Bipolar depression can be profoundly disabling, but successful treatment requires more than a rapid improvement in depressive symptoms. Recovery also depends on preserving sleep, judgement and stability across time. Psilocybin has produced encouraging antidepressant findings in small studies of bipolar II depression; its capacity to destabilise mood remains a central clinical concern.
Early findings justify carefully controlled investigation in selected patients with bipolar II depression. They do not establish psilocybin as routine treatment for bipolar disorder, demonstrate safety in bipolar I disorder, or show that an apparently positive psychedelic experience protects against a later mood episode.
On this page
- Why the bipolar diagnosis changes treatment
- Antidepressant findings in bipolar II depression
- Recovery and activation are different outcomes
- What experiences outside trials reveal
- Medication interactions and the danger of stopping treatment
- Monitoring must extend beyond the session
- What remains unresolved
Why the bipolar diagnosis changes treatment
Bipolar II disorder involves episodes of depression and hypomania. Bipolar I disorder involves at least one manic episode, which may include severe impairment or psychosis. Bipolar II should not be understood as an inconsequential version of the illness: its depressive episodes can carry substantial disability. A history of increased energy, reduced need for sleep and unusual activity can be missed when assessment concentrates on the current depression.1
The distinction matters because depression and activation must be assessed together. More energy accompanied by ordinary sleep, improved concentration and restored daily functioning has a different meaning from escalating activity accompanied by little need for sleep, impulsive decisions or an unusually expansive sense of capability. A depression score alone cannot make that distinction.
Assessment therefore needs a longitudinal account of mood, previous treatment responses and periods of activation. Information from someone who knows the patient well can be valuable, with consent. Current mixed symptoms, recent hypomania, previous mania, substance use and changes in medication alter the clinical question; evidence from a selected bipolar II depression sample cannot simply be transferred to each of these circumstances.2
Antidepressant findings in bipolar II depression
A nonrandomised study of 15 adults with treatment-resistant bipolar II depression combined a single psilocybin session with psychotherapy. The mean Montgomery–Åsberg Depression Rating Scale (MADRS) score fell by 24 points at three weeks, and improvement persisted during the 12-week observation period. Measures of mania, hypomania and suicidality did not increase. These were substantial changes in a small, selected group, but there was no control condition to separate drug effects from psychological treatment, expectation or the course of the episode.3
A separate report described four participants with bipolar II depression treated within a wider programme for treatment-resistant depression. Depression ratings improved, while mania ratings remained stable; no mania, hypomania or psychosis was reported. This study allowed existing conventional mood stabilisers. Four participants provide useful feasibility observations, but almost no precision about uncommon adverse events or which patients are likely to benefit.4
A further open-label trial, published in September 2026, enrolled 14 participants and offered a second session when depression persisted. Depression scores declined after treatment. However, three participants experienced notable psychiatric adverse events involving suicidal ideation or hypomania; these resolved with support, and no serious adverse events were reported. The study used adapted diagnostic criteria allowing shorter hypomanic episodes, which affects comparison with other samples. Because allocation to the second session depended on continuing symptoms, its results do not establish that a higher dose or repeated treatment is superior.5
Taken together, these findings support an antidepressant signal and a need for further safety evaluation. The absence of mood switching in one small sample cannot cancel its occurrence in another. Nor can symptom changes from separate uncontrolled studies be combined into a reliable personal probability of remission.
Recovery and activation are different outcomes
| Clinical outcome | What needs to be established |
|---|---|
| Depressive symptoms | Whether low mood, hopelessness and loss of interest improve, including for people who do not complete treatment. |
| Mood stability | Whether improvement occurs without hypomania, mania, mixed symptoms or increased cycling. |
| Sleep and behaviour | Whether sleep remains adequate and activity, spending, relationships and decision-making remain proportionate. |
| Sustained recovery | Whether benefits persist through ordinary stresses and subsequent months, with fewer relapses and preserved function. |
These are complementary measures of treatment success, not interchangeable ones. A person may report relief while relatives observe a marked change in behaviour. Conversely, a return of healthy motivation should not automatically be labelled hypomania. Clinical interpretation depends on the pattern, duration and consequences of the change, rather than the intensity of a single positive report.
What experiences outside trials reveal
An international survey included 541 people who reported both a bipolar diagnosis and psilocybin use. New or worsening symptoms were reported by 174 respondents (32.2%), particularly manic symptoms, sleep difficulties and anxiety. Eighteen respondents (3.3%) reported use of emergency medical services. Many also described benefits and rated their experiences as more helpful than harmful. Positive personal meaning and clinically significant adverse effects can therefore coexist.6
Those percentages describe this particular survey, not the risk from a supervised clinical treatment. Participants volunteered, diagnoses and exposures were self-reported, and there was no untreated comparison group. Recall, other substances and differences in setting limit causal interpretation. Nevertheless, sleep disturbance and activation are sufficiently prominent to require active follow-up rather than reassurance based only on an initially pleasant experience.6
A published case report also describes mania after psilocybin use in a man previously diagnosed with bipolar II disorder. Prior vulnerability and alcohol and cannabis exposure complicate attribution. A case cannot estimate how often this happens or isolate a single cause, but it documents a serious outcome that small feasibility studies cannot exclude.7
Medication interactions and the danger of stopping treatment
Lithium deserves particular attention. An analysis of online accounts identified seizures in reports of lithium taken with classic psychedelics. These selectively reported experiences cannot produce a dependable incidence rate, and their findings concern several psychedelics rather than a controlled psilocybin regimen. They nevertheless raise a substantial safety concern. The absence of reported seizures with another medicine in the same dataset does not establish that combination as safe.8
Interaction concerns are not a reason to discontinue prescribed lithium or other bipolar treatment in order to take psilocybin. Medication changes require the treating clinician’s involvement. NICE warns that poor adherence or rapid lithium discontinuation increases relapse risk.2
Medication exclusions in a research protocol serve the needs of that particular study. They do not show that a patient would benefit from stopping treatment, nor that an excluded medicine is invariably incompatible. Safety depends on the specific drug, the clinical history, the consequences of withdrawal and the available evidence. A study that permits mood stabilisers and a study that requires medication changes are evaluating different treatment circumstances.
Monitoring must extend beyond the session
The clinically relevant observation period continues after the acute psychedelic effects have ended. Sleep, agitation, irritability, unusual confidence, impulsivity and suicidal thinking need explicit attention. Follow-up should make it possible to identify deterioration promptly, including when the patient interprets escalating activation as a therapeutic breakthrough. The recent bipolar II trial demonstrates why both antidepressant outcomes and psychiatric adverse events must be recorded.5
Qualitative interviews with people who had used psilocybin while living with bipolar disorder describe varied experiences of benefit and difficulty. Such accounts help identify concerns that standard rating scales may miss, but they cannot establish the effectiveness of a clinical intervention. They support asking about the effects on ordinary life, relationships and sleep, as well as the experience itself.9
Markedly reduced need for sleep with escalating activity, emerging psychosis, dangerous behaviour or suicidal intent warrants urgent clinical assessment. These changes should not be treated as a necessary stage of integration or left to resolve through further psychedelic exposure.
What remains unresolved
Randomised comparisons are needed to establish how much benefit is attributable to psilocybin within the treatment package. Larger samples and longer follow-up must address mood switching, relapse, repeat treatment, hospital admission and functioning. Diagnostic definitions, concurrent medication and baseline activation need to be reported clearly so that different populations are not treated as equivalent.
New feasibility protocols indicate continuing interest in bipolar II depression, but a published protocol supplies a plan, not a treatment result.10 The central unanswered question is whether depressive relief can be obtained reliably without trading it for instability later. Until that balance is demonstrated, established specialist bipolar care remains essential, including phase-appropriate medication, psychological treatment and relapse planning.2
References
- National Institute of Mental Health. Bipolar Disorder. Patient information; accessed 11 October 2026.
- National Institute for Health and Care Excellence. Bipolar disorder: assessment and management (CG185). Recommendations; accessed 11 October 2026.
- Aaronson ST, van der Vaart A, Miller T, 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;81:555–562. doi:10.1001/jamapsychiatry.2023.4685.
- Meshkat S, Kaczmarek E, Doyle Z, et al. Psilocybin-Assisted Psychotherapy for Treatment-Resistant Depression in Bipolar II Disorder. Psychedelic Med. 2025;3:53–58. doi:10.1089/psymed.2024.0032.
- Downey AE, Szigeti B, Bradley ER, et al. An Open-Label, Dose-Escalation Trial of Psilocybin-Assisted Therapy for Bipolar II Depression. J Clin Psychiatry. 2026;87:25m16227. doi:10.4088/JCP.25m16227.
- Morton E, Sakai K, Ashtari A, et al. Risks and benefits of psilocybin use in people with bipolar disorder: An international web-based survey on experiences of 'magic mushroom' consumption. J Psychopharmacol. 2023;37:49-60. doi:10.1177/02698811221131997.
- Halim HJ, Burk BG, Fargason RE, et al. Manic episode following psilocybin use in a man with bipolar II disorder: a case report. Front Psychiatry. 2023;14:1221131. doi:10.3389/fpsyt.2023.1221131.
- Nayak SM, Gukasyan N, Barrett FS, et al. Classic Psychedelic Coadministration with Lithium, but Not Lamotrigine, is Associated with Seizures: An Analysis of Online Psychedelic Experience Reports. Pharmacopsychiatry. 2021;54:240–245. doi:10.1055/a-1524-2794.
- DellaCrosse M, Pleet M, Morton E, et al. "A sense of the bigger picture:" A qualitative analysis of follow-up interviews with people with bipolar disorder who self-reported psilocybin use. PLoS One. 2022;17:e0279073. doi:10.1371/journal.pone.0279073.
- Meyer TD, Vale LN, Ibrahim M, et al. Acceptability and safety of two sequential doses of psilocybin in bipolar II depression: protocol for an open-label single-arm feasibility study. BMJ Open. 2026;16:e118471. doi:10.1136/bmjopen-2026-118471.