ADHD symptoms experience benefits from it. Participants scored higher on well-being after two and four weeks. The study design, a prospective survey, makes it possible to draw causal inferences (microdosing causing the improvement), but the study had a large drop-out rate (n=46 at 4 weeks).
ADHD in adulthood is often overlooked, which negatively affects the individual’s well-being. First-line pharmacological interventions are effective in many ADHD patients, relieving symptoms rapidly. However, there seems to be a proportion of individuals who discontinue, or fail to respond to these treatments. For these individuals, alternative treatment options should be explored. A retrospective survey study reported that using classic psychedelics in low, repeated doses, so called microdosing (MD), was more effective than conventional treatments for ADHD. The current prospective study aimed to measure the effect of MD on ADHD symptoms, well-being and temporal processing, since this cognitive domain is often impaired in ADHD. Adults with ADHD who had the intention to start MD on their initiative to self-treat their symptoms were measured before MD and two- and four weeks later. We expected a decrease in ADHD symptoms, an increase in well-being, and enhanced performance on a time perception task after MD. We explored if conventional medication use alongside MD and comorbidities alongside ADHD influenced the effect of MD. Sample sizes included N=226, N=65, and N=46, respectively. We found decreases and increases in ADHD symptoms and well-being, respectively. Time perception was only affected in individuals using conventional medication alongside MD, by over-reproducing the 1000 ms time interval. Conventional medication use and having comorbidities did not change the effect of MD on ADHD symptomatology and well-being after four weeks of MD. Placebo-controlled experimental studies are needed to explore whether there is a beneficial effect of MD for ADHD, beyond the placebo-effect.
Papers cited by this study that are also in Blossom
Hutten, N. P. W., Mason, N. L., Dolder, P. C. et al. · International Journal of Neuropsychopharmacology (2019)
Hutten, N. R. P. W., Mason, N. L., Dolder, P. C. et al. · European Neuropsychopharmacology (2020)
Bershad, A. K., Schepers, S. T., Bremmer, M. P. et al. · Biological Psychiatry (2019)
Yanakieva, S., Polychroni, N., Family, N. et al. · Psychopharmacology (2018)
Haijen and colleagues frame adult attention deficit hyperactivity disorder (ADHD) as a prevalent condition with persistent functional impairments and high comorbidity, and note that many adults either discontinue first-line pharmacological treatments or do not respond adequately to them. The introduction summarises how microdosing (MD) with classical psychedelics (small, repeated doses of LSD, psilocybin, etc.) is reported anecdotally to improve concentration, productivity and mood, and that limited experimental work in neurotypical volunteers has shown small-dose effects on attention and time perception. The authors argue that prospective, behaviourally assessed data are needed in adults with ADHD because prior work has relied largely on retrospective self-reports. The study therefore set out to prospectively assess changes in self-reported ADHD symptoms, well-being, and temporal processing in adults who intended to begin MD on their own initiative. Using a naturalistic design, the investigators hypothesised that ADHD symptoms would decrease and well-being would increase after four weeks of MD compared to baseline, that improvements in ADHD symptoms would correlate with changes in well-being, and that time perception performance would improve. Secondary aims examined whether concurrent use of conventional ADHD medication or presence of psychiatric comorbidity modified outcomes and explored characteristics of participants who did not improve after four weeks.
This was an online, prospective naturalistic survey of adults who intended to start microdosing to self-manage ADHD symptoms. Recruitment occurred via an online advertisement placed on a microdosing information website. Interested individuals provided written informed consent one to three days before initiating MD and completed a baseline survey; follow-up surveys were administered at two and four weeks after enrolment. Participants were asked to keep a diary recording the substances and doses they used during the study period. Surveys could be paused and completed later. Data collection occurred during the COVID-19 pandemic period noted in the paper. Inclusion for analysis required scoring above a clinical cut-off on at least one subscale of the Conners' Adult ADHD Rating Scale—Short Version (CAARS-S:SV), with a t-score ≥ 65 used to indicate clinically elevated symptoms. The sample therefore included adults with a formal ADHD diagnosis and individuals reporting severe ADHD complaints. Baseline measures collected demographic information, psychiatric/neurological/physiological diagnoses (allowing multiple selections), lifetime psychedelic experience, and current or prior use of conventional ADHD medications. Primary and secondary outcomes were assessed at baseline, 2 weeks and 4 weeks. ADHD symptoms were measured with the CAARS-S:SV (subscales: inattention; hyperactivity/impulsivity; ADHD index; plus a DSM-IV total symptom score). Well-being was measured with the WHO-5 Well-Being Index. Temporal processing was assessed with an auditory time reproduction task (TRT) presenting six intervals (1000 ms, 1500 ms, 3200 ms, 3700 ms, 5500 ms, 6000 ms); performance was expressed as a relative difference score (estimated minus actual interval, divided by actual interval). Instructions discouraged explicit counting and participants performed two practice and twelve test trials (each interval twice). Statistical analysis used linear mixed models (LMMs) with Time (0W, 2W, 4W) as a within-subject factor; Interval was an additional within-subject factor for the TRT analysis. Akaike's information criterion determined the best-fitting covariance structure. Missing data were handled via restricted maximum-likelihood estimation. Significant main effects were followed by Bonferroni-corrected pairwise comparisons. An alpha of 0.05 was used. The investigators also calculated frequencies of non-responders (no improvement or worsening in CAARS DSM-IV total t-score) and Pearson correlations between change scores of ADHD symptoms and WHO-5 well-being.
Of 356 individuals who consented and began the survey, 247 completed the baseline survey; after excluding two responses judged non-serious and 12 participants without elevated CAARS t-scores, 233 respondents were analysed. Most participants had prior psychedelic experience (80%). ADHD was reported by 159 participants (68% of the analysed sample), and 54% of those with ADHD reported at least one comorbid diagnosis; depression, anxiety, PTSD and dyslexia were common comorbidities. More than half of the diagnosed participants had previously stopped conventional ADHD medication; current conventional medication use during the study varied. About half of respondents completed the diary item reporting the substance used: among those, 78% reported psilocybin/psilocin, 12% novel lysergamides, 9.5% LSD, and one respondent reported ayahuasca. Self-reported mean doses were reported for these substances but diary completion was incomplete. On the primary outcome, CAARS-S:SV DSM-IV total symptoms t-scores decreased over time (main effect of Time, p = .000, ηp2 reported). Pairwise comparisons showed significant reductions from baseline to 2 weeks (Δ2W–0W = −10.17, p = .000) and to 4 weeks (Δ4W–0W = −15.43, p = .000), and a further reduction between 2 and 4 weeks (Δ4W–2W = −5.26, p = .007). A Time × Medication use interaction was observed (F(3,241.4) = 2.86, p = .038, ηp2 = .034): individuals using conventional ADHD medication alongside MD had higher CAARS scores at 2 weeks (β = 8.7, p = .006), indicating a smaller early decrease, but no difference was found at baseline or 4 weeks. The Time × Comorbidity interaction for the DSM-IV total score was not significant. Analyses of CAARS subscales paralleled the total score findings. Time produced main effects on ADHD index, inattention, and hyperactivity/impulsivity t-scores. Significant Time × Medication use interactions were present for the ADHD index (F(3,247.6) = 3.65, p = .013, ηp2 = .042) and Inattention (F(3,250.6) = 3.45, p = .017, ηp2 = .040), with medication users showing smaller decreases at 2 weeks (ADHD index β = 8.04, p = .001; Inattention β = 10.92, p = .002). A Time × Comorbidity interaction emerged for the ADHD index (F(3,232.7) = 2.73, p = .045, ηp2 = .034) that reflected higher baseline ADHD index scores in those with comorbid diagnoses (β = 3.30, p = .011); no comorbidity effects were found at follow-ups. No Time interactions were observed for the Hyperactivity/Impulsivity subscale. At the four‑week time point, 9 of 47 respondents (19.1%) were classified as non-responders (no improvement or worsening in CAARS DSM-IV total t-score); three of these nine were using conventional medication alongside MD. Some non-responders had transient improvement at 2 weeks and worsened by 4 weeks. Well-being (WHO-5) improved significantly over time (F(2,119.4) = 31.50, p = .000, ηp2 = .345). Well-being scores increased from baseline to 2 weeks (Δ2W–0W = 16.47, p = .000) and to 4 weeks (Δ4W–0W = 17.80, p = .000), with no significant change between 2 and 4 weeks. A Time × Comorbidity interaction indicated lower baseline well-being for participants with comorbid diagnoses (β = −6.61, p = .009), but comorbidity did not alter follow‑up scores. No Time × Medication interaction on WHO-5 was detected. Changes in ADHD symptom scores correlated moderately and negatively with changes in well-being (2W: r = −0.367, p = .003; 4W: r = −0.471, p = .001), indicating that larger reductions in ADHD symptoms were associated with greater increases in well-being. For time perception, there was no main effect of Time on TRT relative difference scores (F(2,59.2) = 0.63, p = .534). Interval length produced the expected effects on reproduction accuracy (longer intervals more likely underestimated). A three-way Time × Interval × Medication use interaction was significant (F(18,97.78) = 2.52, p = .002, ηp2 = .317); follow-up estimates showed that, for the shortest interval (1000 ms), individuals using conventional ADHD medication alongside MD had higher relative difference scores (over-reproduction) at 2 weeks (β = .186, p = .038) and 4 weeks (β = .353, p = .004) compared with those not using conventional medication. No medication effects were found for other intervals. The authors note small sample sizes for medication subgroups at follow-ups (n = 16 at 2W; n = 9 at 4W) and substantial variability in TRT performance, which limit confidence in these specific findings. Attrition was substantial: only 47 respondents completed the four‑week assessment (about 20% of the analysed baseline sample), and diary completion about substances was incomplete, leaving many substance- and dose-related data missing.
Haijen and colleagues interpret the findings as preliminary evidence that self-initiated microdosing is associated with reductions in self-reported ADHD symptoms and increases in well-being over a four‑week period in adults with diagnosed or clinically elevated ADHD symptoms. They emphasise that symptom reductions were observable at two weeks and continued to improve by four weeks, and that symptom change was moderately correlated with enhanced well-being. The authors report that concurrent use of conventional ADHD medication appeared to delay early symptom improvement (at 2 weeks) but did not alter outcomes at 4 weeks; comorbidity did not change MD effects on main outcomes at 4 weeks. Regarding temporal processing, the investigators did not find overall improvement after MD; instead, an over-reproduction of the shortest interval was observed in participants using conventional medication alongside MD, a finding they treat cautiously because of small subgroup sizes and high variability. The discussion situates these results alongside prior retrospective reports and limited experimental findings in healthy volunteers, noting concordance with anecdotal and survey-based claims that MD can aid concentration and mood. The authors consider their prospective, naturalistic design a strength because it reduces some uncertainty inherent in retrospective self-reports and allowed repeated behavioural and self-report assessments using validated instruments. Inclusion of a cognitive task alongside clinical questionnaires is highlighted as an additional asset. Key limitations acknowledged by the investigators include the lack of a placebo-control group and the consequent inability to rule out expectancy/placebo effects, recruitment via a MD website that likely biased the sample towards individuals favourably disposed to MD, self-selection and high dropout (only about 20% of baseline participants completed the four‑week measures), incomplete diary data about the substances and doses used, and limited information about formulation, storage and route of administration. The CAARS-S:SV is based on DSM‑IV criteria and uses older normative data, which the authors note as a limitation. Concerning the time reproduction task, they point out that two trials per interval may have been insufficient to detect subtle MD effects. Finally, small numbers of participants using conventional medication at follow-up limit interpretation of interactions involving medication. For future work, the authors recommend placebo-controlled and blinded studies (including approaches that allow participant blinding in naturalistic settings), standardised dosing and administration protocols, more extensive cognitive testing (working memory, attention, executive function), and exploration of MD effects in other disorders that include attentional symptoms. Despite the caveats, they conclude that this study provides initial prospective evidence suggesting MD may have therapeutic value for adults with ADHD or severe ADHD complaints, and that controlled trials are needed to confirm the magnitude and specificity of the effects.
This online prospective naturalistic survey study aimed to assess changes in ADHD symptoms, well-being, and time perception using validated questionnaires and a time reproduction task in individuals with an ADHD diagnosis or severe ADHD complaints, who started MD on their own initiative. The primary hypothesis that MD would reduce ADHD symptoms was confirmed, as findings showed decreased (self-report) ADHD symptoms after two weeks of MD, with additional decrements two weeks later. Using conventional ADHD medication seemed to delay the decrease in ADHD symptoms after MD. In line with expectations, increased well-being was reported at two and four weeks after MD. Additionally, MD-related changes in well-being and ADHD symptoms J o u r n a l P r e -p r o o f were negatively associated. Using conventional medication alongside MD, or having comorbidities alongside ADHD, did not change the effect of MD on ADHD symptoms and wellbeing after four weeks of MD in the current study. Lastly, time perception seemed to be altered after MD for individuals using conventional medication, illustrated by an over-reproduction of the shortest (1000 ms) time interval used in a time reproduction task. The results do not find support for the hypothesis that performance on a time perception task would be improved after MD in individuals with an ADHD diagnosis or severe ADHD complaints. The decrease in ADHD symptoms after MD was in line with earlier findings showing that MD as self-medication used by people diagnosed with ADHD was rated as being more effective than conventional treatments and increasing their quality of life. Also, the findings were in line with anecdotes of individuals who microdosed to self-treat their ADHD. The strength of the present study over retrospective reports is that the current design allows causal inferences to be made about MD and the observed and self-rated effects. Based on this, it can be said with more certainty that MD could be beneficial and of therapeutic value for individuals diagnosed with ADHD or having severe ADHD complaints, even in addition to first-line pharmacological interventions. After four weeks of MD, mean CAARS-S:SV t-scores were below the used cut-off of 65 for three out of four subscales. The current study sample showed similar changes in the CAARS-S:SV DSM-IV total symptoms scores compared to studies investigating the effects of several weeks of mindfulness-based cognitive therapyand treatment with methylphenidatein adults diagnosed with ADHD. Almost twenty percent of the sample at the four-week time point did not show improvements in ADHD symptoms. This lack of improvement did not seem related to using conventional medication alongside MD or having comorbidities alongside ADHD. Other aspects most likely underlay the lack of improvement in the nonresponding participants. A potential explanation might be that individuals had difficulties determining the right dose, a question that future controlled dose-titration studies could investigate. Well-being was increased after two weeks of MD compared to baseline and remained elevated two weeks later. These scores were put into context by comparing them to normative data collected during the COVID-19 pandemic because the current data was collected within the same period. Compared to the normative dataset collected from almost 15 thousand respondents from 14 countries, the current study sample reported low well-being scores at baseline (mean WHO-5: 42.7). After two and four weeks of MD, the current sample showed well-being scores (mean WHO-5 (2W): 59, mean WHO-5 (4W): 60) that were more in line with the average well-J o u r n a l P r e -p r o o f being scores of West-European countries during the COVID-19 pandemic. This shows that, during MD, the sample evolved from below-average to average well-being scores. No support was found for improved performance on a time perception task after MD. However, a decrease in performance was found after MD compared to baseline in individuals using conventional medication. The hypothesis was based on previous research showing that individuals diagnosed with ADHD tended to underestimate presented time intervalsand a study showing that microdoses of LSD led to an over-reproduction of 2000-4000 milliseconds time intervals. The only effect of MD on time perception found here included an overestimation of the 1000 milliseconds interval for individuals using conventional ADHD medication next to MD, after two-and four weeks of MD. This finding should be interpreted with caution, given the small number of respondents using conventional ADHD medication at the two-(n = 16; 25% of respondents at 2W) and four-week time points (n = 9; 20% of respondents at 4W). Furthermore, the large variability in the accuracy of time estimations suggests that two trials per time interval were potentially not enough to capture a potential main effect of MD. The effect found in some of the respondents in the current study provides enough reason to further investigate the possible impact of MD on time perception in this population in a controlled environment and using more trials in the time perception task. Strengths of the current study included using a naturalistic, prospective design. This allowed drawing causal inferences with less uncertainty compared to asking retrospectively about individuals' MD experiences. Widely used, validated questionnaires were used to assess ADHD symptoms and well-being, which allowed comparison across studies investigating conventional treatments and comparison to normative non-clinical data. In addition, the inclusion of a cognitive task combined with the subjective self-report questionnaires provided the opportunity to assess the effects of MD in adults diagnosed with ADHD or having serious ADHD complaints on a clinical, psychological and cognitive level. Although the current study design provided an easy way to collect data and observe the effects of MD in an ADHD population without manipulating MD practices, it comes, of course, with its limitations. Due to lacking a placebo-control group, the design could not test whether the MD group behaved differently than a group receiving a placebo. In the current study, participants were recruited via a MD website and respondents had the intention to microdose to self-treat their ADHD symptoms. It can only be expected that individuals who choose MD as self-treatment are positively oriented towards the practice of MD, which might enhance a potential placebo effect. Future research could instruct participants on how to blind themselves, including a placebo-control group, as has been done recently by Szigeti and colleagues. When J o u r n a l P r e -p r o o f comparing the effectiveness of MD for ADHD to other interventions for ADHD, it should be noted that the participants of the current study chose MD to self-treat their ADHD complaints on their own initiative, while in controlled, experimental studies participants are assigned to a treatment. In both cases, participants choose to be enrolled in research, however, it can be suggested that choosing a treatment may increase the positivity about the treatment and the treatment's efficacy. Furthermore, participants enrolling themselves in the study might have led to selfselection bias. Perhaps only individuals who were willing to adhere to completing several questionnaires participated in the study, resulting in a study sample that may not be representative of the general ADHD population. Further, the study suffered from a large dropout rate, with only 20% of the sample at baseline completing the four-week time point, which is common in prospective survey studies. Perhaps only those with a positive MD experience continued in the study, biasing the findings. Future studies could follow up and ask participants about their reason to drop out. Furthermore, participants were requested to report in a diary what substance they had used to microdose. About half of the respondents at baseline did not fill in the diary, leading to a large amount of missing data about the substances and doses that were used during the study. No additional information about the substances was collected, such as the formulation, storage conditions, and route of administration, which may be important factors influencing experienced effects. Lastly, the CAARS-S:SV originates from 1999, therefore the change from DSM-IV to DSM-5 has not been taken into account by using this scale. The DSM-5 devotes more attention to ADHD in adulthood, by including more examples of how ADHD is expressed in adults and by raising the age of symptom onset. Also, the normative data of the CAARS-S:SV could be considered outdated. Future research should include a more up-to-date ADHD scale taking into account the changes from DSM-IV to DSM-5 and more recent norm scores. Future placebo-controlled studies are needed to assess the possible therapeutic value of MD in adults diagnosed with ADHD. In this context, it would be interesting to compare another type of practice/therapy as self-treatment to MD, to assess if MD potentially has effects superior to the effect of treatment choice and placebo in adults self-treating their ADHD symptoms. Further, a more standardized test administration should be used to assess the effects of MD on time perception in adults with ADHD. Other cognitive domains that are impaired in ADHD, such as working memory, attention, and executive functioning, should also be investigated in future studies investigating the effects of MD in adults with ADHD, to assess in what domain MD might exert beneficial effects. Lastly, since inattention, hyperactivity and impulsivity symptoms may occur in other disorders, such as substance use disorder, bipolarand borderline J o u r n a l P r e -p r o o f personality disorder, future studies should investigate the effects of MD in these populations. To conclude, the present study provides the first evidence that MD may have therapeutic value in adults diagnosed with ADHD or experiencing severe ADHD complaints. Given the limitations of the current study design, studies including placebo-treated and/or other control groups could confirm the magnitude of the therapeutic effect of MD in ADHD. Age (mean ± standard deviation) ☐ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☒ The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: KPC Kuypers reports a relationship with Mind Medicine Inc. that includes: funding grants. KPC Kuypers reports a relationship with The Beckley Foundation that includes: funding grants.
Create a free account to open full-text PDFs.
Fadiman, J., Korb, S. · Journal of Psychoactive Drugs (2019)
Andersson, M., Kjellgren, A. · Harm Reduction Journal (2019)
Szigeti, B., Kartner, L., Blemings, A. et al. · eLife (2021)
Hübner, S., Haijen, E. C. H. M., Kaelen, M. et al. · Journal of Medical Internet Research (2021)
Papers in Blossom that reference this study
Mueller, L., de Jesus, N. M. S., Schmid, Y. et al. · JAMA Psychiatry (2025)
Hutten, N. R. P. W., Quaedflieg, C. W. E. M., Mason, N. L. et al. · Translational Psychiatry (2024)
Polito, V., Liknaitzky, P. · Journal of Psychopharmacology (2024)
Haijen, E. C. H. M., Hurks, P. P. M., Kuypers, K. P. C. · European Psychiatry (2024)
Haijen, E. C. H. M., Hurks, P. P. M., Kuypers, K. P. C. · Frontiers in Psychiatry (2023)