Psychedelic‐assisted therapy for treating anxiety, depression, and existential distress in people with life‐threatening diseases

Abstract

Background

Psychedelic‐assisted therapy refers to a group of therapeutic practices involving psychedelics taken under therapeutic supervision from physicians, psychologists, and others. It has been hypothesised that psychedelic‐assisted therapy may reduce symptoms of anxiety, depression, and existential distress in patients facing life‐threatening diseases (e.g. cancer). However, these substances are illegal in most countries and have been associated with potential risks.

Objectives

To assess the benefits and harms of psychedelic‐assisted therapy compared to placebo or active comparators (e.g. antidepressants) for treatment of anxiety, depression, and existential distress in people with life‐threatening diseases.

Search methods

We searched CENTRAL, MEDLINE, Embase, and two trial registers on 30 March 2024. In addition, we undertook reference checking, citation searching, and contact with study authors to identify additional studies. We used no language or date restrictions.

Selection criteria

We included randomised controlled trials (RCTs), with no restrictions regarding comorbidity, sex, or ethnicity. Interventions comprised a substance‐induced psychedelic experience preceded by preparatory therapeutic sessions and followed by integrative therapeutic sessions.

Data collection and analysis

We used the standard methodological procedures expected by Cochrane.

Main results

We included six studies in the review, which evaluated two different interventions: psychedelic‐assisted therapy with classical psychedelics (psilocybin ('magic mushrooms') and lysergic acid diethylamide (LSD)), and psychedelic‐assisted therapy with 3,4‐methylenedioxymethamphetamine (MDMA or 'Ecstasy'). The studies randomised 149 participants with life‐threatening diseases and analysed data for 140 of them. The age range of participants was 36 to 64 years. The studies lasted between 6 and 12 months, and were conducted in outpatient settings in the USA and in Switzerland. Drug companies were not involved in study funding, but funding was provided by organisations that promote psychedelic‐assisted therapy.

Primary outcomes (at 1 to 12 weeks)

Anxiety

Psychedelic‐assisted therapy using classical psychedelics (psilocybin, LSD) may result in a reduction in anxiety when compared to active placebo (or low‐dose psychedelic): State Trait Anxiety Inventory (STAI‐Trait, scale 20 to 80) mean difference (MD) −8.41, 95% CI −12.92 to −3.89; STAI‐State (scale 20 to 80) MD −9.04, 95% CI −13.87 to −4.21; 5 studies, 122 participants; low‐certainty evidence. The effect of psychedelic‐assisted therapy using MDMA on anxiety, compared to placebo, is very uncertain: STAI‐T MD −14.70, 95% CI −29.45 to 0.05; STAI‐S MD −16.10, 95% CI −33.03 to 0.83; 1 study, 18 participants; very low certainty evidence.

Depression

Psychedelic‐assisted therapy using classical psychedelics (psilocybin, LSD) may result in a reduction in depression when compared to active placebo (or low‐dose psychedelic): Beck Depression Inventory (BDI, scale 0 to 63) MD −4.92, 95% CI −8.97 to −0.87; 4 studies, 112 participants; standardised mean difference (SMD) −0.43, 95% CI −0.79 to −0.06; 5 studies, 122 participants; low‐certainty evidence. The effect of psychedelic‐assisted therapy using MDMA on depression, compared to placebo, is very uncertain: BDI‐II (scale: 0 to 63) MD −6.30, 95% CI −16.93 to 4.33; 1 study, 18 participants; very low certainty evidence.

Existential distress

Psychedelic‐assisted therapy using classical psychedelics (psilocybin, LSD) compared to active placebo (or low‐dose psychedelic) may result in a reduction in demoralisation, one of the most common measures of existential distress, but the evidence is very uncertain (Demoralisation Scale, 1 study, 28 participants): post treatment scores, placebo group 39.6 (SEM 3.4), psilocybin group 18.8 (3.6), P ≤ 0.01). Evidence from other measures of existential distress was mixed. Existential distress was not measured in people receiving psychedelic‐assisted therapy with MDMA.

Secondary outcomes (at 1 to 12 weeks)

Quality of life

When classical psychedelics were used, one study had inconclusive results and two reported improved quality of life, but the evidence is very uncertain. MDMA did not improve quality of life measures, but the evidence is also very uncertain.

Spirituality

Participants receiving psychedelic‐assisted therapy with classical psychedelics rated their experience as being spiritually significant (2 studies), but the evidence is very uncertain. Spirituality was not assessed in participants receiving MDMA.

Adverse events

No treatment‐related serious adverse events or adverse events grade 3/4 were reported. Common minor to moderate adverse events for classical psychedelics were elevated blood pressure, nausea, anxiety, emotional distress, and psychotic‐like symptoms (e.g. pseudo‐hallucination where the participant is aware they are hallucinating); for MDMA, common minor to moderate adverse events were anxiety, dry mouth, jaw clenching, and headaches. Symptoms subsided when drug effects wore off or up to one week later.

Certainty of the evidence

Although all six studies had intended to blind participants, personnel, and assessors, blinding could not be achieved as this is very difficult in studies investigating psychedelics. Using GRADE criteria, we judged the certainty of evidence to be low to very low, mainly due to high risk of bias and imprecision (small sample size).

Authors' conclusions

Implications for practice

Psychedelic‐assisted therapy with classical psychedelics (psilocybin, LSD) may be effective for treating anxiety, depression, and possibly existential distress, in people facing a life‐threatening disease. Psychedelic‐assisted therapy seemed to be well tolerated, with no treatment‐emergent serious adverse events reported in the studies included in this review. However, the certainty of evidence is low to very low, which means that we cannot be sure about these results, and they might be changed by future research. At the time of this review (2024), psychedelic drugs are illegal in many countries.

Implications for research

The risk of bias due to 'unblinding' (participants being aware of which intervention they are receiving) could be reduced by measuring expectation bias, checking blinding has been maintained before cross‐over, and using active placebos. More studies with larger sample sizes are needed to reduce imprecision. As the US Drug Enforcement Administration (DEA) currently classifies psychedelics as Schedule I substances (i.e. having no accepted medical use and a high potential for abuse), research involving these drugs is restricted, but is steadily increasing.

Plain language summary

Psychedelic‐assisted therapy for treating anxiety, depression and existential distress in people with life‐threatening diseases

Key messages

‐ In people with life‐threatening diseases (e.g. cancer), we found that psychedelic‐assisted therapy with “classical” psychedelics (i.e. psilocybin ('magic mushrooms'), LSD), may result in a reduction of anxiety and depression symptoms.

‐ We also found that existential distress (such as feeling that life has no meaning) and quality of life may be improved by classical psychedelics, but the evidence is mixed and very uncertain.

‐ Data regarding these outcomes is minimal for psychedelic‐assisted therapy with MDMA ('Ecstasy'), and we are very uncertain about the results.

‐ No severe negative side effects of psychedelic‐assisted therapy were reported in the studies we identified, but the evidence is very uncertain. Moderate side effects subsided after the drug effects wore off or within the following week.

What is the significance of anxiety, depression, and existential distress in people with life‐threatening diseases?

Anxiety, depression and existential distress are frequently experienced by people facing a life‐threatening disease and negatively impact their quality of life as well as the quality of life of their caregivers.

How are anxiety, depression, and existential distress treated?

Treatment of these symptoms is challenging, especially in end‐of‐life care, because well‐established antidepressant or anti‐anxiety medication may not work and psychotherapy may take a long time or be difficult to access.

What is psychedelic‐assisted therapy?

Psychedelics are illegal in most countries, but a few places have allowed restricted access, and research into the potential use of these drugs is increasing. Psychedelic‐assisted therapy involves the intake of a psychedelic drug (e.g. LSD, psilocybin ('magic mushrooms'), MDMA ('Ecstasy')) under close supervision by a therapist (physicians, psychologists, and others). Psychedelic‐assisted therapy includes three treatment phases: preparation sessions take place first, then the drug dosing (intake) session, and afterwards, integration sessions are conducted to reflect on the experience.

What did we want to find out?

We wanted to find out whether psychedelic‐assisted therapy was better than active‐placebo‐assisted therapy to improve anxiety, depression, and existential distress. We also wanted to find out whether psychedelic‐assisted therapy was linked with any unwanted effects or risk of harm.

What did we do? 
Our aim was to systematically collect and evaluate available evidence for psychedelic‐assisted therapy to treat anxiety, depression, and existential distress. We searched for high quality clinical studies. We compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as study methods.

What did we find?

We found six studies testing psychedelic‐assisted therapy using psilocybin (3 studies), LSD (2 studies), and MDMA (1 study). The studies involved 149 adults with anxiety, depression, or existential distress. The largest study tested psilocybin in 56 people and the two smallest studies tested psilocybin or LSD in 12 people. The studies took place in the USA and Switzerland. Most studies had a follow‐up of 6 to 12 months. Drug companies were not involved in study funding, but funding was provided by organisations that promote psychedelic‐assisted therapy.

Psychedelic‐assisted therapy with classical psychedelics (psilocybin, LSD) compared to an active placebo (e.g. a very low dose of the same drug) may reduce anxiety and depression. However, our confidence in the effect estimate is limited; we may find that the true effect is substantially different when more studies are conducted. Psychedelic‐assisted therapy with classical psychedelics may reduce existential distress, but the evidence is mixed and very uncertain. For psychedelic‐assisted therapy with MDMA, the data on anxiety or depression is inconclusive, and existential distress was not investigated.

Psychedelic‐assisted therapy with classical psychedelics (psilocybin, LSD) compared to an active placebo may improve quality of life and induce spiritually significant experiences, but the evidence is very uncertain. For MDMA‐assisted therapy, quality of life did not improve, but the evidence is very uncertain. Spirituality outcomes were not assessed for MDMA.

Psychedelic‐assisted therapy seems to be well tolerated, but evidence about side effects is very uncertain. The studies reported no serious side effects. The classical psychedelic studies reported mild to moderate side effects such as nausea, anxiety, dry mouth, psychotic‐like symptoms (e.g. pseudo‐hallucination, where people know that they are hallucinating), and high blood pressure, which subsided when the drug effects wore off or the following day. The MDMA study reported anxiety, dry mouth, jaw clenching, and headaches, which subsided when the drug effects wore off or within the following week.

What are the limitations of the evidence?

There were several limitations to the evidence. Importantly, participants were often aware of the treatment they were receiving, which may influence their results. In addition, the studies were small. It is important to note that future studies might change the conclusions of this review. As the US Drug Enforcement Administration (DEA) currently classifies psychedelics as Schedule I substances (i.e. having no accepted medical use and a high potential for abuse), research involving these drugs is restricted, but is steadily increasing.

How up to date is this evidence?

This review is based on a search for evidence up to March 2024.

Summary of findings

Summary of findings 1. Summary of findings table: classical psychedelics (psilocybin, LSD) compared to placebo in life‐threatening disease.

Classical psychedelics (psilocybin, LSD) compared to placebo in life‐threatening disease
Patient or population: people with life‐threatening disease Setting: outpatient Intervention: classical psychedelics (psilocybin, LSD) Comparison: placebo
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo	Risk with classical psychedelics (psilocybin, LSD)
Anxiety assessed with STAI‐T Scale 20 to 80, lower is better Follow‐up: 1 to 12 weeks	Mean anxiety ranged from ‐5.7 to +1.9 (change pre‐ to post‐treatment)	MD 8.41 lower (12.92 lower to 3.89 lower)	‐	122 (5 RCTs)	⊕⊕⊝⊝ Lowa,b	When measured with STAI‐S (scale 20 to 80, lower is better): MD −9.04, 95% CI −13.87 to −4.21; 5 studies, 122 participants; low‐certainty evidence.
Depression assessed with BDI, HADS‐D Follow‐up: 1 to 12 weeks	‐	SMD 0.43 SD lower (0.79 lower to 0.06 lower)	‐	122 (5 RCTs)	⊕⊕⊝⊝ Lowa,b	When measured with BDI (scale 0 to 63, lower is better): MD −4.92, 95% CI −8.97 to −0.87, 4 studies; 112 participants; low‐certainty evidence.
Existential distress Follow‐up: 1 to 12 weeks	Classical psychedelic‐assisted therapy reduced existential distress on some scales and not on others when compared to active placebo, but the evidence is very uncertain. Ross 2016 reported benefit from psilocybin for existential distress measured on the scale we prioritised, the Demoralisation Scale (high demoralisation > 36): post treatment scores: placebo group 39.6 (SEM 3.4), psilocybin group 18.8 (3.6), P ≤ 0.01) (1 study, 28 participants). Ross 2016 also reported finding a benefit from classical psychedelics using HAI, but no evidence of benefit on DAS or DTS. Griffiths 2016 reported a benefit from classical psychedelics when on LAP‐R but no evidence of a benefit on DTS.		‐	79 (2 RCTs)	⊕⊝⊝⊝ Very lowa,b,	Self‐report measures: DS‐I measures cancer‐related demoralisation syndrome (>36 high demoralisation, lower is better) HAI measures hopelessness in advanced cancer DAS measures level of death anxiety LAP‐R Death Acceptance subscale measures attitude towards death DTS measures positive attitudes and adaptations to the finitude of life
Quality of life Follow‐up: 1 to 12 weeks	Individual studies reported that classical psychedelics improved quality of life compared to active placebo, but the evidence is very uncertain. Gasser 2014 reported benefit on EORTC, but did not calculate P values. Griffiths 2016 used MQOL: placebo group (low‐dose psilocybin): 6.2 (SEM 0.32), psilocybin 7.1 (SEM 0.3), P < 0.05. Ross 2016 reported statistically significant increases on quality of life measures, but did not specify scores.		‐	90 (3 RCTs)	⊕⊝⊝⊝ Very lowa,b,c	Measures used: EORTC, scale 0 to 100; MQOL, scale 0 to 10. Higher scores are better.
Spirituality Follow‐up: 1 to 12 weeks	Individual studies reported that psychedelics induced spiritually significant experiences compared to active placebo, but the evidence is very uncertain. Griffiths 2016: Spiritual significance question: placebo (low‐dose psilocybin) 3.2 (SEM 0.24), psilocybin 4.5 (SEM 0.2), P ≤ 0.001 Purpose in Life Test: placebo (low‐dose psilocybin): 101.8 (SEM 3.78), psilocybin 106.2 (SEM 3.0), P value not significant between groups Ross 2016: Spiritual significance question: placebo (low‐dose psilocybin) 2.1 (SEM 0.33), psilocybin 4.5 (SEM 0.35), P ≤ 0.001 FACIT‐Sp: reported statistically significant increases, but did not specify scores.		‐	79 (2 RCTs)	⊕⊝⊝⊝ Very lowa,b,c	Measures used: spiritual significance question: how spiritually significant was the experience? (scale 1 to 6); Purpose in Life Test, self‐rated measure of life meaningfulness (scale 20 to 140). Higher scores are better.
Serious adverse events	No participant in the intervention or placebo groups had a treatment‐related serious adverse event.		‐	122 (5 RCTs)	⊕⊕⊝⊝ Lowa,b	‐
Adverse events grade 3 and 4	No participant in the intervention or placebo groups had a grade 3 or 4 treatment‐related adverse event.		‐	122 (5 RCTs)	⊕⊕⊝⊝ Lowa,b	Common minor to moderate adverse events were elevated blood pressure, nausea, anxiety, emotional distress, and psychotic‐like symptoms (e.g. pseudo‐hallucination where the participant is aware they are hallucinating). Symptoms subsided when drug effects wore off or up to a day later.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; SEM: standard error of the mean; SMD: standardised mean difference; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_443367815595315556.

^aWe downgraded the certainty of the evidence by one level due to serious risk of bias: unblinding.
^bWe downgraded the certainty of evidence by one level due to serious imprecision caused by there being fewer than 400 participants. 
^cWe downgraded the certainty of evidence by one level due to serious inconsistency caused by partially contradictory data.

Scales: BDI: Beck Depression Inventory; DAS: Death Anxiety Scale; DTS: Death Transcendence Scale; DS‐I: Demoralisation Scale; EORTC: European Cancer Quality of Life Questionnaire; FACIT‐Sp: Functional Assessment of Chronic Illness Therapy‐Spiritual Well‐Being; HADS‐D: Hospital Anxiety and Depression Scale ‐Depression; HAI: Hopelessness Assessment and Illness; LAP‐R: Life Attitude Profile Revised; MQOL: McGill Quality of Life Questionnaire; STAI‐S: State‐Trait Anxiety Inventory‐State; STAI‐T: State‐Trait Anxiety Inventory‐Trait

Summary of findings 2. Summary of main findings: 3,4‐methyl​enedioxy​methamphetamine (MDMA) compared to placebo in life‐threatening disease.

3,4‐methyl​enedioxy​methamphetamine (MDMA) compared to placebo in life‐threatening disease
Patient or population: people with life‐threatening disease Setting: outpatient Intervention: 3,4‐methyl​enedioxy​methamphetamine (MDMA) Comparison: placebo
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo	Risk with 3,4‐methyl​enedioxy​methamphetamine (MDMA)
Anxiety assessed with: STAI‐T Scale from 20 to 80, lower is better Follow‐up: 1 to 12 weeks	Mean anxiety was 48.6 (post‐treatment score)	MD 14.7 lower (29.45 lower to 0.05 higher)	‐	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b	Measured with STAI‐S (scale 20 to 80, lower is better): risk with placebo 45.8, risk with MDMA MD 16.1 lower (CI 33.03 lower to 0.83 higher)
Depression assessed with: BDI‐II Scale from 0 to 63, lower is better Follow‐up: 1 to 12 weeks	Mean depression was 15.4 (post‐treatment score)	MD 6.3 lower (16.93 lower to 4.33 higher)	‐	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,b	‐
Existential distress	‐	‐	‐	‐	‐	Existential distress was not assessed.
Quality of life Follow‐up: 1 to 12 weeks	One study reported that quality of life did not improve with MDMA compared to placebo, but the evidence is very uncertain.		‐	17 (1 RCT)	⊕⊝⊝⊝ Very lowa,b	‐
Spirituality	‐	‐	‐	‐	‐	Spirituality was not assessed.
Serious adverse events	No participant in the intervention or placebo group had a treatment‐related serious adverse event, but the evidence is very uncertain.		‐	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,c	‐
Adverse events grade 3 and 4	No participant in the intervention or placebo group had a treatment‐related adverse event of grade 3 or 4, but the evidence is very uncertain.		‐	18 (1 RCT)	⊕⊝⊝⊝ Very lowa,c	Common minor to moderate adverse events for MDMA were anxiety, dry mouth, jaw clenching, and headaches. Symptoms subsided when drug effects wore off or up to one week later.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; N: number of participants; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
See interactive version of this table: https://gdt.gradepro.org/presentations/#/isof/isof_question_revman_web_443368057416602010.

^aWe downgraded the certainty of the evidence by one level due to serious risk of bias: unblinding.
^bWe downgraded the certainty of evidence by two levels due to serious imprecision caused by fewer than 400 participants and the CI including no‐effect.
^cWe downgraded the certainty of evidence by two levels due to serious imprecision caused by fewer than 400 participants; study was not sufficiently powered to detect rare events.

Scales: BDI: Beck Depression Inventory; STAI‐S: State‐Trait Anxiety Inventory‐State; STAI‐T: State‐Trait Anxiety Inventory‐Trait

Background

According to the World Health Organization (WHO), palliative care is an approach that improves the quality of life of patients and their families facing problems associated with life‐threatening diseases. The discipline focusses on the prevention and relief of suffering by means of early identification, comprehensive assessment, and effective treatment of pain and other physical, psychosocial, and spiritual problems (WHO 2020). Depending on the underlying disease, physical symptoms can vary widely amongst people receiving palliative care. Psychological, social, and existential distress is driven not so much by the particular disease, but results from the changes and losses that stem from a diagnosis with no hope for a cure. Often, anxiety, depression, and existential distress coexist and may lead to a multitude of negative clinical outcomes, such as a reduction of quality of life, increased suicide rates, and decreased survival (Brown 2020; Pitman 2018; Zho 2017).

Description of the condition

Anxiety in people with life‐threatening disease

Anxiety is a common symptom in people nearing or facing death. The prevalence of anxiety disorders in people suffering from cancer in different studies is around 10% (Grassi 2023; Mitchell 2011), while anxiety as a symptom is even higher at 35% (Zeilinger 2022). The high prevalence of anxiety is not limited to people suffering from oncological diseases, as studies investigating other terminal diseases point out. For example, in people with end‐stage chronic obstructive pulmonary disease (COPD), the prevalence of anxiety ranges from 10% to 33% (Hill 2008). Other terminal diseases are also associated with a high prevalence of anxiety (Murtagh 2007; Solano 2006).

Anxious cancer patients tend to overgeneralise and catastrophise, and they view themselves as helpless in a hopeless situation (Roth 2007). Anxiety is considered one of the most impactful symptoms in terms of directly negatively influencing quality of life (Smith 2003).

Depression in people with life‐threatening disease

Depression is a state of low mood and loss of interest or pleasure in nearly all activities (Kennedy 2008; Zwart 2019). Depressed mood is a common symptom of mood disorders such as major depressive disorder, bipolar spectrum disorder, and dysthymia (DSM‐5 2013). Symptoms of depression include anhedonia (the inability to experience pleasure), insomnia, sadness, difficulty with thinking and concentration, and significant changes in appetite (Clarke 2005). People experiencing depression may report feelings of dejection, hopelessness, isolation, and suicidal thoughts (NHS 2023).

The diagnosis of depression in adult palliative care is challenging. As psychological distress and situational sadness are inevitable parts of the disease process in terminal illness, palliative care practitioners may often underestimate the prevalence of treatable depression (Rayner 2009). Although low mood, fatigue, weight loss, and sleep disturbance are diagnostic symptoms of a depressive disorder, they may also be features of a life‐limiting illness. As such, grief as a normal response to loss is difficult to distinguish from a depressive disorder. Wide variation in existing prevalence estimates (3% to 45%) highlights the difficulty in detecting depression in palliative care patients (Reeve 2008). As with anxiety, there is a negative correlation between quality of life and the existence and intensity of depression. Moreover, depression is a predictor of desire for a hastened death in the terminally ill (Chochinov 1999).

Existential distress in people with life‐threatening disease

The construct of existential distress is inconsistently defined in healthcare literature (Boston 2011). Associated terms include existential, spiritual, holistic, or psycho‐existential suffering; existential, spiritual, or total pain; psychospiritual distress; and demoralisation (Best 2015a; Saunders 1996). A concept analysis of existential distress in cancer patients identified five core attributes of existential distress: lack of meaning, loss of autonomy, loss of dignity, hopelessness, and death anxiety (Chen 2022).

In the context of a life‐threatening disease, real, perceived, or threatened loss can trigger existential distress. Loss can occur on any level: physical, psychological, social, and spiritual (Wilson 2007). Existential distress varies in severity and fluctuates over time (Vehling 2018).

One clinically relevant existential construct is 'demoralisation': a persistent perceived inability to cope, a sense of helplessness and hopelessness, impaired self‐esteem, and lack of meaning and purpose in life due to a lack of a worthwhile future (Bovero 2019; Vehling 2019). People experience an "entrapped feeling that nothing can be done" (Vehling 2014). A systematic review of people suffering from all stages of cancer found an overall prevalence of significant demoralisation of 13% to 18% (Robinson 2015). Demoralisation prevalence could increase to more than 50% as death approaches and symptoms burden cumulate (Bovero 2019; Vehling 2019).

Description of the intervention

During the first wave of psychedelic research in the 1960s and 1970s, psychedelic‐assisted therapy showed promising results for reducing depression and fear of death in participants with cancer (Grof 1973; Kast 1966), but the studies also had considerable methodological shortcomings, including their lack of control groups and lack of blinding (Ross 2022). 'Modern' RCTs in participants with life‐threatening diseases have been published since 2011 and are reviewed here.

Psychedelics are a group of psychoactive substances. They can be divided into four different classes, as defined by their pharmacological mechanism of action and chemical structure (Garcia‐Romeu 2016):

• serotonin 2A receptor (5‐HT_2AR) agonists such as lysergic acid diethylamide (LSD), psilocybin (also known as 'magic mushrooms'), and N, N‐dimethyltryptamine (DMT), often referred to as classic hallucinogens or psychedelics (Garcia‐Romeu 2016);

• mixed serotonin, norepinephrine, and dopamine reuptake inhibitors and releasers such as 3,4‐methylenedioxy‐methamphetamine (MDMA, also known as 'Ecstasy'), referred to as empathogens or entactogens (Nichols 1986);

• N‐methyl‐D‐aspartate (NMDA) antagonists such as ketamine and dextromethorphan (DXM), also known as dissociative anaesthetics (Morris 2014); and

• atypical hallucinogens such as the kappa‐opioid receptor agonist salvinorin A, the indole alkaloid ibogaine, which affects multiple neurotransmitter systems, anticholinergics such as atropine and datura, also known as deliriants, and cannabis (Garcia‐Romeu 2016).

Psychedelic‐assisted therapy

Psychedelic‐assisted therapy refers to a group of therapeutic practices involving psychedelics. As opposed to the use of conventional psychiatric medication, which is usually taken on a daily basis, psychedelics are administered only once or a few times (Reiff 2020).

Psychedelic‐assisted therapy consists of three distinct phases: preparation, psychedelic experience, and integration (Reiff 2020). During the preparatory phase, the person and the therapist build a therapeutic alliance and rapport that may improve therapeutic outcomes (Murphy 2022). The person explores their life history, elaborates on personal motives and goals for treatment, and is informed about the drug effects during the psychedelic experience (Cavarra 2022). During the substance session, the person takes a psychedelic drug in the presence of one or two therapists, whose role is to non‐directionally support the person in keeping the attention inward and to provide support during challenging moments (Cavarra 2022). This session takes place in a pleasant and calming environment. The person may recline on a couch or bed to feel comfortable. Therapists often offer people eyeshades and music to use throughout the experience to facilitate introspection (Johnson 2014; Mithoefer 2015). After the acute psychedelic experience, the therapist and the person debrief the nature of the experience in one or several integration sessions. During integration sessions, emphasis is placed on the personal significance of the psychedelic experiences and how the individual can integrate the insights gained into their daily life, including new meaning, perspectives, attitudes, and behaviours (Schenberg 2018).

Legal status of psychedelics

Psychedelics are illegal in many countries of the world. The US Drug Enforcement Administration (DEA) currently classifies psychedelics as Schedule I substances. However, several US states (Siegel 2023), as well as Canada and Jamaica, have started towards decriminalisation and legalisation (Cormier 2020; De La Haye 2023). Some countries, for example, Switzerland, allow limited medical use of psychedelics in patients suffering from various treatment‐resistant disorders (referred to as 'compassionate care') (Gasser 2022; Schmid 2021). Other countries, for example, Portugal and the Netherlands, do not criminalise the personal use of some psychedelics (Marks 2021). Australia was the first country worldwide to approve the use of psychedelics to treat some mental health conditions: specialised psychiatrists are allowed to prescribe MDMA to treat post‐traumatic stress disorder (PTSD) and psilocybin for treatment‐resistant depression (Haridy 2023).

How the intervention might work

Neurobiological mechanisms

Due to psychedelics' complex neurobiological and psychological mechanisms, we will only introduce a basic overview of these. It is widely accepted that classic psychedelics like LSD and psilocybin exert their main action by activating the 5‐HT_2A receptor (Carhart‐Harris 2017; Preller 2019); a comprehensive review of pharmacokinetics and binding profiles can be found elsewhere (Holze 2024). This G‐protein‐coupled receptor is expressed throughout the cortex, especially in the higher‐level associative cortex, whereby its activation leads to excitatory effects of the host neurone (Carhart‐Harris 2017). The psychedelic experience is characterised by altered somatosensory, visual, auditory, and proprioceptive sensations (Holze 2022; Nichols 2016; Turton 2014). For example, 5‐HT_2A agonists change sensory perceptions, enhance mental imagery, intensify affectivity, and modify thinking processes (Nichols 2016; Passie 2008). It is hypothesised that the activation of 5‐HT_2A receptors mediates active coping, neuroplasticity, and open‐mindedness (Carhart‐Harris 2017; Vollenweider 2020), improving mood and thereby reducing depression and anxiety (Carhart‐Harris 2017).

Through transient production of a neuroplastic (entropic, unpredictable, desynchronised) brain state (Carhart‐Harris 2014; Siegel 2024), the brain is enabled to adapt, change, and reset. This hypothesis is supported by neuroimaging studies, which suggest that psychedelics increase functional connectivity in and between specific brain networks while decoupling others, affecting the integrity of the default mode network (DMN), which is hypothesised to be a neurological functional correlate for the sense of self (Carhart‐Harris 2010; Carhart‐Harris 2016). The DMN is persistently over‐activated in various pathologies such as depression, anxiety, and post‐traumatic stress disorders, as well as in chronic pain disorders (Otti 2012). The inhibition of the DMN by psychedelics allows the 'stuck', pathologically self‐referential brain to establish new associations with other brain areas. As a result, the brain is able to diminish self‐reference and restore the lost relation to the world (environment) (Carhart‐Harris 2012). In summary, psychedelic‐assisted therapy may break a fixated neurobiological matrix characterised by impaired responsiveness and reduced emotional and cognitive flexibility, as is often seen in depressive states (Vollenweider 2020).

Psychological mechanisms

Interestingly, long‐lasting psychological changes are commonly reported after just a single intake of psychedelics (Schmid 2018). Some studies suggest that the positive psychological effects of psychedelics are mediated mainly by positive changes in personality, such as an increase in openness (MacLean 2011). Another key aspect of the psychedelic experience is its profound spiritual and mystical nature (Griffiths 2006). Several studies suggest that positive outcomes correlate with the intensity of mystical experience (Breitbart 2015; Griffiths 2016; Johnson 2014; Liechti 2017). Thus, mystical experiences may facilitate personal transformative processes that can lead to dramatic changes in a person's sense of self and behaviour.

People suffering from anxiety, depression, or existential distress who are facing or nearing death are often in a negative state of anticipation. Psychedelics may ameliorate this anticipation, enabling people to become more aware of the pleasures of the present moment and facilitating a new will to live against a background of preoccupying fear (Sheehan 1972). By doing so, people are able to engage in the process of creating meaning out of their experience (Bauereiß 2018). In summary, the effect of psychedelic‐assisted therapy is not limited to the physiological drug effect, but rather relies on insights and memories conferred by the unique phenomenology of the psychedelic experience, which allows for the possibility of discovering new perspectives regarding the nature of self, death, and consciousness (Grob 2013).

Why it is important to do this review

Pharmacological treatments for depressed people with cancer show a potential beneficial effect of antidepressants, but the evidence is very uncertain (Vita 2023). Similarly, there is insufficient high‐quality evidence supporting the effectiveness of psychotherapy for patients suffering from advanced cancer and clinically diagnosed depression (Okuyama 2017). For symptoms of anxiety, there is a lack of evidence about the effectiveness of existing drug therapies in people receiving palliative care (Salt 2017).

Several interventions address existential distress in cancer and enhance the spiritual aspects of dying (LeMay 2008): meaning‐centered, hope‐centered therapy, dignity therapy, hope‐centred interventions, stress‐reduction interventions, supportive‐expressive therapy, psychoeducational interventions, and religious or spiritual interventions (Bauereiß 2018; Best 2015b; Candy 2012; Chochinov 2005; Chochinov 2008). However, there remains an ongoing need to improve how we address existential distress.

There is a need for new, more efficient, and faster‐acting interventions for anxiety, depression, and existential distress in a population that is not afforded the luxury of time. Promising early‐phase clinical research (in the 1960s to early 1970s) suggested therapeutic potential for serotonergic psychedelics (e.g. psilocybin, LSD, dipropyltryptamine (DPT)) in treating cancer‐related psychiatric distress (Ross 2018); however, research utilising psychedelics was halted for several reasons. These included political reasons, as Nixon's 'War on Drugs' which led to increased barriers to conducting psychedelic research, and other reasons, such as the failure of controlled clinical trials to live up to the claims of psychedelic advocates and the pharmaceutical industry's lack of interest in funding clinical trials (Hall 2021).

In summary, conducting this intervention systematic review with meta‐analysis is important because:

• anxiety, depression, and existential distress are common in people facing a life‐threatening diseases and palliative care. Modern medicine (e.g. immunotherapies in cancer treatment) allows people to live for longer but with 'chronic' non‐curable diseases and the demands of coping with them;

• psychedelics have been investigated for the treatment of anxiety, depression, and existential distress in small randomised controlled trials (RCTs). The results are promising in terms of efficacy and safety, but the numbers of participants are low. Legalisation of the medical use of psychedelics for other indications (PTSD, treatment‐resistant depression) is likely to occur within the next few years. 'Compassionate' use of psychedelics for anxiety, depression, and existential distress in palliative care is already legal in several countries (e.g. the expanded access programme of the FDA in the US or in Switzerland). Thus, an up‐to‐date systematic review of the literature, with meta‐analysis, can inform evidence‐based decision‐making.

Objectives

To assess the benefits and harms of psychedelic‐assisted therapy compared to placebo or active comparators (e.g., antidepressants) for the treatment of anxiety, depression, and existential distress in people with life‐threatening diseases.

Methods

Criteria for considering studies for this review

Types of studies

We published our planned methods as a protocol (Schipper 2022).

We included randomised controlled trials (RCTs) (including cross‐over trials and cluster‐RCTs) that investigated the effects of psychedelic‐assisted therapy in adults with life‐threatening disease. We limited our inclusion to RCTs because, if performed appropriately, this design provides the most reliable evidence on the effects of experimental interventions in a highly controlled therapeutic setting.

We included studies reported in peer‐reviewed journal articles, clinical trial registry entries, conference proceedings, summaries of otherwise unpublished clinical trials, and abstracts. If there were insufficient data to assess eligibility, or extract or assess data, we attempted to procure more information by contacting the study authors.

We excluded other study designs, such as non‐randomised studies, case reports, and clinical observations.

Types of participants

We included studies with participants:

• suffering from anxiety, depression, and existential distress (diagnosed with one or more of the conditions or meeting predefined score thresholds on questionnaires (e.g. the State Trait Anxiety Inventory (STAI));

• with a life‐threatening disease such as advanced cancer, HIV/AIDS, heart failure, renal failure, liver failure, etc;

• aged 18 years or older;

• in inpatient and outpatient settings.

Some studies included participants with curable and incurable cancer or participants with non‐malignant and advanced‐stage diseases. As the life‐threat is the main reason for the psychological condition (anxiety, depression, existential distress), as opposed to the stage of the cancer or other disease, we included participants with curable or incurable disease (see Differences between protocol and review).

Types of interventions

We included all interventions that contained the following two elements:

• intake of psychedelics at any dosage;

• preparatory and integration therapy.

The interventions were compared to therapy assisted by placebo or active placebo (e.g. low‐dose psychedelic), or active comparators (e.g. antidepressants).

Types of outcome measures

Primary outcomes

• Anxiety

  • Generalised Anxiety Disorder 7‐Item scale (GAD‐7) (Spitzer 2006), Hospital Anxiety and Depression Scale (HADS) (Zigmond 1983), State‐Trait Anxiety Inventory (STAI) (Spiegelberger 1983), Brief Symptom Inventory (BSI) (Derogatis 1992), Numeric Rating Scale (NRS) (Downie 1978)

• Depression

  • Patient Health Questionnaire–9 (PHQ‐9) (Kroenke 2001), HADS (Zigmond 1983), Beck Depression Inventory (BDI) (Beck 1987), Hamilton Depression Rating Scale (HDRS) (Hamilton 1960), BSI (Derogatis 1992), NRS (Downie 1978)

• Existential distress

  • Demoralisation scale I (DS‐I) (Kissane 2004), Hopelessness Assessment and Illness (HAI) (Rosenfeld 2011), Death Anxiety Scale (DAS) (Templer 1970), Life Attitude Profile‐Revised (LAP‐R) Death Acceptance (Reker 1992)

We analysed two time points:

• 1 to 12 weeks after the end of the intervention;

• more than 3 months after the end of the intervention.

If multiple time points were presented within the specified range (< 12 weeks, > 3 months), we used the latest time point because this seemed to be the most relevant, considering our interest in the long‐term efficacy of treatments.

Secondary outcomes

• Quality of life

  • WHO Quality of Life Scale (Whoqol Group 1998), McGill Quality of Life Questionnaire (MQOL) (Cohen 1995), European Cancer Quality of Life Questionnaire (EORTC)‐QLQ (Aaronson 1993)

• Spirituality (spiritual experience/well‐being)

  • Functional Assessment of Chronic Illness Therapy‐Spiritual Well‐Being (FACIT‐SP) (Peterman 2002)

• Serious adverse events related to the treatment (number of participants experiencing an event)

• Adverse events grade 3 and 4 (as described by the Common Terminology Criteria of Adverse Events (CTCAE Version 5)) related to the treatment (number of participants experiencing an event)

Search methods for identification of studies

Electronic searches

We searched the following databases on 30 March 2024 without language or date restrictions.

• The Cochrane Central Register of Controlled Trials (CENTRAL, 2024, Issue 3) (in the Cochrane Library)

• MEDLINE (via Ovid) 1946 to 30 March 2024

• Embase (via Ovid) 1980 to 2024 week 13

We tailored searches to individual databases. The search strategies can be found in the appendices (Appendix 1; Appendix 2; Appendix 3). We used the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE and Embase (Lefebvre 2023).

The Cochrane Pain, Palliative and Supportive Care (PaPaS) Group's Information Specialist (Joanne Abbott) developed the search strategy and performed the searches. We had planned to search for studies in any language to limit language biases. Had we identified records in a language that our review team could not translate, we would have asked for assistance via Cochrane Task Exchange, but this was not necessary as only articles in English were found.

Adverse effects

We did not perform a separate search for adverse effects of the target interventions. We considered the adverse effects described in the included studies only.

Searching other resources

We searched the US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organisation International Clinical Trials Registry Platform (WHO ICTRP) (apps.who.int/trialsearch) for ongoing trials using the search strategies in Appendix 4 and Appendix 5. In addition, we checked reference lists of similar reviews and retrieved articles for additional studies, and performed citation searches on key articles. We contacted experts in the field for unpublished and ongoing trials. We contacted study authors for additional information where necessary.

Data collection and analysis

Selection of studies

Some sections of the following methods were derived from a Cochrane Review published by authors of this review (Andreas 2021).

Two review authors (Kabir Nigam (KN) and Christopher Boehlke (CB)) independently determined the eligibility of each study identified by the search. We coded the abstracts as either 'include' or 'exclude' using the software Rayyan (Ouzzani 2016). The review authors eliminated studies that did not satisfy inclusion criteria and obtained complete copies of the remaining studies. Two review authors (KN, CB) read these studies independently to select relevant studies, and in the event of disagreement, a third author adjudicated (Vanessa Piechotta (VP)). We did not anonymise the studies in any way before the assessment. We included a PRISMA flowchart in this review (Page 2021). We included studies in the review irrespective of whether measured outcome data were reported in a 'usable' way.

Data extraction and management

We conducted data extraction and assessment according to the guidelines proposed by Cochrane (Li 2023). Two review authors (KN, CB) independently extracted data using a standard piloted form and checked for agreement before entry into Review Manager (RevMan Web 2024). In the event of disagreement, a third author adjudicated (VP). We collated multiple reports of the same study, so that each study is the unit of interest in the review rather than each report. We collected the characteristics of the included studies in sufficient detail to populate a table of Characteristics of included studies in this review. We extracted data using a customised data extraction form developed in Microsoft Excel (Microsoft 2018), and extracted the following information.

• Methods: study design, total study duration, number of study centres and location, study setting, date of the study.

• Participants: number randomised, number lost to follow‐up or withdrawn, number analysed, mean age, age range, sex/gender, severity of the condition, diagnostic criteria, inclusion and exclusion criteria.

• Interventions: intervention, comparison, concomitant medications, and excluded medications.

• Outcomes: outcomes as specified in Types of outcome measures, time points reported.

• Notes: funding for the trial and notable conflicts of interest of trial authors. Information needed to assess bias (e.g. any deviations from intended interventions, data imputed for key outcomes, etc).

• Information needed to assess GRADE (e.g. baseline risk in the control group for key outcomes).

Following the advice in Chapter 5 of the Cochrane Handbook for Systematic Reviews of Interventions, we used free online software to extract data for meta‐analyses from published graphs when authors did not provide data in another format (Plotdigitizer 2023).

Assessment of risk of bias in included studies

Two authors (KN, CB) independently assessed the risk of bias for each study (RoB1), using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We discussed any disagreement and a third author adjudicated (VP), if necessary. We completed a risk of bias table for each included study using the risk of bias tool in Review Manager (RevMan Web 2024).

We assessed the following biases for each included study.

• Random sequence generation (checking for possible selection bias). We assessed the method used to generate the allocation sequence as:

  • low risk of bias (any truly random process; e.g. random number table, computer random number generator);

  • unclear risk of bias (insufficient detail about the method of randomisation to be able to judge);

  • we excluded studies that used a non‐random process (e.g. odd or even date of birth, hospital or clinic record number).

• Allocation concealment (checking for possible selection bias). The method used to conceal allocation to interventions prior to assignment determines whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment. We assessed the methods as:

  • low risk of bias (e.g. telephone or central randomisation, consecutively numbered sealed opaque envelopes);

  • unclear risk of bias (insufficient detail about the method of randomisation to be able to judge);

  • we excluded studies that did not conceal allocation (e.g. open list).

• Blinding of participants and personnel (checking for possible performance bias). We assessed the methods used to blind study participants and personnel from knowledge of which intervention a participant received. We assessed methods as:

  • low risk of bias (study states that it was blinded and describes the method used to achieve blinding, such as identical tablets matched in appearance or smell, or a double‐dummy technique (i.e. both groups receiving drug and placebo at the same time));

  • unclear risk of bias (study states that it was blinded but does not provide an adequate description of how it was achieved);

  • we considered studies that were not double‐blinded to have a high risk of bias.

• Blinding of outcome assessment (checking for possible detection bias). We assessed the methods used to blind study participants and outcome assessors from knowledge of which intervention a participant received. We assessed the methods as:

  • low risk of bias (study has a clear statement that outcome assessors were unaware of treatment allocation and describes how this was achieved);

  • unclear risk of bias (study states that outcome assessors were blinded to treatment allocation but lacks a clear statement on how this was achieved);

  • we considered studies at high risk of bias where outcome assessment was not blinded.

• Incomplete outcome data (checking for possible attrition bias due to the amount, nature, and handling of incomplete outcome data). We assessed the methods used to deal with incomplete data as:

  • low risk (no missing outcome data; reasons for missing outcome data unlikely to be related to true outcome (for survival data, censoring unlikely to be introducing bias); missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups; missing data have been imputed using 'baseline observation carried forward’ analysis);

  • unclear risk of bias (insufficient reporting of attrition and exclusions to permit a judgement (e.g. number randomised not stated, no reasons for missing data provided, or the study did not address this outcome));

  • high risk of bias (reason for missing outcome data is likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups; 'as‐treated' analysis done with substantial departure of the intervention received from that assigned at randomisation; potentially inappropriate application of simple imputation).

• Selective reporting (checking for reporting bias). We assessed reporting biases due to selective outcome reporting. We judged studies as:

  • low risk of bias (the study protocol is available and all of the study's prespecified outcomes that are of interest in the review have been reported in the prespecified way);

  • unclear risk of bias (insufficient information available to permit a judgement);

  • high risk of bias (not all the study's prespecified primary outcomes have been reported; one or more primary outcomes have been reported using measurements, analysis methods, or subsets of the data (e.g. subscales) that were not prespecified; one or more reported primary outcomes were not prespecified (unless clear justification for their reporting is provided, such as an unexpected adverse effect); one or more outcomes of interest in the review have been reported incompletely so that they cannot be entered in a meta‐analysis; the study report failed to include results for a key outcome that would be expected to have been reported for such a study).

• Other bias

  • low risk of bias: the study appears to be free of other components that could put it at risk of bias.

  • unclear risk of bias: insufficient information available to permit a judgement about whether the study is free of other components that could put it at risk of bias.

  • high risk of bias: there are other factors in the study that could put it at risk of bias, e.g. for‐profit involvement (e.g. pharmaceutical companies), authors have conducted trials on the same topic, etc.

Overall risk of bias

We judged the overall risk of bias to be low if we assessed the study as being at low risk of bias in all the specified key domains; unclear risk of bias if one or more of the specified key domains were judged as unclear risk of bias and none of the key domains were judged as high risk; and high risk of bias if we judged one or more of the specified key domains to be at high risk of bias. Due to the highly subjective outcomes, key domains were: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, and selective reporting.

Measures of treatment effect

For continuous outcomes using the same scale, we performed analyses using the mean difference (MD) with 95% confidence intervals (CIs). During the review process, we decided to use change scores as pre‐ and post‐values were available. However, no information regarding correlation between pre‐ and post‐values was reported. Therefore, the standard error for the MD was calculated by assuming no correlation between pre‐ and post‐values, which leads to wider confidence intervals.

For continuous outcomes measured on different scales, we performed analyses using the standardised mean difference (SMD). In our interpretation of SMDs, we re‐expressed the SMD in the original units of a particular scale with the most clinical relevance and impact (BDI). We entered data presented as a scale with a consistent direction of effect.

For dichotomous outcomes, we described the percentage of participants experiencing a certain event.

If data were not reported in a format that we could enter directly into a meta‐analysis, we planned to convert it to the required format using the information in Chapter 6 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2023), obtaining standard deviations from standard errors for group means, and imputing standard deviations for changes from baseline.

Unit of analysis issues

The unit of analysis was the individual participant.

Events that may re‐occur

If events were observed multiple times, we considered the number of participants experiencing any event and not the number of experienced events for analysis.

Multiple treatment attempts

We considered the number of participants and not the number of assigned treatments for analysis. Since psychological changes through psychedelic‐assisted therapy are long‐lasting and sometimes permanent, we considered only the first period of cross‐over trials for our analyses.

Studies with multiple treatment groups

For studies with multiple treatment groups, we had planned to combine arms when they could be regarded as subtypes of the same intervention. If they could not, we planned to compare each arm with the common comparator separately. For pairwise meta‐analyses, we would split the 'shared' group into two or more groups with smaller sample sizes, and include two or more independent comparisons. For this purpose, for dichotomous outcomes, both the number of events and the total number of participants would be divided up by the common comparator. For continuous outcomes, the total number of participants would be divided up by unchanged means and standard deviations (SDs) for the common comparator. However, no multiple treatment groups were identified.

Cluster‐randomised trials

Had we found cluster‐randomised trials, we would have analysed them according to the methods described in Chapter 23 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2023b).

Dealing with missing data

Several potential sources of missing data are suggested in Chapter 6 of the Cochrane Handbook for Systematic Reviews of Interventions, which we took into account: at study level, outcome level, and at summary data level (Higgins 2023). In the first instance, it was of the utmost importance to differentiate between data 'missing at random' and 'not missing at random'. We identified additional and unpublished data by contacting investigators or study sponsors to verify key study characteristics or obtain missing outcome data. If authors did not provide missing data, we had planned to use the risk of bias assessment to decide if data was missing at random or not at random and perform a Sensitivity analysis of studies with a high risk of bias (for selective reporting). As we assessed all included studies as high risk of bias, this was not necessary. We assessed the risk of bias due to missing results by thoroughly assessing selective non‐reporting or under‐reporting results in the studies identified.

When possible, we calculated missing standard deviations (SDs) or other necessary data using additional data from the trial, such as CIs, based on methods outlined in Chapter 6 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2023). When the reported results did not allow calculation of the standard error (SE) of the mean difference in a paired analysis, we tried to approximate the SE by estimating the correlation within participants. When there were insufficient data to calculate the correlation coefficient, we assumed a correlation of zero, which resulted in a conservative scenario, i.e., SE was slightly overestimated.

Statistical heterogeneity

We assessed statistical heterogeneity of treatment effects between studies as recommended in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions, using a Chi² test with a significance level at P < 0.1 and through visual examination of forest plots to consider the direction and magnitude of effects and the degree of overlap between CIs. We used the I² statistics to quantify possible heterogeneity, with an I² over 30% signifying moderate heterogeneity and an I² over 75% signifying considerable heterogeneity (Deeks 2023). We had planned to explore potential causes of heterogeneity through subgroup analyses (Subgroup analysis and investigation of heterogeneity), but heterogeneity in our meta‐analyses was low.

Clinical heterogeneity

We did not pool the results in cases of significant clinical diversity.

Assessment of reporting biases

We did not plan to assess reporting bias (e.g. using funnel plots (Egger 1997)), because we only expected a small number of RCTs to be included in the review. We will use a funnel plot to assess reporting bias in the future if there are 10 or more studies (Sterne 2011).

Meta‐analysis of numerical data

If the clinical and methodological characteristics of individual studies were sufficiently homogeneous, we pooled data in meta‐analysis. We performed analyses according to the recommendations in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2023).

We used Review Manager Web for analyses (RevMan Web 2024). One review author (CB) entered the data into the software, and a second review (KN) author checked the data for accuracy.

We used the random‐effects model for all analyses, as we anticipated that true effects would be related but not the same for the included studies. We based the estimation of the between‐study variance on the DerSimonian‐Laird method (DerSimonian 1986). We used the inverse‐variance method in all random‐effects meta‐analyses. We explored heterogeneity with subgroup analyses (Subgroup analysis and investigation of heterogeneity).

We made two treatment comparisons:

• classical‐psychedelic‐assisted therapy (lysergic acid diethylamide (LSD), psilocybin) versus placebo‐assisted (low dose) therapy;

• entactogen‐assisted therapy (3,4‐methylenedioxy‐methamphetamine (MDMA)) versus placebo‐assisted therapy.

We had planned to assess psychedelic‐assisted therapy versus active comparator, but no studies were identified that evaluated this comparison.

Synthesis using other methods

If meta‐analysis was not possible, we followed Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions (McKenzie 2023). For example, we used structured tabulation of results across studies for secondary outcomes and adverse events.

Subgroup analysis and investigation of heterogeneity

We conducted subgroup analyses only for the primary outcome measures (anxiety, depression, existential distress). We compared LSD and psilocybin as they have different serum half‐lives (t_1/2) (Holze 2022). We compared subgroups using the difference in effects between subgroups approach presented in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2023).

We had also planned to conduct subgroup analyses by number of sessions (one psychedelic therapy session versus several (two to three)) and medical condition, for example, participants suffering from cancer versus participants suffering from other diseases. However, there were too few studies to undertake these analyses.

Sensitivity analysis

To test the robustness of results, we conducted sensitivity analyses of the primary outcomes based on the following characteristics.

• Influence of the meta‐analysis model used, i.e. random‐effects versus fixed‐effect model. We reported effects based on a fixed‐effect model only if they differed from estimates of using the random‐effects model. We planned to use the Mantel‐Haenszel method for binary outcomes under the fixed‐effect model.

• Influence of overall risk of bias (as defined in Assessment of risk of bias in included studies): we planned to restrict primary analyses to studies judged to be at overall low risk of bias, but we judged all studies to be at high risk of bias overall. We included studies at high risk of bias in our primary analyses, because all studies were at high risk of bias due to the unblinding of participants and personnel that is inherent to studies with psychedelics. We discussed the possible influence of unblinding thoroughly in the discussion section and decided to interpret the results very cautiously (compare Summary of findings and assessment of the certainty of the evidence).

Summary of findings and assessment of the certainty of the evidence

We used GRADEpro GDT software to create the summary of findings tables (GRADEpro GDT 2024).

In accordance with Chapter 14 of the updated Cochrane Handbook for Systematic Reviews of Interventions, we included the "most critical and/or important health outcomes, both desirable and undesirable, limited to seven or fewer outcomes" in the summary of findings tables (Schünemann 2023). We prioritised outcomes most relevant for individuals with a terminal illness, as follows:

• anxiety;

• depression;

• existential distress;

• quality of life;

• spirituality;

• serious adverse events;

• adverse events Grade 3 and 4.

We reported the scales that allowed us to include most studies/participants.

Assessment of the certainty of the evidence

We used the GRADE approach to assess the certainty of the evidence for the outcomes listed above. Two review authors (Sivan Schipper (SS), CB) independently rated the certainty of the body of evidence for the outcomes, with disagreements resolved by discussion or involving a third review author (VP). We justified, documented, and incorporated judgements into the reporting of results for each outcome.

The GRADE approach uses five domains (risk of bias, consistency of effect, imprecision, indirectness, and publication bias) on which to assess the certainty of the body of evidence for each outcome. According to GRADE guidance 18 (Schünemann 2019), the initial rating for randomised trials is high certainty.

Factors that may decrease the certainty level of a body of evidence are:

• serious or very serious study limitations (risk of bias as defined in Assessment of reporting biases);

• important or serious inconsistency of results;

• some or major indirectness of evidence;

• serious or very serious imprecision;

• probability of publication bias.

The GRADE system definitions for the level of certainty we can have in a body of evidence are:

• high: we are very confident that the true effect lies close to that of the estimate of the effect;

• moderate: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of effect, but there is a possibility that it is substantially different;

• low: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect; or

• very low: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

We followed the current GRADE guidance for assessing the certainty of the evidence in its entirety, as recommended in Chapter 14 of the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2023). We used our overall risk of bias judgements to inform decisions on downgrading for study limitations. We phrased the findings and certainty of the evidence as suggested in the informative statement guidance of GRADE guidelines 26 (Santesso 2020).

We included two summary of findings tables to present the main findings for 'classical' psychedelics (LSD, psilocybin) versus placebo and entactogens (MDMA) versus placebo in a transparent and simple tabular format. In particular, we included key information concerning the certainty of the evidence, the magnitude of effect of the interventions examined, and the sum of available data on the outcomes.

Results

Description of studies

Please see the Characteristics of included studies table for full information about the six studies we included in this review. We contacted five corresponding authors requesting more information and additional data; three responded (Gasser 2014; Griffiths 2016; Holze 2022). We found seven ongoing studies.

Results of the search

We identified 440 records through searching bibliographic databases and 13 records through trial registries. We had 304 records in total after removal of duplicates (Figure 1). Review authors (KN, CB) evaluated the titles and abstracts, selecting relevant studies if they met the inclusion criteria. At this stage, we rejected 272 records. We obtained a full copy of 30 records that were potentially eligible to do a more detailed evaluation. After evaluating the records, we included 6 studies (17 records) and excluded 6 studies (6 records), while 7 studies (7 records) are still ongoing. We were able to include data from the six studies in our quantitative analysis.

1.

Included studies

In total, we identified six RCTs for inclusion in the review, with 149 randomised participants, 140 of whom were analysed: Gasser 2014 (12 participants randomised, 11 analysed); Griffiths 2016 (56 randomised, 51 analysed); Grob 2011 (12 randomised and analysed); Holze 2022 (20 randomised and analysed); Ross 2016 (31 randomised, 28 analysed); Wolfson 2020 (18 randomised and analysed). Ross 2016 had two follow‐up publications (Agin‐Liebes 2020; Ross 2021). Holze 2022 investigated participants suffering from anxiety with or without a life‐threatening disease; we included data from participants with a life‐threatening disease only, in accordance with our inclusion criteria (Types of participants). Drug companies were not involved in study funding, but funding was provided by organisations that promote psychedelic‐assisted therapy.

All six studies were cross‐over studies. They took place in the USA and Switzerland. Five of the studies were conducted in a single centre, four in the USA and one in Switzerland (Gasser 2014), and Holze 2022 was conducted in two centres in Switzerland. All studies had small sample sizes, with fewer than 50 participants per arm. Three studies compared psilocybin (0.2 to 0.4mg/kg) to an active placebo (low‐dose psilocybin (Griffiths 2016) or niacin (Grob 2011; Ross 2016)); two studies compared LSD (200 μg) to low‐dose LSD (Gasser 2014) or placebo (Holze 2022); and one study compared MDMA (75 to 125 mg ± 62.5 mg) to placebo (Wolfson 2020). Primary outcomes were measured between 2 and 8 weeks after one (Griffiths 2016; Grob 2011; Ross 2016) or two (Gasser 2014; Holze 2022; Wolfson 2020) psychedelic‐assisted sessions accompanied by preparation and integration therapy sessions.

Ongoing studies

We identified seven ongoing studies (ACTRN12619001225101; ACTRN12623000478617; NCT05214417; NCT05398484; NCT05403086; NCT05883540; EudraCT 2020‐005037‐32), which are investigating psilocybin, LSD (psychedelic and microdose), and ketamine in participants facing life‐threatening disease. For full details, see Characteristics of ongoing studies.

Excluded studies

We excluded six records from the full‐text screening because they did not meet the inclusion criteria: three were studies of ineligible design, two were commentaries, and one was a study with an ineligible population. See Characteristics of excluded studies.

Risk of bias in included studies

We present the risk of bias in the included studies graphically (Figure 2; Figure 3), and report details justifying our decisions in the Characteristics of included studies tables. All six studies had a high risk of bias in specified key domains, leading to a judgement of an overall high risk of bias in all studies. The main reason for giving a judgement of high risk of bias was a lack of blinding (causing performance and detection bias). Although all six studies had intended to blind participants, personnel, and assessors, blinding could not be achieved due to reasons inherent to studies investigating psychedelics; we discuss this in the Discussion section of the review.

2.

3.

Allocation

Overall, we judged four studies to be at low risk of selection bias (Gasser 2014; Griffiths 2016; Holze 2022; Wolfson 2020), and two studies to be unclear (Grob 2011; Ross 2016).

Random sequence generation (selection bias)

We judged four studies to be at low risk of bias as they used appropriate methods of randomisation (e.g. computer‐generated table of random numbers) (Gasser 2014; Griffiths 2016; Holze 2022; Wolfson 2020). We judged two studies to have an unclear risk of bias because they reported using randomisation but did not state the method of randomisation (Grob 2011; Ross 2016).

Allocation concealment (selection bias)

We judged five studies to have concealed allocation adequately; we judged one study to be unclear due to insufficient information describing the randomisation process (Grob 2011).

Blinding

Although blinding was attempted, blinding of participants, personnel, and assessors could not be achieved in any study for reasons that are inherent to studies investigating psychedelics, i.e. participants know if they have received the psychedelic or the placebo. Therefore, we judged all studies to be at high risk of performance and detection bias.

Incomplete outcome data

We judged the risk of attrition bias to be low in all six studies as the reasons for dropouts were provided or the overall low dropout rates were low.

Selective reporting

We judged the risk of reporting bias to be low in all six studies. Two studies reported outcomes that were not prespecified in their protocols (Grob 2011; Ross 2016). However, adding these outcomes (e.g. BDI) seemed reasonable, and the outcomes did not seem to have been selected based on the results from multiple analyses of the data.

Other potential sources of bias

We found carry‐over effects in all six cross‐over studies, which were caused by the long‐term effects of psychedelics; we therefore used data from the first period only in our quantitative analyses. Because the six studies seemed free of other potential sources of bias, we judged them to be at low risk of bias in this domain.

Effects of interventions

See: Table 1; Table 2

We investigated anxiety, depression, and existential distress as the primary outcomes. We were able to perform meta‐analyses comparing classical psychedelics versus placebo for the primary outcomes of anxiety and depression at up to 12 weeks. See Table 1. As all studies were cross‐over studies, with the carry‐over effects inherent to psychedelics, we included only the first study period (before cross‐over) in our meta‐analyses. We performed subgroup and sensitivity analyses as described in our methods sections: Subgroup analysis and investigation of heterogeneity and Sensitivity analysis.

Since all outcomes that were more than three months were measured after cross‐over, a control group was lacking due to carry‐over effects in all studies. We did not find any studies that used active comparators (e.g. antidepressants).

We present additional outcomes measured in the studies in the Characteristics of included studies table, and secondary outcomes in Table 3.

1. Secondary outcomes.

Study	Method/scale	Results
Quality of life
Gasser 2014	European Cancer Quality of Life Questionnaire (EORTC)‐QLQ V.1.0 (scale 0 to 100) Higher values mean better quality of life.	Baseline Active placebo (low‐dose LSD) group: 44.3 (SD 12.7), N = 3 High‐dose LSD group: 37.4 (SD 10.0), N = 8 2 months (before cross‐over) Active placebo (low‐dose LSD): 36.0 (SD 12.8), N = 3 High‐dose LSD: 50.0 (SD 14.9), N = 8 No P values reported
Griffiths 2016	McGill Quality of Life Questionnaire (MQOL), a self‐rated measure of overall quality of life (scale 0 to 10) Higher values mean better quality of life.	Baseline Active placebo (low‐dose psilocybin) group: 5.7 (SEM 0.24), N = 25 High‐dose psilocybin group: 5.3 (SEM 0.29), N = 26 Endpoint 5 weeks Active placebo (low‐dose psilocybin): 6.2 (SEM 0.32), N = 25 High‐dose psilocybin: 7.1 (SEM 0.29), N = 25 P < 0.05, between groups
Ross 2016	World Health Organisation Qualify of Life scale, brief version (WHO‐Bref), self‐report measure of quality of life in four domains (4 domains: physical, psychological, social relationships, environment, scale for each domain 4 to 20) Higher values mean better quality of life.	No scores are reported in the text. Post‐session 1 (2 weeks) between‐group differences: physical (P ≤ 0.01), psychological (P ≤ 0.01), social ( P value NS), environmental (P < 0.05)
Wolfson 2020	Global Assessment of Functioning (GAF), rates subjectively the social, occupational, and psychological functioning of an individual (scale 1 to 100) Administered by an independent rater. Higher values mean better quality of life.	Baseline Placebo group: 69.8 (SD 13.4), N = 5 MDMA group: 68.5 (SD 5.4), N = 13 1 month after 2 experimental sessions Placebo group: 72.8 (SD 7.7), N = 5 MDMA group: 75.1 (SD 9.9), N = 13 P = 0.52, between groups
	Functional Assessment of Chronic Illness Therapy Scale (FACIT), specifically designed for use in populations with LTDs (scales: physical‐well‐being 0 to 28; social/family well‐being 0 to 28, emotional well‐being 0 to 24; functional well‐being 0 to 28; additional concerns scale unclear because not specified). Higher values mean better quality of life.	Baseline (physical well‐being, social/family well‐being, emotional well‐being, functional well‐being, additional concerns) Placebo group: 19.8 (SD 6.7), 20.0 (SD 9.7), 14.0 (SD 5.9), 19.5 (SD 5.9), 24.8 (SD 14.4), N = 4 MDMA group: 21.6 (SD 4.2), 17.6 (SD 2.9), 14.7 (SD 3.0), 14.5 (SD 2.8), 24.0 (SD 9.2), N = 13 1 month after 2 experimental sessions Placebo group: 21.4 (SD 3.0), 17.6 (SD 6.3), 15.0 (SD 3.9), 18.8 (SD 7.1), 24.2 (SD 10.3), N = 5 MDMA group: 23.0 (SD 4.3), 18.5 (SD 3.8), 16.3 (SD 6.7), 19.3 (SD 6.3), 28.5 (SD 14.1), N =13 P values between groups: P = 0.61, P = 0.15, P = 0.87, P = 0.22, P = 0.45
Spirituality
Griffiths 2016	Question: how spiritually significant was the experience? (scale 1 to 6) Higher values mean more significance.	Endpoint 5 weeks Active placebo (low‐dose): 3.2 (SEM 0.24) High‐dose psilocybin: 4.5 (SEM 0.19) P ≤ 0.001, between groups
	Purpose in Life Test, a self‐rated measure of life meaningfulness (scale 20 to 140) Higher values mean more meaningfulness.	Baseline Active placebo (low‐dose psilocybin) group: 96.2 (SEM 3.32), N = 25 High‐dose psilocybin group: 91.0 (SEM 3.43), N = 26 Endpoint 5 weeks Active placebo (low‐dose psilocybin): 101.8 (SEM 3.78), N= 25 High‐dose psilocybin: 106.2 (SEM 3.0), N = 25 P value NS, between groups
Ross 2016	Question: How spiritually significant was the experience? (scale 1 to 6) Higher values mean more significance.	Endpoint 2 weeks Active placebo (niacin): 2.1 (SEM 0.33) High‐dose psilocybin: 4.5 (SEM 0.35) P ≤ 0.001, between groups
	Functional Assessment of Chronic Illness Therapy‐Spiritual Well‐Being (FACIT‐Sp), a self‐report measure of spiritual well‐being generating three scales: meaning/peace, faith, total spiritual well‐being (each subscale ranges from 0 to 32, 0 to 16, and 0 to 48, respectively) Higher values mean greater spiritual well‐being.	No values are reported in the text. Post‐session 1 (2 weeks) between‐group differences: meaning/peace (P ≤ 0.01), faith (P < 0.05), spiritual well‐being (P < 0.05)
Adverse events grade 3 and 4
Grob 2011		None
Gasser 2014		None
Griffiths 2016		None
Ross 2016		None
Holze 2022		None
Wolfson 2020		None
Serious adverse events
Grob 2011		None
Gasser 2014		None
Griffiths 2016		None
Ross 2016		None
Holze 2022		Four SAEs (two LSD periods, two placebo periods), which were considered unrelated to the treatment. Other SAEs reported by the authors were related to participants without a life‐threatening disease and are therefore not included here.
Wolfson 2020		None

AE: adverse event; EORTC‐QLQ: European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire; FACIT: Functional Assessment of Chronic Illness Therapy; FACIT‐Sp: Functional Assessment of Chronic Illness Therapy‐Spiritual Well‐Being; GAF: Global Assessment of Functioning; LTD: life‐threatening disease; LSD: lysergic acid diethylamide; MD: mean difference; MDMA: 3,4‐methylenedioxy‐methamphetamine; MQOL: McGill Quality of Life Questionnaire; N: number of participants; NA: not applicable; NS: not significant; PEQ: Persisting Effects Questionnaire; SAE: severe adverse event; SD: standard deviation; SEM: standard error of the mean; V: version; WHO‐Bref: World Health Organisation Qualify of Life scale, brief version

Classical psychedelics (psilocybin, LSD)

Five studies evaluated classical psychedelics (Gasser 2014; Griffiths 2016; Grob 2011; Holze 2022; Ross 2016). See Table 1.

Anxiety

Two subscales of the State‐Trait Anxiety Inventory (STAI) were used to measure different aspects of fear: fear as a condition (STAI‐state (STAI‐S)) and fear as a trait (STAI‐trait (STAI‐T)), both with a range of 20 to 80, with lower scores being better. Classical psychedelics (LSD and psilocybin) may result in a reduction of anxiety on both scales when compared to active placebo at time points up to 12 weeks (STAI‐T, MD −8.41, 95% CI −12.92 to −3.89; 5 studies, 122 participants; Analysis 1.1; low certainty of evidence; STAI‐S, MD −9.04, 95% CI −13.87 to −4.21; 5 studies, 122 participants; Analysis 1.2; low certainty of evidence). We downgraded the certainty of evidence to low because of the risk of bias and imprecision. The pooled estimates remained unchanged in sensitivity analysis using the fixed‐effect model. Subgroup analysis comparing studies with psilocybin and LSD did not show any evidence of a difference between the two psychedelics (STAI‐T: Chi² = 0.61, df = 1 (P = 0.43); STAI‐S: Chi² = 0.51, df = 1 (P = 0.48)).

1.1. Analysis.

Comparison 1: Classical psychedelics versus placebo, Outcome 1: Anxiety on STAI‐T (scale 20 to 80)

1.2. Analysis.

Comparison 1: Classical psychedelics versus placebo, Outcome 2: Anxiety on STAI‐S (scale 20 to 80)

Depression

Four studies with 112 participants measured this outcome using the Beck Depression Inventory (BDI) (Griffiths 2016; Grob 2011; Holze 2022; Ross 2016). Classical psychedelics (LSD and psilocybin) may reduce depression measured on the BDI (scale range 0 to 63) when compared to active placebo at time points up to 12 weeks (MD −4.92, 95% CI −8.97 to −0.87; Analysis 1.3; low certainty of evidence). We downgraded the certainty of the evidence to low because of imprecision and risk of bias. The pooled estimate remained robust in sensitivity analysis using the fixed‐effect model.

1.3. Analysis.

Comparison 1: Classical psychedelics versus placebo, Outcome 3: Depression on BDI (scale 0 to 63)

To put all five studies in one meta‐analysis, we calculated the SMD to combine results for depression measured on different scales (SMD −0.43, 95% CI −0.79 to −0.06; 5 studies, 122 participants; Analysis 1.4; low certainty of evidence; the SMD re‐expressed on the BDI scale was ‐3.11, 95% CI −5.72 to −0.43). We downgraded the certainty of evidence to low because of risk of bias and imprecision. The pooled estimate remained robust in sensitivity analysis using the fixed‐effect model. Subgroup analysis comparing studies with psilocybin and LSD did not show any evidence of a difference between the two psychedelics (Chi² = 0.07, df = 1 (P = 0.79)).

1.4. Analysis.

Comparison 1: Classical psychedelics versus placebo, Outcome 4: Depression on different scales (BDI, HADS‐D), SMD

Existential distress

Two studies with 79 participants presented outcomes measuring different aspects of existential distress on different scales (Griffiths 2016; Ross 2016). At the time point of 2 weeks, existential distress was lower in the high‐dose psilocybin group, when measured on the Demoralisation Scale (DS‐I), a self‐report measure of the cancer‐related demoralisation syndrome (active placebo (low‐dose psilocybin) group 39.6 (SEM 3.4), psilocybin group 18.8 (3.6), P ≤ 0.01). It was also lower on the Hopelessness Assessment and Illness (HAI) measure, a self‐report measure of hopelessness in advanced cancer (active placebo group 6.1 (SEM 0.8), psilocybin group 2.8 (0.7), P ≤ 0.01). However, on the Death Anxiety Scale (DAS), a self‐report questionnaire assessing the level of death anxiety, another aspect of existential distress remained unchanged at 2 weeks (active placebo group 7.8 (SEM 0.6), psilocybin group 6.6 (0.6), P value not significant) (Ross 2016).

Surrogate measures for the outcome of existential distress were inconclusive: Death Transcendence Scale (DTS), a self‐report measure of positive attitudes and adaptations to the finitude of life, was not higher in the high‐dose psilocybin group at 2 weeks (active placebo group 14.4 (SEM 5.9), psilocybin group 28.2 (5.7), P value not significant) (Ross 2016), or after 5 weeks (active placebo group MD 5.5 (SD 29.4), psilocybin group MD 10.6 (SD 26.5), P value not significant) (Griffiths 2016); however, the Life Attitude Profile Revised (LAP‐R Death Acceptance) was higher after 5 weeks (active placebo group: MD 1.4 (SD 15.2), psilocybin group MD 7.1 (SD 18.9), P ≤ 0.05) (Griffiths 2016). Summarising these findings, psychedelic‐assisted therapy may reduce existential distress more than active placebo when measured on some scales and not when measured on others, but the evidence is very uncertain.

Secondary outcomes

Quality of life, spirituality, serious adverse events, and adverse events grade 3/4 are summarised here and in Table 3.

Quality of life

Quality of life was reported in some studies and showed improvement before cross‐over (Gasser 2014; Griffiths 2016; Ross 2016; Table 1). Gasser 2014 (11 participants) did not calculate P‐values for secondary endpoints, while Griffiths 2016 (51 participants) reported a small improvement in quality of life (as measured by the McGill Quality of Life Questionnaire (MQOL), a self‐rated measure of overall quality of life, P < 0.05). Ross 2016, with 28 participants, reported statistically significant increases in quality of life measures, but did not specify the scores in the text. Summarising these findings, psychedelic‐assisted therapy using classical psychedelics may improve quality of life when compared to active placebo, but the evidence is very uncertain.

Spirituality

Griffiths 2016 (50 participants) and Ross 2016 (28 participants) reported measures related to spirituality (see Table 1). Participants were asked how spiritually significant the psychedelic experience was for them (on a scale of 1 to 6). The psilocybin group gave a higher rating to the experience compared to the active placebo group: active placebo group 3.2 (SEM 0.24) and 2.1 (SEM 0.33), psilocybin group 4.5 (SEM 0.2) and 4.5 (SEM 0.35), P ≤ 0.001, respectively. More comprehensive measures like the Purpose in Life test, a self‐rated measure of life meaningfulness, and the Functional Assessment of Chronic Illness Therapy‐Spiritual Well‐Being (FACIT‐Sp), a self‐report measure, showed conflicting results after treatment with psilocybin compared to active placebo (Table 3). Summarising these findings, psychedelics may induce spiritually significant experiences when compared to active placebo, but the evidence is very uncertain.

Serious adverse events (SAEs) and adverse events grade 3/4

No treatment‐related SAEs or adverse events grade 3/4 related to the intake of psychedelics or (active) placebo were reported (Table 1; Table 3). Holze 2022 reported one treatment‐related SAE in a participant without a life‐threatening disease who was not included in the analysis, which consisted of acute transient anxiety and delusions during an LSD session. The participant was successfully treated with lorazepam and olanzapine.

Other related adverse events were elevated blood pressure, nausea, anxiety, emotional distress, and psychotic‐like symptoms (Table 4), which subsided when drug effects wore off.

2. Adverse events (not grade 3, 4 or 5).

Study	Results
Grob 2011	"No clinical significant adverse events"
Gasser 2014	Related AEs during the sessions Affect lability: high‐dose session: 13.6% of participants, low‐dose session 0% Anger: high‐dose session: 4.5% of participants, low‐dose session 33.3% Anxiety: high‐dose session: 22.7% of participants, low‐dose session 50.0% Bradyphrenia: high‐dose session: 4.5% of participants, low‐dose session 0% Depersonalisation: high‐dose session: 4.5% of participants, low‐dose session 0% Derealisation: high‐dose session: 9.1% of participants, low‐dose session 0% Emotional distress: high‐dose session: 36.4% of participants, low‐dose session 33.3% Euphoric mood: high‐dose session: 4.5% of participants, low‐dose session 0% Feeling abnormal: high‐dose session: 40.9% of participants, low‐dose session 33.3% Feeling cold: high‐dose session: 45.4% of participants, low‐dose session 0% Feeling of relaxation: high‐dose session: 4.5% of participants, low‐dose session 0% Gait disturbance: high‐dose session: 31.8% of participants, low‐dose session 0% Hallucination: high‐dose session: 4.5% of participants, low‐dose session 0% Hyperhidrosis: high‐dose session: 13.6% of participants, low‐dose session 0% Secondary hypertension: high‐dose session: 4.5% of participants, low‐dose session 0% Illusion: high‐dose session: 72.7% of participants, low‐dose session 16.7% Mydriasis: high‐dose session: 18.2% of participants, low‐dose session 0% Perseveration: high‐dose session: 4.5% of participants, low‐dose session 0% Tachyphrenia: high‐dose session:. 0% of participants, low‐dose session 16.7% Thinking abnormally: high‐dose session: 9.1% of participants, low‐dose session 33.3% Vision blurred: high‐dose session: 0% of participants, low‐dose session 16.7%
Griffiths 2016	Related AEs during the sessions Nausea or vomiting: 15% of participants in the high‐dose session, none in the low‐dose session Physical discomfort (any type): high‐dose session: 21% of participants, low‐dose session 8% of participants Psychological discomfort (any type): high‐dose session: 32% of participants, low‐dose session 12% of participants Anxiety: high‐dose session: 26% of participants, low‐dose session 15% of participants Paranoid ideation: high‐dose session: 2% of participants, low‐dose session 0% of participants There were no cases of hallucinogen persisting perception disorder or prolonged psychosis.
Ross 2016	Related AEs during the sessions Elevations in BP and HR (76%, clinically not significant) Headaches/migraines (28%) Nausea (14%) Anxiety (17%) Near psychotic‐like symptoms (7%) Paranoid ideation (3%) Thought disorder (3%)
Wolfson 2020	Anxiety: MDMA: 3 (23.1%), placebo: 0 Dry mouth: MDMA 9 (69.2%), placebo 1 (20.0%) Headache: MDMA: 8 (61.5%), placebo 1 (20.0) Insomnia: MDMA: 2 (15.4%), placebo 1 (20.0) Jaw clenching, tight jaw: MDMA: 11 (84.6%), placebo 1 (20.0) Lack of appetite: MDMA: 4 (30.8%), placebo 0 Nausea: MDMA: 3 (23.1%), placebo 1 (20.0) Perspiration: MDMA: 9 (69.2%), placebo 0 Sensitivity to cold: MDMA: 2 (15.4%), placebo 1 (20.0) Thirst: MDMA: 11 (84.6%), placebo 2 (40.0)
Holze 2022	All participants included (with life‐threatening disease and non‐life‐threatening disease) During the sessions: Anxiety: LSD session: 7% of participants, placebo session: not reported Nausea: LSD session: 10% of participants, placebo session: not reported Headache: LSD session: 2% of participants, placebo session: not reported A total of 229 additional AEs were reported during the entire trial duration (105 during the LSD period and 124 during the placebo period). None of these AEs were considered related to treatment.

AE: adverse event; BP: blood pressure; HR: heart rate; LSD: lysergic acid diethylamide; MDMA: 3,4‐methylenedioxy‐methamphetamine

Follow‐up measurements after more than 3 months

Anxiety, depression, existential distress, quality of life, and spirituality were all measured after cross‐over, but a meaningful comparison of the psychedelic group and the placebo group was impossible due to carry‐over effects. Symptom reduction from pretreatment to posttreatment was sustained in most cases after 24 to 54 months. Two follow‐up studies of Ross 2016 suggested sustained reduction after an average of 3.2 and 4.5 years (Agin‐Liebes 2020; Ross 2021).

Entactogens (MDMA)

One study evaluated MDMA (Wolfson 2020). We assessed the certainty of evidence as very low for all primary endpoints; we downgraded by one level because of risk of bias and by two levels because of imprecision (Table 2).

Anxiety

Wolfson 2020 reported that measures of anxiety did not decrease in the MDMA group compared to the placebo group, but the evidence is very uncertain: STAI‐T (scale range 20 to 80), endpoint 1 month after the second session, MD −14.70, 95% CI −29.45 to 0.05; 18 participants; very low certainty evidence; Analysis 2.1; STAI‐S: MD −16.10, 95% CI −33.03 to 0.83; 18 participants; very low certainty evidence; Analysis 2.2. We downgraded the certainty of the evidence by one level because of risk of bias and by two levels because of imprecision.

2.1. Analysis.

Comparison 2: MDMA versus placebo, Outcome 1: Anxiety on STAI‐T (scale 20 to 80)

2.2. Analysis.

Comparison 2: MDMA versus placebo, Outcome 2: Anxiety on STAI‐S (scale 20 to 80)

Depression

Similarly, depression measured by the BDI‐II did not decrease in the MDMA group compared to the placebo group, but the evidence is very uncertain (MD −6.30, 95% CI −16.93 to 4.33; very low certainty evidence; Analysis 2.3).

2.3. Analysis.

Comparison 2: MDMA versus placebo, Outcome 3: Depression on BDI (scale 0 to 63)

Existential distress

Existential distress was not measured.

Secondary outcomes

Quality of life did not improve in the MDMA group compared to the placebo group, but the evidence is very uncertain. MDMA was well tolerated, and no participant had to stop the treatment due to side effects.

No treatment‐related SAEs or adverse events grade 3/4 related to the intake of MDMA were reported (Table 3). Reported adverse events were anxiety, dry mouth, jaw clenching, and headaches (Table 4). These reactions were short and mostly resolved by the end of the experimental session or during the following week.

Discussion

Summary of main results

This systematic review aimed to synthesise all available evidence from randomised controlled trials of psychedelic‐assisted therapy in adults with life‐threatening diseases. We identified six studies that assessed the effects of two different interventions (Gasser 2014; Griffiths 2016; Grob 2011; Holze 2022; Ross 2016; Wolfson 2020).

Classical psychedelic‐assisted therapy (psilocybin, LSD) may reduce symptoms of anxiety and depression compared to active placebo‐assisted therapy; existential distress may be reduced, but the evidence is very uncertain. Studies reported improved quality of life, but the evidence is very uncertain. When participants were asked how spiritually significant the psychedelic experience had been for them, they reported greater spiritual significance when receiving psychedelics compared to active placebo, but the evidence is also very uncertain. No treatment‐related serious adverse events and adverse events at grades 3 and 4 were reported by participants. The adverse events most commonly reported were mild or moderate, and subsided when drug effects wore off or soon after. These included elevated blood pressure, nausea, anxiety, emotional distress, and sometimes psychotic‐like symptoms.

The effect of MDMA‐assisted therapy on anxiety, depression, and quality of life compared to placebo‐assisted therapy was inconclusive, but the evidence is very uncertain. Future studies with larger sample sizes might change these findings. No treatment‐related serious adverse events and adverse events at grades 3 and 4 were reported by participants. The most common adverse events were anxiety, dry mouth, jaw clenching, and headaches, which subsided when drug effects wore off or soon after.

We derived our main results from endpoints before cross‐over (less than 12 weeks), because of carry‐over effects in all studies.

Overall completeness and applicability of evidence

The included studies reported data for all of our prespecified outcomes. We were able to meta‐analyse data for measures of anxiety and depression, but sample sizes were small. The main measure of existential distress (demoralisation) was only investigated in one study (Ross 2016), while another study reported potential surrogates for existential distress (Griffiths 2016).

While some studies included participants with advanced cancer (Griffiths 2016; Grob 2011; Ross 2016); other studies also included non‐malignant diseases (e.g. Parkinson’s disease, multiple sclerosis) (Gasser 2014; Holze 2022; Wolfson 2020). Due to the diversity and different character of the participant groups being investigated, the applicability of evidence is restricted to the particular patient group targeted by the specific intervention. All participants faced a life‐threatening disease causing a psychological condition. In terms of ethnicity, most of the participants in the studies were described as 'White'. Underrepresentation of ethnoracial minority populations is known in studies examining psychedelic therapies (Hughes 2024).

All the included studies were cross‐over studies, and so longer term outcomes could not be analysed due to carry‐over effects.

We identified six ongoing RCTs investigating the effectiveness of classical psychedelics (psilocybin, LSD) in participants with life‐threatening diseases (EudraCT 2020‐005037‐32; ACTRN12619001225101; ACTRN12623000478617; NCT05398484; NCT05403086; NCT05883540), and one study investigating the atypical psychedelic ketamine (NCT05214417). Some of these studies use ketamine as an active placebo, possibly reducing unblinding (EudraCT 2020‐005037‐32). Results from these studies might add relevant information to this review and possibly necessitate an update of it.

Ethical considerations in psychedelic‐assisted therapy

Patients in altered states of consciousness are, possibly more than in other therapeutic modalities, susceptible to ethical violations such as abuse by therapists (McNamee 2023). The challenge for psychedelic therapists is to develop a system of rigorous peer review and supervision to mitigate these risks (Anderson 2020). Professional groups and emerging professional societies are beginning to identify relevant core competencies and an ethical framework for psychedelic therapists regarding equity‐based obligations, special features of informed consent, and professional boundaries (McGuire 2024).

Quality of the evidence

We created two summary of findings tables, structured according to the GRADE system (Schünemann 2023), which show that we judged the certainty of evidence for the primary outcomes to be low or very low, i.e. the true population effects could be substantially different from the estimated effects in our review. We found low‐certainty evidence about the effect of psychedelic therapy using classical psychedelics (psilocybin, LSD) on anxiety and depression, and very low certainty of evidence for existential distress. We found very low‐certainty evidence for the effect of MDMA‐assisted therapy on anxiety and depression; measures of existential distress were not provided. The main factors influencing our downgrading of the certainty of the evidence were the risk of bias due to unblinding inherent to studies with psychedelics and imprecision due to small sample sizes. See Table 1 and Table 2.

Risk of bias

The overall risk of bias in all studies was high, mostly due to the unblinding of participants and personnel. Carry‐over effects in all cross‐over studies were not judged as high in the domain of 'other bias' because we included only data measured before the cross‐over in our meta‐analyses.

Blinding in RCTs of psychedelic‐assisted therapies is highly problematic. The attribution to placebo or psychedelic group was correctly guessed by most participants and therapists (Gasser 2014; Griffiths 2016; Holze 2022; Ross 2016; Wolfson 2020); one study reported on unblinding of participants and therapists narratively: "the drug order was almost always apparent to subjects and investigators whether the treatment was psilocybin or placebo"(Grob 2011). Furthermore, studies reported that participants often had prior experiences with psychedelics (Characteristics of included studies), possibly facilitating unblinding (White 2023). Unblinding can result in positive confirmation effects in the psychedelic group and reactions of disappointment in the placebo group that may reinforce positive or negative therapy results and can thus lead to an overestimation of effect sizes (Muthukumaraswamy 2021). Although this was not the case in the included studies in this review, unblinding may lead to attrition bias if more participants attributed to the placebo group leave the study due to disappointment at belonging to the placebo arm of the study. This is why strict implementation of intention‐to‐treat analyses is necessary. To minimise the risk of unblinding, active instead of inactive placebos can be used. Active placebos can be substances that trigger similar though not corresponding psychotropic effects compared to the substance under investigation, or, alternatively, smaller doses of the same psychedelic substance (Aday 2022). In this review, four of the six included studies used an active placebo (Gasser 2014; Griffiths 2016; Grob 2011; Ross 2016). Active placebos, however, are only partially helpful. Due to the exceedingly strong effects of high‐dose psychedelic substances, the use of active placebo mostly cannot prevent unblinding and the use of an active placebo may underestimate the true strength of the effect.

In the context of critically ill patients enroled in studies of psychedelic‐assisted therapy, high hopes and expectations may lead to expectancy bias. Additionally, the alliance between participants and therapists that is formed through the screening process may lead to various subjective bias phenomena such as rater bias, participant bias, and the Hawthorne effect. The media hype around psychedelics could further increase the expectations of people participating in a study investigating psychedelic‐assisted therapy (Aday 2022).

Since higher expectations correlate with better treatment outcomes, future studies should make expectations transparent. Expectation can be measured, for example, using the Credibility/Expectancy Questionnaire (CEQ) (Borkovec 1972; Devilly 2000), or the Stanford Expectation of Treatment Scale (SETS) (Younger 2012), and could be collected prospectively as a baseline screening. An alternative to these questionnaires could be questions ("How useful do you think the therapy will be to improve your symptom X?") (Aday 2022; Sherman 2010). With this, expectations can be recorded, analysed, and considered in the interpretation of the study results.

Another problematic aspect of studies of psychedelic‐assisted therapy potentially leading to bias is the setting, consisting of the therapeutic ritual (e.g. music, room design, scents, light or darkness, other integrated elements) and the interaction with the therapist or the therapist‐team. There is a clear positive correlation between therapeutic alliance and treatment outcome, therefore future studies could measure it (Flückiger 2018). In addition, the experimental conditions under which the therapy study takes place should be described in as much detail as possible. The setting can theoretically be standardised (e.g. the same music playlist was used for all test participants in Griffiths 2016), but this measure contradicts the basic idea of ​​psychedelic‐assisted therapy, since context‐related therapeutic rituals and interventions cannot and should not be standardised. Using an "enrichment factorial study design" could help to distinguish between substance effect and different domains of the setting (e.g. choice of music, design of the therapy room, etc) (Muthukumaraswamy 2021).

All six included studies used cross‐over design. However, this design is problematic. As the therapy effects are expected to be sustained and persistent (Agin‐Liebes 2020; Ross 2021), the risk of a carry‐over effect is very high. Cross‐over designs are presumably conducted to increase the statistical power of the studies and to give every participant the opportunity to receive the psychedelic. In terms of the latter, an alternative would be inclusion in an open‐label arm with administration of the psychedelic after the end of the study, i.e. a secondary open‐label cross‐over design (Wolfson 2020).

Psychedelic‐assisted therapy interweaves pharmacological and non‐pharmacological elements and is prone to unblinding, and as such poses real problems for the scientific premise of the gold standard "randomised controlled trial". This raises the question of alternative study designs, i.e. pragmatic clinical studies. While pragmatic studies lack internal validity, the primary focus of a well‐planned pragmatical clinical study is the external validity, generalisability, and effectiveness of the overall therapy composed of the totality of the therapy factors (Aday 2022; Carhart‐Harris 2022; Uthaug 2021). Unsuccessful application of blinding is not a shortcoming unique to clinical studies on psychedelic therapy but is also observed in other psychopharmacological studies (Hovmand 2023).

Imprecision

As suggested by Schünemann 2023, small sample sizes (less than 400 participants) prompted us to downgrade the certainty of evidence by one level. In some cases, we downgraded the certainty of evidence by two levels because, in addition to the small sample size, the confidence interval was wide and crossed the line of no effect.

Sources of heterogeneity

Pooling psilocybin and LSD in meta‐analyses

We pooled classical psychedelics (psilocybin, LSD) in these meta‐analyses because they both act at the 5‐HT_2A‐receptor. Subjective effects of the psychedelic experience and positive mood effects induced by both LSD and psilocybin were prevented by pre‐treatment with the 5‐HT_2A receptor antagonist ketanserin (Holze 2021; Kometer 2012; Kraehenmann 2017). Furthermore, the acute effects of the two substances could not be differentiated (Holze 2022), and both induce mystical‐type experiences and ego dissolution at higher dosages, which were positively correlated with treatment effects including the reduction of depression and anxiety (Griffiths 2006; Holze 2020).

Different therapy paradigms as sources of heterogeneity

Our definition of psychedelic‐assisted therapy means that the intake of psychedelics in all studies was accompanied by preparatory and integration sessions. Although the number and length of these sessions, the number of therapists (one or two), and the number of psychedelic‐assisted sessions (one to three) were different across the studies, heterogeneity remained low in all meta‐analyses. Differences in the therapeutic practices did not seem to influence statistical heterogeneity in this review. This could be the case because all studies more or less adhered to the therapeutic paradigm of providing preparation and integration sessions that was established in the 1960s and 1970s (Grof 1973; Turton 2014). Therapeutic frameworks in psychedelic‐assisted therapy are reviewed elsewhere (Thal 2024). Currently, it is unclear which therapeutic approach, e.g. cognitive‐behavioral and/or psychodynamic, may be beneficial to people (Burton 2022; Guss 2022; Yaden 2022).

Potential biases in the review process

To avoid potential bias in the review process, we were committed at all times to conducting a systematic review that followed published guidance in the Cochrane Handbook for Systematic Reviews of Interventions.

However, we made a post hoc change to the scope of our review (Differences between protocol and review). Our original intention was to focus on adults in palliative care. We had planned to include participants with advanced (incurable) disease and stated in our protocol that we would include studies when >75% of participants met these criteria (Schipper 2022). What we saw during the screening process was that the studies for potential inclusion often referred to 'life‐threatening diseases', but less than 75% of participants were incurable. Hence, we decided to change the inclusion criteria to "participants facing life‐threatening disease" as we considered the life threat to be the main reason for the psychological condition (anxiety, depression, existential distress), rather than the stage of the cancer or other terminal disease; we felt that excluding studies that did not meet our '>75%' criterion would substantially diminish the meaningfulness of this review.

We had to extract data from graphs in two studies because other data was not available (Grob 2011; Ross 2016).

We identified no other potential sources of bias in our review process.

Agreements and disagreements with other studies or reviews

Several reviews on psychedelic‐assisted therapy for people with life‐threatening diseases have been published over the last few years.

Our review agrees with the findings of Oliveira Maia 2022. The Oliveira Maia 2022 review used the Joanna Briggs Institute Critical Appraisal Checklist for randomised controlled trials and non‐randomised controlled trials (quasi‐experimental studies) to appraise the quality of studies and bias, but did not carry out a thorough risk of bias assessment or assessment of the certainty of the evidence through GRADE. No meta‐analyses were performed. Regarding RCTs, the authors included the same studies as we did in our review. The results suggest positive effects of psychedelic‐assisted therapies for symptom control in patients diagnosed with serious illness, with safety of use.

Schimmers 2022 conducted a systematic narrative review including RCTs, quasi‐RCTs, controlled clinical trials, before‐after studies, case series and case reports, and qualitative studies. They conclude that both classical and atypical psychedelics are promising treatment options in patients with terminal illness, but to draw conclusions about the effectiveness and safety of psychedelics, larger high‐quality studies of classical psychedelics and MDMA are needed. These conclusions are in agreement with ours.

Andersen 2021 conducted a comprehensive systematic review, including participants with life‐threatening diseases and other diseases. Holze 2022 and Wolfson 2020 were not included due to the publication date of the review. No risk of bias assessment or certainty of evidence judgements were performed. Meta‐analyses were not conducted because the authors deemed studies to be heterogenous (different study designs and drug types). In our opinion, classical psychedelics can be meta‐analysed together due to their common agonism at the 5HT_2A‐receptor (Quality of the evidence). The authors conclude that the resurrection of clinical psychedelic research provides early evidence for treatment efficacy and safety for a range of psychiatric conditions, which is, in principle, in agreement with our conclusions.

Ko 2022 conducted a systematic review of adults with or without life‐threatening disease with depressive symptoms. They performed risk of bias assessments and meta‐analyses. As the PICO (participants, interventions, comparators, outcomes) of the review differed substantially from ours, the results of the meta‐analyses cannot be directly compared. The analyses were characterised by high heterogeneity (I² of 84 to 97%), possibly due to the different populations being compared. In contrast, our meta‐analyses were not heterogenous. Furthermore, Holze 2022 was not yet included. Their risk of bias assessment (overall low risk of bias) is in disagreement with ours: for example, the Ko 2022 authors mostly judge performance and detection bias to be low, while we judge them to be high due to unblinding of participants and personnel. The authors conclude that the review and meta‐analysis suggest that administration of psychedelics significantly reduces depressive symptoms at several time points, most notably at short‐ and medium‐term follow‐up.

A narrative review by Ross 2022 concluded that historical and contemporary research suggests a role for psychedelic‐assisted therapy in patients with life‐threatening diseases. They claim that more rigorous research is needed to determine the efficacy, effectiveness, and mechanisms of action of psychedelic therapies to treat psychiatric and existential distress in these patients. Our results are in agreement with these statements.

Hovmand 2023 reviews the risk of bias in psychedelic‐assisted therapy studies using the RoB2 tool, which was in contrast to us as we used the RoB1 tool. They state that unblinding may inflate effect sizes (Moncrieff 2004), while using an active placebo might decrease effect sizes (Laursen 2023). Media coverage and hype may lead to expectancy bias. The overall conclusion of "overall high risk of bias" is congruent with our findings, although judgement on some subdomains may be different due to the different assessment tools used.

Yerubandi 2024 conducted a systematic review and meta‐analysis of adverse events of trials of psilocybin and included Griffiths 2016 and Ross 2016 and four other trials investigating psilocybin for depressed participants without a life‐threatening disease. The main identified adverse events were headaches, nausea, anxiety, dizziness, paranoia, blood pressure and heart rate elevation, visual perceptual effects, physical discomfort and mood changes typically resolving within 24 to 48 hours, which is in line with what is reported in our review.

Authors' conclusions

Implications for practice.

In those countries or regions that permit its use, psychedelic‐assisted therapy with classical psychedelics (psilocybin ('magic mushrooms') and LSD) may be an effective therapy to treat anxiety, depression, and possibly existential distress, in people facing a life‐threatening disease. Due to low to very‐low certainty of evidence, these results should be interpreted cautiously.

Psychedelic‐assisted therapy may be safe when provided in a medical setting, but the evidence is very uncertain. Screening for risk factors is paramount as studies do not include people with risk factors (e.g. uncontrolled hypertension), certain comorbidities (e.g. organ insufficiencies), or older age.

Implications for research.

Psychedelic drugs are illegal in most countries and research involving these drugs is restricted, though it is steadily increasing. Future studies ought to investigate demoralisation as the main measure of existential distress. The risk of bias due to unblinding necessitates changes in study methodology; for example, expectation could be measured to understand expectation bias, or blinding of participants and personnel could be investigated before cross‐over. Possibly, blinding could be improved by excluding participants with prior experience with psychedelics and by using active placebos that better prevent unblinding. More studies with larger sample sizes are needed to reduce imprecision. For all other than self‐reported measures, blinded assessors are needed. Studies should include participants of different ethnicities. Furthermore, future studies should address which therapeutic approach (e.g. cognitive‐behavourial or psychodynamic or both) may be more beneficial and whether participants with advanced (incurable) disease respond differently to psychedelic‐assisted therapy than those facing a potentially curable life‐threatening disease. Trials of parallel design or secondary cross‐over design would allow assessment of long‐term outcomes.

The publication of the results of ongoing studies may necessitate an update of this review. The conclusions of an updated review could differ from those of the present review and may provide clearer evidence regarding the efficacy and safety of psychedelic‐assisted therapy for people facing a life‐threatening disease.

History

Protocol first published: Issue 11, 2022

Appendices

Appendix 1. MEDLINE (OVID)

1. hallucinogens/ or 3,4‐methylenedioxyamphetamine/ or lysergic acid diethylamide/ or psilocybin/ or ketamine/ or Banisteriopsis/ or Ibogaine/ or Mescaline/

2. (hallucinogen* or psychedelic* or entactogen).tw.

3. (MDMA or LSD or DPT or DMT).tw.

4. (psilocybin or lysergic acid diethylamide or 3 4‐methylenedioxyamphetamine* or dipropyltryptamine or dimethyltryptamine or ayahuasca or ketamin* or ibogaine or mescaline or peyote or esketamine).tw.

5. 1 or 2 or 3 or 4

6. exp Psychotherapy/

7. psychotherap*.tw.

8. (psycho* adj3 therap*).tw.

9. Counseling/

10. counsel*.tw.

11. (relaxation or imagery or (behavio#r adj3 therap*)).tw.

12. biofeedback.tw.

13. (stress adj2 manag*).tw.

14. Depression/ or Demoralization/ or Adjustment Disorders/

15. anxiety/ or catastrophization/ or psychological distress/

16. (anxiet* or anxious or demoral* or depress* or "existential distress" or agitat* or panic* or hopeless* or catastrophi?ation or adjustment disorder*).tw.

17. 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16

18. 5 and 17

19. ((advance* or late or last or end or final) adj4 (stage* or phase*)).tw.

20. (palliat* or terminal* or endstage or end stage).tw.

21. ((end adj3 life) or (care adj3 dying)).tw.

22. ((advance* or progressi* or terminal*) adj6 (ill* or disease* or condition*)).tw.

23. hospice*.tw.

24. (hospice or (nursing adj3 home*)).tw.

25. (life adj3 threat*).tw.

26. Palliative Care/ or terminally ill/

27. terminal care/ or hospice care/

28. 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27

29. 18 and 28

30. randomized controlled trial.pt.

31. controlled clinical trial.pt.

32. randomized.ab.

33. placebo.ab.

34. drug therapy.fs.

35. randomly.ab.

36. trial.ab.

37. 30 or 31 or 32 or 33 or 34 or 35 or 36

38. exp animals/ not humans.sh.

39. 37 not 38

40. 29 and 39

Appendix 2. CENTRAL

#1 MeSH descriptor: [Hallucinogens] this term only

#2 MeSH descriptor: [3,4‐Methylenedioxyamphetamine] this term only

#3 MeSH descriptor: [Lysergic Acid Diethylamide] this term only

#4 MeSH descriptor: [Psilocybin] this term only

#5 MeSH descriptor: [Ketamine] this term only

#6 MeSH descriptor: [Banisteriopsis] this term only

#7 MeSH descriptor: [Ibogaine] this term only

#8 MeSH descriptor: [Mescaline] this term only

#9 ((hallucinogen* or psychedelic* or entactogen)):ti,ab,kw (Word variations have been searched)

#10 ((MDMA or LSD or DPT or DMT)):ti,ab,kw (Word variations have been searched)

#11 ((psilocybin or dipropyltryptamine or dimethyltryptamine or ayahuasca or ketamin* or ibogaine or mescaline or peyote or esketamine)):ti,ab,kw (Word variations have been searched)

#12 (lysergic acid diethylamide):ti,ab,kw (Word variations have been searched)

#13 (methylenedioxyamphetamine*):ti,ab,kw (Word variations have been searched)

#14 {OR #1‐#13}

#15 MeSH descriptor: [Psychotherapy] explode all trees

#16 (psychotherap*):ti,ab,kw (Word variations have been searched)

#17 ((psycho* Near/3 therap*)):ti,ab,kw (Word variations have been searched)

#18 MeSH descriptor: [Counseling] this term only

#19 (counsel*):ti,ab,kw (Word variations have been searched)

#20 ((relaxation or imagery or (behavio#r near/3 therap*))):ti,ab,kw (Word variations have been searched)

#21 (biofeedback):ti,ab,kw (Word variations have been searched)

#22 (stress near/2 manag*):ti,ab,kw (Word variations have been searched)

#23 MeSH descriptor: [Depression] this term only

#24 MeSH descriptor: [Demoralization] this term only

#25 MeSH descriptor: [Adjustment Disorders] this term only

#26 MeSH descriptor: [Anxiety] this term only

#27 MeSH descriptor: [Catastrophization] this term only

#28 MeSH descriptor: [Psychological Distress] this term only

#29 ((anxiet* or anxious or demoral* or depress* or "existential distress" or agitat* or panic* or hopeless* or catastrophi?ation or adjustment disorder*)):ti,ab,kw (Word variations have been searched)

#30 {OR #15‐#29}

#31 (((advance* or late or last or end or final) near/4 (stage* or phase*))):ti,ab,kw (Word variations have been searched)

#32 ((palliat* or terminal* or endstage or end stage)):ti,ab,kw (Word variations have been searched)

#33 (((end near/3 life) or (care near/3 dying))):ti,ab,kw (Word variations have been searched)

#34 (((advance* or progressi* or terminal*) near/6 (ill* or disease* or condition*))):ti,ab,kw (Word variations have been searched)

#35 (hospice*):ti,ab,kw (Word variations have been searched)

#36 ((hospice or (nursing near/3 home*))):ti,ab,kw (Word variations have been searched)

#37 ((life near/3 threat*)):ti,ab,kw (Word variations have been searched)

#38 MeSH descriptor: [Palliative Care] this term only

#39 MeSH descriptor: [Terminally Ill] this term only

#40 MeSH descriptor: [Terminal Care] this term only

#41 MeSH descriptor: [Hospice Care] this term only

#42 {OR #31‐#41}

#43 #14 and #30 and #42

Appendix 3. Embase

1 psychedelic agent/ or mescaline/ or psilocybine/

2 3,4 methylenedioxyamphetamine/

3 ketamine/

4 Banisteriopsis/

5 ibogaine/

6 lysergide/

7 (hallucinogen* or psychedelic* or entactogen).tw.

8 (MDMA or LSD or DPT or DMT).tw.

9 (psilocybin or lysergic acid diethylamide or 3 4‐methylenedioxyamphetamine* or dipropyltryptamine or dimethyltryptamine or ayahuasca or ketamin* or ibogaine or mescaline or peyote or esketamine).tw.

10 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9

11 exp psychotherapy/

12 psychotherap*.tw.

13 (psycho* adj3 therap*).tw.

14 counseling/

15 counsel*.tw.

16 (relaxation or imagery or (behavio#r adj3 therap*)).tw.

17 biofeedback.tw.

18 (stress adj2 manag*).tw.

19 depression/

20 demoralization/

21 adjustment disorder/

22 anxiety/

23 catastrophizing/

24 distress syndrome/

25 (anxiet* or anxious or demoral* or depress* or "existential distress" or agitat* or panic* or hopeless* or catastrophi?ation or adjustment disorder*).tw.

26 or/11‐25

27 10 and 26

28 ((advance* or late or last or end or final) adj4 (stage* or phase*)).tw.

29 (palliat* or terminal* or endstage or end stage).tw.

30 ((end adj3 life) or (care adj3 dying)).tw.

31 ((advance* or progressi* or terminal*) adj6 (ill* or disease* or condition*)).tw.

32 hospice*.tw.

33 (hospice or (nursing adj3 home*)).tw.

34 (life adj3 threat*).tw.

35 exp palliative therapy/

36 terminally ill patient/

37 terminal care/

38 hospice care/

39 or/28‐38

40 27 and 39

41 random$.tw.

42 factorial$.tw.

43 crossover$.tw.

44 cross over$.tw.

45 cross‐over$.tw.

46 placebo$.tw.

47 (doubl$ adj blind$).tw.

48 (singl$ adj blind$).tw.

49 assign$.tw.

50 allocat$.tw.

51 volunteer$.tw.

52 Crossover Procedure/

53 double‐blind procedure.tw.

54 Randomized Controlled Trial/

55 Single Blind Procedure/

56 41 or 42 or 43 or 44 or 45 or 46 or 47 or 48 or 49 or 50 or 51 or 52 or 53 or 54 or 55

57 (animal/ or nonhuman/) not human/

58 56 not 57

59 40 and 58

Appendix 4. ClinicalTrials.gov

(depression OR anxiety OR existential distress) AND (psychedelic OR LSD OR psilocybin OR MDMA OR Ayahuasca)

Appendix 5. WHO ICTRP

depression OR anxiety OR existential distress AND psychedelic (intervention)

Data and analyses

Comparison 1. Classical psychedelics versus placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Anxiety on STAI‐T (scale 20 to 80)	5	122	Mean Difference (IV, Random, 95% CI)	‐8.41 [‐12.92, ‐3.89]
1.1.1 LSD versus placebo	2	31	Mean Difference (IV, Random, 95% CI)	‐11.34 [‐19.93, ‐2.74]
1.1.2 Psilocybin versus placebo	3	91	Mean Difference (IV, Random, 95% CI)	‐7.29 [‐12.60, ‐1.99]
1.2 Anxiety on STAI‐S (scale 20 to 80)	5	122	Mean Difference (IV, Random, 95% CI)	‐9.04 [‐13.87, ‐4.21]
1.2.1 LSD versus placebo	2	31	Mean Difference (IV, Random, 95% CI)	‐11.89 [‐21.43, ‐2.35]
1.2.2 Psilocybin versus placebo	3	91	Mean Difference (IV, Random, 95% CI)	‐7.85 [‐13.56, ‐2.14]
1.3 Depression on BDI (scale 0 to 63)	4	111	Mean Difference (IV, Random, 95% CI)	‐4.92 [‐8.97, ‐0.87]
1.3.1 LSD versus placebo	1	20	Mean Difference (IV, Random, 95% CI)	‐5.80 [‐19.59, 7.99]
1.3.2 Psilocybin versus placebo	3	91	Mean Difference (IV, Random, 95% CI)	‐4.83 [‐9.07, ‐0.60]
1.4 Depression on different scales (BDI, HADS‐D), SMD	5	122	Std. Mean Difference (IV, Random, 95% CI)	‐0.43 [‐0.79, ‐0.06]
1.4.1 LSD versus placebo	2	31	Std. Mean Difference (IV, Random, 95% CI)	‐0.34 [‐1.08, 0.40]
1.4.2 Psilocybin versus placebo	3	91	Std. Mean Difference (IV, Random, 95% CI)	‐0.46 [‐0.87, ‐0.04]

Comparison 2. MDMA versus placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Anxiety on STAI‐T (scale 20 to 80)	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
2.1.1 MDMA versus placebo	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
2.2 Anxiety on STAI‐S (scale 20 to 80)	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
2.2.1 MDMA versus placebo	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
2.3 Depression on BDI (scale 0 to 63)	1		Mean Difference (IV, Random, 95% CI)	Totals not selected
2.3.1 MDMA versus placebo	1		Mean Difference (IV, Random, 95% CI)	Totals not selected

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Gasser 2014.

Study characteristics
Methods	Study design: cross‐over Total study duration: 12 months Number of study centres: 1 Location: Switzerland Study setting: outpatient Date of the study: 2008 to 2012
Participants	Number randomised: 12 (8 high dose, 4 low dose) Number analysed High dose in the 1st period: baseline (8), 2‐months follow up, (8), 12‐month follow‐up (6) Low dose in 1st period and high dose after cross‐over: baseline (4), 2‐month follow up (3), 12‐month follow‐up (3) Mean age: 51.7 years (SD 9.7) Age range: 39 to 64 years Sex: 4 females (33%) Ethnicity: 11 white (100%) Severity of the condition: "life‐threatening disease" Diagnostic criteria: Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID) Inclusion criteria: a score > 40 on the state or trait STAI scale; life‐threatening disease Exclusion criteria: current alcohol or drug dependence (except caffeine or nicotine); primary psychotic, bipolar affective, or dissociative disorder; neurocognitive impairment; pregnant or nursing women Prior psychedelic use: 11 of 12 participants had no experience with LSD
Interventions	High‐dose LSD group: 2 sessions of 200 µg LSD (weight of free base) Active placebo group (low‐dose LSD): 2 sessions of 20 µg LSD (weight of free base) 2 preparation sessions before substance‐assisted sessions, 6 integration sessions (3 after each substance‐assisted session), open‐label cross‐over for the placebo group, male and female therapist Comparison: NA Concomitant medications or excluded medications: taper off of antidepressants and antianxiety medications; avoid alcohol and recreational drugs for 24 hours before substance‐assisted session
Outcomes	Outcomes as specified and collected in Types of outcome measures: STAI‐State‐Anxiety, STAI‐Trait‐Anxiety, HADS‐Anxiety, HADS‐Depression, EORTC‐QLQ Outcomes reported at: baseline, [2 substance sessions], 2‐month follow‐up, [cross‐over, open label], [2 substance sessions], 2‐month follow‐up, 12‐month follow up
Notes	Funding for study: Schweizerische Aerztegesellschaft für Psycholytische Therapie (SAEPT, Swiss Medical Society for Psycholytic Therapy) and Multidisciplinary Association for Psychedelic Studies (MAPS) Possible conflicts of interest of the study authors: Rick Doblin (president) and Berra Yazar‐Klosinski were employees of the main sponsor MAPS. Peter Gasser was president of SAEPT.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	A computer program was used (information was sought from the author).
Allocation concealment (selection bias)	Low risk	Quote: "Capsules were of identical size, colour, and shape and were bottled in sequentially numbered containers."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "For all 24 blinded sessions, all participants correctly guessed the dose of LSD that was administered, and both therapists guessed incorrectly in one active placebo session."
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "For all 24 blinded sessions, all participants correctly guessed the dose of LSD that was administered, and both therapists guessed incorrectly in one active placebo session." Most efficacy outcomes were self‐reported by participants.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No dropouts before cross‐over.
Selective reporting (reporting bias)	Low risk	Primary outcome (STAI‐state/trait) was prespecified in the protocol (NCT00920387).
Other bias	Low risk	The study appears to be free of other sources of bias.

Griffiths 2016.

Study characteristics
Methods	Study design: cross‐over Total study duration: 6 months Number of study centres: 1 Location: Maryland, USA Study setting: outpatient Date of the study: 2007 to 2014
Participants	Number randomised: 56 (high‐dose‐1st group 29, low‐dose‐1st group 27) Number analysed "High‐dose‐1st" group: baseline (26), post session 1 (25 or 26), post session 2 (25), 6 months (24) "Low‐dose‐1st" group: baseline (25), post session 1 (25), post session 2 (24), 6 months (22) Mean age: 56.3 (SEM 1.4) Age range: NR Sex: 25 females (49%) Ethnicity: 48 white (94%) Severity of the condition: potentially life‐threatening cancer diagnosis (> 65% recurrent or metastatic) Diagnostic criteria: DSM‐IV diagnosis that includes anxiety or mood symptoms, or both Inclusion criteria (detailed list in the supplementary of the original report): 21 to 80 years old; ECOG performance status of 0, 1, or 2; has or has had a cancer diagnosis that is potentially life‐threatening; DSM‐IV diagnosis: chronic adjustment disorder with anxiety, chronic adjustment disorder with mixed anxiety and depressed mood, dysthymic disorder, generalised anxiety disorder (GAD), major depressive disorder (MDD), or a dual diagnosis of GAD and MDD, or GAD and dysthymic disorder Exclusion criteria (detailed list in the supplementary material of the original report): cancer with known CNS involvement, or other major CNS disease; hepatic dysfunction; cardiovascular conditions; epilepsy with a history of seizures; renal insufficiency; various psychiatric exclusion criteria (e.g. schizophrenia, psychotic disorder) Prior psychedelic use: lifetime use of hallucinogens 45% of participants
Interventions	High‐dose psilocybin group: 1 session, 30/22 mg/70 kg (trialists reduced dosage in the study from 30 to 22 after 2 withdrawals) Low‐dose psilocybin group: 1 session, 3/1 mg/70 kg 3 preparation sessions (together with an average of 7.9 hours) before substance‐assisted session, 2‐3 integration sessions (3.4 hours); after cross‐over: 2‐3 integration sessions (2.5 hours), 2 therapists Concomitant medications: 6 participants reported initiating medication treatment with an anxiolytic (2 participants), antidepressant (3), or both (1) between the post‐session‐2 and the 6‐month follow‐up assessments Excluded medications (detailed list in the supplementary material of the original paper) Weaning of medications having a primary pharmacological effect on serotonin neurones, or medications that are MAO inhibitors Potent metabolic inducers or inhibitors
Outcomes	Outcomes as specified and collected in Types of outcome measures: HAM‐Anxiety, HADS‐Anxiety, STAI‐State‐Anxiety, GRID‐HAM‐Depression‐17, BDI, HADS‐Depression, LAP‐R Death Acceptance, DTS, MQOL, Purpose in Life Test Outcomes reported at: baseline, 5 weeks after session 1 (period 1), [cross‐over], 5 weeks after session 2 (period 2), 6 months
Notes	Funding for study: Heffter Research Institute, the Riverstyx Foundation, William Linton, the Betsy Gordon Foundation, the McCormick Family, the Fetzer Institute, George Goldsmith, and Ekaterina Malievskaia, NIH grant RO1DA03889 Possible conflicts of interest of study authors: Roland Griffiths was on the Board of Directors of the Heffter Research Institute. "Sponsors did not have roles in the collection, management, analysis, and interpretation of the data or in the preparation and approval of the manuscript." The one exception to this was that Cody Swift from the Riverstyx Foundation served as a session monitor for two volunteers.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	A randomised number table from a statistical handbook was used (information was sought from the author).
Allocation concealment (selection bias)	Low risk	Quote: "The randomization was managed by the Research Pharmacy and investigators, research staff, and participants were blinded to the randomization sequence."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Although efforts were made to minimise differential effects between low and high doses, unblinding was likely due to the effects of psychedelics.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Although efforts were made to minimise differential effects between low and high doses, unblinding was likely due to the effects of psychedelics. Most efficacy outcomes were self‐reported by participants.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups (e.g. disease progression).
Selective reporting (reporting bias)	Low risk	Primary outcome (STAI state/trait) was prespecified in the protocol (NCT00465595).
Other bias	Low risk	The study appears to be free of other sources of bias.

Grob 2011.

Study characteristics
Methods	Study design: cross‐over Total study duration: 6 months Number of study centres: 1 Location: California, USA Study setting: admitted for substance‐assisted session, but otherwise outpatient Date of the study: 2004 to 2008
Participants	Number randomised: 12 (psilocybin 6, niacin 6) Number analysed: 12 at 3‐month follow‐up, 11 at 4‐month follow‐up, 8 at 6‐month follow‐up Mean age: NR Age range: 36 to 58 years Sex: 11 females (92%) Ethnicity: NR Severity of the condition: "advanced stages of illness" Diagnostic criteria: DSM IV Inclusion criteria: advanced‐stage cancer and anxiety; 18 to 70 years old; diagnosis of acute stress disorder, GAD, anxiety disorder due to cancer, adjustment disorder with anxiety Exclusion criteria: central nervous system cancer; severe cardiovascular disease; untreated hypertension; abnormal hepatic or renal function; diabetes; life‐time event of schizophrenia, bipolar disease, other Prior psychedelic use: 4/12 no use
Interventions	Psilocybin group: 0.2 mg/kg Active placebo (niacin) group: 250 mg Two preparation sessions before the substance‐assisted session, 1 integration session, "therapeutic team" Concomitant medications: prescription or over‐the‐counter non‐narcotic pain medications at any time and narcotic pain medications up to 8 hours before and 6 hours after administration Excluded medications: active cancer chemotherapy, anti‐seizure medications, insulin and oral hypoglycemics, and psychotropic medications in the previous 2 weeks
Outcomes	Outcomes as specified and collected inTypes of outcome measures: STAI‐State‐Anxiety, STAI‐Trait‐Anxiety, BDI Outcomes reported at: baseline, [session 1, period 1], 1 day after session 1, 2 weeks after session 1, [cross‐over], [session 2], 1 to 6 months (monthly)
Notes	Funding for study: Heffter Research Institute; Betsy Gordon Foundation; Nathan Cummings Foundation; grant M01‐RR00425 from National Institutes of Health to Harbor‐UCLA General Clinical Research Center Possible conflicts of interest of study authors: Dr. Grob and Dr. Greer are on the board of directors of the Heffter Research Institute.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Sequence generation was not reported.
Allocation concealment (selection bias)	Unclear risk	Not enough information about the randomisation process.
Blinding of participants and personnel (performance bias) All outcomes	High risk	The drug order was almost always apparent to participants and investigators whether the first treatment was psilocybin or placebo.
Blinding of outcome assessment (detection bias) All outcomes	High risk	The drug order was almost always apparent to subjects and investigators whether the first treatment was psilocybin or placebo. Most efficacy outcomes were self‐reported by participants.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No dropouts.
Selective reporting (reporting bias)	Low risk	Several reported primary outcomes were not prespecified in the protocol (NCT00302744), but the addition of these outcomes (e.g. BDI) was reasonable and does not seem to have been selected based on the results from multiple analyses of the data.
Other bias	Low risk	The study appears to be free of other sources of bias.

Holze 2022.

Study characteristics
Methods	Study design: cross‐over Total study duration: 54 weeks (12.9 months) Number of study centres: 2 Location: Switzerland Study setting: outpatient Date of the study: 2017 to 2021
Participants	Number randomised: 20 participants with life‐threatening disease (LSD 10, placebo 10) Number analysed: 20 Mean age: 46 years (SD 13) Age range: NR Sex: 11 females (55%) Ethnicity: 20 white (100%) Severity of the condition: score ≥ 40 on the state or trait STAI scale Diagnostic criteria: DSM‐IV criteria for an anxiety disorder, including generalised anxiety disorder, social phobia, and panic disorder Inclusion criteria (detailed list in the supplementary material of the original report): > 25 years old; DSM‐IV anxiety related disorder or STAI state or trait > 40; 40% or more of the participants should have a diagnosis of a fatal illness (autoimmune, neurological, or cancer without CNS involvement) and life expectancy > 12 months; discontinuation of psychiatric medications (5 times the particular's drug half live, typically 3 to 7 days) Exclusion criteria (detailed list in the supplementary material of the original report): age < 25 years; past or present diagnosis of a primary psychotic disorder; first‐degree relative with psychotic disorders; past or present bipolar disorder (according to the DSM‐IV); current substance use disorder (within the past 2 months according to the DSM‐IV, except nicotine); suicide risk; likely to require psychiatric hospitalisation during the course of the study; central nervous system involvement of the cancer Prior psychedelic use: NR
Interventions	LSD group: 2 sessions a 200 µg LSD (weight of free base) Placebo group: ethanol (1 ml) Two treatment periods consisted of 2 treatment sessions and 5 study visits. Treatment sessions were separated by 6 weeks; the length of study was 62 weeks. Study visits were conducted at baseline (1 preparation session), between the sessions, and 2, 8, and 16 weeks (4 integration sessions per period) after the second treatment session. The week‐16 visit in the second period also served as the end‐of‐study visit. Concomitant medications: long‐term pain medication, caffeine, nicotine Excluded medications: psychiatric medications (tapered before substance‐assisted session: for at least 5 times t1/2), psychoactive drugs at least 24 hours before substance‐assisted session
Outcomes	Outcomes as specified and collected in Types of outcome measures: STAI‐Trait‐Anxiety, STAI‐State‐Anxiety, BDI Outcomes reported at Period 1: baseline, [session 1], 5 weeks, [session 2], 10, 16, 24 weeks, [cross‐over] Period 2: baseline (26 weeks), [session 1], 31 weeks, [session 2], 36, 32, 50 weeks
Notes	Funding for study: University Hospital Basel (to Dr. Liechti, MEL), Swiss Medical Society for Psycholytic Therapy (to Dr. Gasser, PG) Possible conflicts of interest of trial authors: data associated with this work and owned by the University Hospital Basel were licensed by Mind Medicine, Inc. Mind Medicine, Inc had no role in financing, planning, or conducting this study or this publication. MEL is a consultant for Mind Medicine, Inc. PG is a consultant for Mind Medicine, Inc., CompassPathways, and Reconnect Foundation. The other authors report no biomedical financial interests or potential conflicts of interest. The funders of the study had no role in the study design, data collection, data analysis, data interpretation, or writing the report.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computerised block randomisation (balanced blocks of 2 and 4 participants) was used.
Allocation concealment (selection bias)	Low risk	Quote: "For allocation concealment medication was prepackaged and numbered by Patient‐ID and Session (1‐4)."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "The characteristic effects of LSD unblinded the treatment order assignment in most patients once the effects of LSD were perceived."
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "The characteristic effects of LSD unblinded the treatment order assignment in most patients once the effects of LSD were perceived." Most efficacy outcomes were self‐reported by participants.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No dropouts.
Selective reporting (reporting bias)	Low risk	Primary outcome (STAI) was prespecified in the protocol (NCT03153579).
Other bias	Low risk	The study appears to be free of other sources of bias.

Ross 2016.

Study characteristics
Methods	Study design: cross‐over Total study duration: 9 months Number of study centres: 1 Location: New York State, USA Study setting: outpatient Date of the study: 2009 to 2014
Participants	Number randomised: 31 (psilocybin 16, niacin 15) Number analysed Psilocybin 1st session and niacin on the 2nd: 14, 14, 12, and 11 (baseline (14), post session 1 (14), post session 2 (12), and 12 months (11) Niacin 1st session and psilocybin on the 2nd: baseline (15), post session 1 (14), post session 2, (12) and 12 months (12) Mean age: 56.28 (SD 12.93) Age range: 22 to 75 years Sex: 18 females (62%) Ethnicity: 26 white (90%) Severity of the condition: nearly two‐thirds of participants (62%) had advanced cancers (stages III or IV) Diagnostic criteria: DSM IV Inclusion criteria: diagnosis of acute stress disorder, generalised anxiety disorder, anxiety disorder due to cancer, or adjustment disorder with anxiety +/‐ depression Exclusion criteria: epilepsy; renal disease; diabetes; abnormal liver function; severe cardiovascular disease; malignant hypertension; baseline blood pressure more than 140/90; personal history or immediate family members with schizophrenia, bipolar affective disorder, delusional disorder, schizoaffective disorder or other psychotic spectrum illness; current substance use disorder Prior psychedelic use: 55% of participants had used psychedelics at one point in their life.
Interventions	Psilocybin group: 0.3 mg/kg Active placebo group: 250 mg niacin 3 preparation sessions (2 hours each) before the substance‐assisted session, 3 integration sessions (2 hours each) after; 2 therapists Concomitant medications: "none were on any psychotropics at the time of study enrolment" Excluded medications: anti‐seizure medications, insulin, oral hypoglycemics, clonidine, aldomet, cardiovascular medications, antipsychotics (first and second generation), antidepressants, mood stabilisers
Outcomes	Outcomes reported in the study used in the review (Types of outcome measures): HADS‐A, HADS‐D, BDI, STAI‐S, STAI‐T, WHO‐Bre, FACIT‐Sp, Demoralisation Scale Outcomes reported at: baseline, 6 weeks after session 1 (period 1), [cross‐over], 6 weeks after session 2 (period 2), 12 months
Notes	Funding for trial: Heffter Research Institute, RiverStyx Foundation, New York University‐Health and Hospitals Corporation (NYU‐HHC) Clinical and Translational Science Institute (CTSI) (NYU CTSA grant UL1 TR000038 from National Center for Advancing Translational Sciences, National Institutes of Health), Carey and Claudia Turnbull, William Linton, Robert Barnhart, Arthur Altschul, Kelly Fitzsimmons, George Goldsmith, and Ekaterina Malievskaia Notable conflicts of interest of trial authors: the authors declared no potential conflicts of interest with respect to the research, authorship, or publication of the article.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Block randomisation methodology (supplement). Unclear how blocks were generated
Allocation concealment (selection bias)	Low risk	Quote: "The random allocation sequence was available only to administrative staff (Dr. Patricia Corby) at the Bluestone Center for Clinical Research."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Quote: "Staff members guessed correctly in 28/29 participants (97%) whether the participant received the psilocybin or the active placebo. Participants were not asked."
Blinding of outcome assessment (detection bias) All outcomes	High risk	The specific effects of psilocybin were most likely known to participants. Most efficacy outcomes were self‐reported by participants.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups (e.g. disease progression).
Selective reporting (reporting bias)	Low risk	Reported primary outcomes were not prespecified in the protocol (NCT00957359), but adding these outcomes (e.g. BDI) was reasonable and does not seem to have been selected on the basis of the results from multiple analyses of the data.
Other bias	Low risk	The study appears to be free of other sources of bias.

Wolfson 2020.

Study characteristics
Methods	Study design: cross‐over Total study duration: 12 months Number of study centres: 1 Location: USA Study setting: outpatient Date of the study: 2015 to 2018
Participants	Number randomised: 18 (MDMA 13, placebo 5) Number analysed MDMA: 13, 12 after session 3 Placebo: 5, no dropouts Mean age: 54.9 (7.9) Age range: NR Sex: 14 females (77.8%) Ethnicity: 15 white (83%) Diagnostic criteria: Clinical Interview for DSM‐IV Axis Disorders (SCID) Inclusion criteria: age 18 years or older; diagnosed with a life‐threatening cancer or non‐dementing neurological illness that is ongoing or in remission with risk of recurrence; estimated life expectancy of at least nine months; STAI‐Trait > 45 Exclusion criteria: ongoing primary treatment for illness, such as initial chemotherapy for cancer; major medical conditions contraindicated for MDMA administration; uncontrolled hypertension; history of significant cerebrovascular or cardiovascular disease; primary or metastatic tumours in the brain, renal disease; dementing neurological disease; diabetes type I or II; history of hyponatremia or hyperthermia; weight less than 48 kg; pregnancy (or lactation); diagnosis of psychotic disorders; bipolar disorder I; dissociative identity disorder; eating disorder with active purging Prior psychedelic (MDMA) use: MDMA group 7 participants (53.9%), placebo group 3 participants (60.0%)
Interventions	MDMA group: 125 mg MDMA, + 62.5 mg MDMA 90 to 150 minutes later Active placebo group: 125 mg lactose "The design consisted of a blinded segment that included two day‐long experimental sessions (MDMA or placebo) scheduled two to four weeks apart, along with nine 60‐ to 90‐min non‐drug psychotherapy sessions; three preparing sessions for the first experimental session and three for integration after each experimental session." Concomitant medications: opiates Excluded medications: psychiatric medication, ongoing treatment for primary illness
Outcomes	Outcomes reported in the study used in the review (Types of outcome measures): BDI‐II, STAI‐S, STAI‐T, GAF, FACIT (physical well‐being, social/family well‐being, emotional well‐being, functional well‐being, additional concerns) Outcomes reported at: baseline; 1 month after the 2nd session; 1 month after 3rd MDMA session (open label)
Notes	Funding for trial: Multidisciplinary Association for Psychedelic Studies (MAPS) Possible conflicts of interest of trial authors: authors received salaries from MAPS.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Used a web‐based randomisation system with unique container numbers
Allocation concealment (selection bias)	Low risk	Quote: "Unique container numbers"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Most investigators and participants (> 80%) guessed assignment to the treatment or placebo group correctly.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Most investigators and participants (> 80%) guessed assignment to the treatment or placebo group correctly. Most efficacy outcomes were self‐reported by participants.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No dropouts at primary assessment.
Selective reporting (reporting bias)	Low risk	Primary outcome (STAI state/trait) was prespecified in the protocol (NCT02427568).
Other bias	Low risk	The study appears to be free of other sources of bias.

BDI: Beck Depression Inventory; BSL: baseline; CNS: central nervous system; DSM‐IV: 4th edition of the Diagnostic and Statistical Manual of Mental Disorder; DTS: Death Transcendence Scale; ECOG: Eastern Cooperative Oncology Group; EORTC‐QLQ: European Organization for Research and Treatment of Cancer‐Quality of Life Questionnaire; FACIT: Functional Assessment of Chronic Illness Therapy; FACIT‐Sp: Functional Assessment of Chronic Illness Therapy–Spiritual Well‐Being Scale; GAD: generalised anxiety disorder; GAF: Global Assessment of Functioning; HADS: Hospital Anxiety and Depression Scale; HAM: Hamilton Rating Scale; LAP‐R: Life Attitude Profile‐Revised; LSD: lysergic acid diethylamide; LTI: life‐threatening illness; MAO: monoamine oxidase; MDMA: 3,4‐methylenedioxy‐methamphetamine; MQOL: McGill Quality of Life Questionnaire; NA: not applicable; NR: not reported; SCID: Structured Clinical Interview for DSM‐IV Axis I Disorders; SD: standard deviation; SCID: Structured Clinical Interview for DSM disorder; SEM: standard error of the mean; SAEPT: Swiss Medical Society for Psycholytic Therapy; STAI: State‐Trait Anxiety Inventory; WHO‐Bre: World Health Organisation Qualify of Life scale, brief version

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Gasser 2015	Ineligible study design ‐ qualitative study
Grof 1973	Ineligible study design ‐ not an RCT
Pahnke 1971	Ineligible population
Rodriguez‐Mayoral 2023	Ineligible study design ‐ no psychotherapy intervention
Rosenbaum 2019	Commentary
Ross 2022	Commentary

RCT: randomised controlled trial

Characteristics of ongoing studies [ordered by study ID]

ACTRN12619001225101.

Study name	Psilocybin‐assisted psychotherapy for the treatment of depression and anxiety associated with life‐threatening illness
Methods	Randomised, double‐blind, placebo‐controlled study
Participants	40
Interventions	Group 1: psilocybin 25 mg Group 2: niacin 100 mg Psychotherapy in both groups
Outcomes	Hospital Anxiety and Depression Scale, Beck Depression Inventory, State‐Trait Anxiety Inventory
Starting date	27 September 2019
Contact information	Ms Virginia Francis, St. Vincent's Hospital, Melbourne Psychosocial Cancer Care, 41 Victoria Parade Fitzroy, 3065 VIC, Australia, virginia.francis@svha.org.au
Notes

ACTRN12623000478617.

Study name	A feasibility study of psychedelic microdosing‐assisted meaning‐centred psychotherapy in advanced stage cancer patients (PAM Trial)
Methods	Randomised, double‐blind, placebo‐controlled study
Participants	40
Interventions	Group 1: LSD‐microdose Group 2: placebo MCP in both groups
Outcomes	Feasibility: measures will include adherence to medication regimen, attendance at MCP sessions, MCP treatment fidelity, and participant recruitment and attrition rates; acceptability; safety (https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=385810)
Starting date	2 December 2024
Contact information	Dr Lisa Reynolds, l.reynolds@auckland.ac.nz
Notes

EudraCT 2020‐005037‐32.

Study name	Psilocybin ‐ a strategy of rapid antidepressant response in depression comorbid with cancer, a randomised double‐blind study with the possibility of entering open extension
Methods	Randomised double‐blind study with the possibility of entering open‐label extension
Participants	60
Interventions	Group 1: psilocybin 20 mg Group 2: ketamine 200 mg/midazolam 5mg Psychotherapy in both groups
Outcomes	Primary outcome: Montgomery–Åsberg Depression Rating Scale score between baseline and the 28th day
Starting date	22 September 2021
Contact information	Dr. Michaela Kubelová, PhD, Masaryk University, michaela.kubelova@med.muni.cz
Notes

NCT05214417.

Study name	Conscious dying/conscious living: ketamine‐assisted psychotherapy (KAP) for patients at end of life ‐ a pilot study for palliative and hospice care
Methods	Randomised, multicentre study
Participants	120
Interventions	Group 1: ketamine 100 mg Group 2: pre‐existing conventional treatment Psychotherapy in both groups
Outcomes	Primary outcomes: State‐Trait Anxiety Inventory, Death and Dying Distress Scale (DADDS)
Starting date	1 May 2022
Contact information	Philip E Wolfson, MD, ketamine.research@gmail.com
Notes

NCT05398484.

Study name	A phase 2b, randomized, double‐blind, placebo‐controlled, multi‐center study of the effects of psilocybin‐assisted psychotherapy on psychiatric and existential distress in advanced cancer
Methods	Randomised, double‐blind, placebo‐controlled, multicentre
Participants	300 Inclusion criteria Aged ≥ 21 Diagnosis of Advanced Cancer defined as: Solid tumors to include stage 3 or 4, metastatic illness, or recurrent illness Hematologic malignancies to include, but not limited to, Stage 3 or 4 non‐Hodgkins lymphoma, late‐stage multiple myeloma or second‐line therapy for multiple myeloma, and all forms of acute myeloid leukaemia Functional status defined as: Eastern Cooperative Oncology Group (ECOG) ≤ 2 and Palliative Performance Scale (PPS) ≥ 60% Clinically significant anxiety defined as SIGH‐A > 17 at screening Have an identified support person: agree to be accompanied home (or to an otherwise safe destination) by the support person, or another responsible party, following dosing Participants of childbearing potential must agree to practice an effective means of birth control throughout the duration of the study. Exclusion criteria Unstable medical conditions or serious abnormalities of complete blood count, chemistries, or ECG that in the opinion of the study physician would preclude safe participation in the trial. Risk for hypertensive crisis defined as screening, baseline, and medication session (prior to dosing) blood pressure >140/90 mmHg Significant central nervous system pathology Primary psychotic or affective psychotic disorders Family history of first‐degree relative with psychotic or serious bipolar spectrum illness High risk of adverse emotional or behavioral reaction based on investigator's clinical evaluation Active substance use disorders defined as: DSM‐5 criteria for alcohol or drug use disorder (excluding caffeine and nicotine) within the past year Extensive use of serotonergic hallucinogens (e.g. LSD, psilocybin) Clinically significant suicidality or high risk of completed suicide History of hallucinogen persisting perception disorder (HPPD) Cognitive impairment as defined by: Montreal Cognitive Assessment Test (MoCA) < 23 Concurrent medications (Note: If taking any supplement containing niacin, agrees to suspend use for at least five days prior to dosing and for the duration of the study) Have a positive urine drug test including amphetamines, barbiturates, buprenorphine, benzodiazepines, cocaine, cannabis, methamphetamine, MDMA, methadone, opiates (morphine, oxycodone), phencyclidine (PCP), and tetrahydrocannabinol (THC). Note: prescribed opiate medications (e.g., cancer‐related pain) will be allowed to continue through the study period for participants who have been on a stable dose of such medicine for at least 1 month prior to screening, as determined during review of concomitant medications. Note: prescribed benzodiazepine medications and non‐benzodiazepine sleeping medications will be allowed to continue through the study period for participants who have been on a stable dose of such a medicine for at least 6 weeks prior to screening, as determined during review of concomitant medications. Note: participants using cannabis, including legal cannabis, for any purposes must agree to refrain from use beginning at screening, as confirmed with a negative baseline drug test, and through to the end of the study. Note: Participants using prescribed psychostimulants (amphetamines and Ritalin), must agree to refrain from use two weeks prior to baseline visit, as confirmed with a negative baseline drug test, and through to the end of the study. Have a psychiatric condition judged to be incompatible with establishment of rapport with the study therapists or safe exposure to psilocybin Participants who are pregnant, as indicated by a positive urine pregnancy test at screening, baseline, or prior to dosing on medication administration sessions. Participants who intend to become pregnant during the study or who are currently nursing. Have any psychological or physical symptom, medication or other relevant finding prior to randomisation, based on the clinical judgment of the PI or relevant clinical study staff that would make a participant unsuitable for the study. Have an allergy or intolerance to any of the materials contained in either drug product Be enrolled in another clinical trial assessing intervention(s) for anxiety, depression, or existential distress (e.g. pharmacologic or psychotherapeutic interventions)
Interventions	Group 1: psilocybin 25 mg Group 2: niacin 100 mg Behavioural psychotherapy in both groups
Outcomes	Primary outcome: change in structured interview guide for the Hamilton Anxiety Scale (HAM‐A): SIGH‐A score
Starting date	4 May 2023
Contact information	Stephen Ross, MD, stephen.ross@nyulangone.org
Notes

NCT05403086.

Study name	Pragmatic trial of psilocybin therapy in palliative care (PT2PC): a multicenter triple‐blind phase 2 randomized controlled trial of psilocybin therapy for demoralized adults near the end of life
Methods	Multicentre, triple‐blind, phase 2, randomised controlled trial
Participants	100 Inclusion criteria Aged 18 years and older Has the capacity to consent to research Is currently a patient in a study‐engaged clinical site Has a life‐threatening illness and a life expectancy of ≤ 2 years Has moderate‐to‐severe demoralisation Ability to take oral medication (capsules and liquid) Exclusion criteria General Treatment with another investigational drug or intervention within 1 month of signing informed consent form If deemed by clinical judgment of the study investigators to be unsafe for undergoing the intervention Neurological Cognitive impairment sufficient to impede the ability to complete study tasks History of intracranial hemorrhage Recent embolic stroke Recent seizure Current intracranial mass Advanced stage of a neurologic disease that elevates risk for psychosis Cardiovascular Uncontrolled hypertension Clinically significant cardiac disease Respiratory Severe pulmonary disease Supplemental oxygen requirement Gastrointestinal Current intractable nausea/vomiting/diarrhoea Recent, clinically significant gastrointestinal bleed Markedly abnormal liver function tests Endocrine, renal, and reproductive Pregnancy or lactation Severe renal insufficiency Unstable insulin‐dependent diabetes mellitus Prohibited medications Antipsychotics Antidepressants (with exceptions) Dopamine agonists Drugs known to have adverse interactions with psilocybin or ketamine
Interventions	Group 1: psilocybin moderate/high dose (single dose) Group 2: ketamine low/moderate dose (single dose) 4 to 5 sessions of a brief, existential psychotherapy
Outcomes	Primary outcome: change from baseline on 16‐item Demoralization Scale ‐ II at weeks 2 and 5 Various secondary outcomes
Starting date	1 October 2023
Contact information	Charles S Grob, MD, psilocybin@lundquist.org
Notes

NCT05883540.

Study name	Lysergic acid diethylamide (LSD) in palliative care: a randomised, double‐blind, active‐placebo controlled phase II study (LPC‐study)
Methods	Randomised, double‐blind, active‐placebo controlled, parallel
Participants	60 Inclusion criteria Age ≥ 25 years End‐stage fatal disease of any cause with a life expectancy ≥ 12 weeks and ≤ 2 years Sufficient understanding of the study procedures and risks associated with the study Participants must be willing to adhere to the study procedures and sign the consent form. Participants must be willing not to drive a traffic vehicle or to operate machines within 24 h after LSD administration. Participants must complete an actual "Emergency Medical Directive". Exclusion criteria Life expectancy < 12 weeks Known hypersensitivity to LSD Requiring ongoing concomitant therapy with a psychoactive prescription drug that might interfere with the study drug, and unable or unwilling to comply with the washout period Current use of a potent drug CYP2D6 inhibitor Women who are pregnant or nursing or intend to become pregnant during the course of the study Somatic disorders including central nervous system involvement of cancer, epilepsy with a history of seizures, history of delirium, end‐stage heart failure, untreated hypertension or insufficiently treated hypertension, angina pectoris, severe liver disease or severely impaired renal function, or other that in the judgement of the investigators pose too great potential for side effects Inability to follow the procedures of the study, e.g. due to language problems, psychological disorders, dementia, etc Participation in another study with an investigational drug within the 30 days preceding and during the present study Concomitant diagnosis of past or present psychotic disorder Concomitant diagnosis of past or present bipolar disorder Substance use disorder (within the last 2 months, except nicotine, opioids used for analgesia, and benzodiazepine treatment for anxiety) Weight < 45 kg Suicidal ideation with active intent or plan to act on suicidal thoughts as assessed by the treating investigator
Interventions	Group 1: LSD 100 or 200 μg p.o. Group 2: LSD 25 μg p.o. "Patients will be allocated in a 2:1 ratio to one of the two intervention arms receiving either two moderate to high doses of LSD (100 µg and 100 µg or 100 µg and 200 µg) as intervention and two low doses of LSD (25 µg and 25 µg) as active‐placebo control."
Outcomes	Primary outcome: changes in state anxiety assessed by questionnaire (State Anxiety Inventory, STAI‐S) compared with active placebo, 2 weeks after the second session of the intervention Various secondary outcomes
Starting date	September 2023
Contact information	Yasmin Schmid, MD, yasmin.schmid@usb.ch
Notes

im: intramuscular; LSD: lysergic acid diethylamide; MCP: meaning‐centred psychotherapy; po: per os

Differences between protocol and review

Changes to title

We changed the title of the review to meet the new inclusion criteria from "Psychedelic/entactogen‐assisted therapy for treatment of anxiety, depression and existential distress in adult palliative care" to "'Psychedelic‐assisted therapy for treating anxiety, depression, and existential distress in people with life‐threatening diseases". The reference to entactogens was omitted because these are a subset of psychedelics. The change from palliative care is explained below.

Changes to inclusion criteria

We had planned to include participants with advanced (incurable) disease and stated in the protocol that we would include studies when more than 75% of participants met this criterion. Only four of our included studies would have met this criterion (Gasser 2014; Grob 2011; Wolfson 2020). Post hoc, we decided to change the inclusion criteria to include studies of "participants facing life‐threatening disease"; our rationale was that as the threat to life, as opposed to the stage of cancer, was the main reason for the psychological condition (anxiety, depression, existential distress), excluding studies that did not meet the 'above 75%' criterion would diminish the meaningfulness of this review. We modified the Background where we had previously mentioned palliative care to reflect the wider range of participant conditions that we had now decided to include.

Changes to methods

Carry‐over effects of cross‐over trials

We deleted the following sentence that we had stated in the protocol: "We will include properly reported cross‐over trials (i.e. analysed with paired t‐test and without a carry‐over or period effect) in meta‐analyses using the generic inverse variance (GIV) method (Deeks 2023)". We also deleted the part of this sentence that related to the second period: "Since psychological changes through psychedelic‐assisted therapy may be long‐lasting and sometimes permanent, we will consider the first period of crossover trials for our analyses and carefully check for carry‐over effects to include the second period" (the part referring to the second period has been deleted). These changes were made because the carry‐over effects due to the long‐lasting and sometimes permanent changes caused by psychedelic‐assisted therapy making cross‐over trials unsuitable, and therefore for meta‐analyses, we used first‐period data only.

Methods explained more fully

• Primary outcomes

We added: "If multiple time points were presented within the specified range (< 12 weeks, > 3 months), we used the latest time point because this seemed to be the most relevant considering our interest in the long‐term efficacy of treatments.". In the protocol, we had not foreseen that "< 12 weeks, > 3 months" would include multiple time points, therefore clarification of the included time points was necessary.

• Data extraction and management

We used the online software Plotdigitizer 2023 to extract data for meta‐analyses from published graphs when authors did not provide data.

• Risk of bias

We added a paragraph describing how we assessed 'other risk of bias', a risk of bias domain that we had not mentioned in the protocol.

• Measures of treatment effect

We explained why we used change scores for mean differences of continuous data as this had not been mentioned in the protocol. We also added a statement about how we dealt with dichotomous data.

• Unit of analysis

We explained how we would analyse cluster‐randomised trials.

Methods not implemented

• Sensitivity analyses

We did not conduct sensitivity analyses with studies at low risk of bias because we judged all studies to be at high risk of bias.

Contributions of authors

• Sivan Schipper (SS): screening the search, writing the manuscript.

• Kabir Nigam (KN): screening the search, data extraction, risk of bias assessment, writing the manuscript.

• Yasmin Schmid (YS): discussing results, writing the manuscript.

• Vanessa Piechotta (VP): risk of bias and GRADE assessments, writing the manuscript.

• Michael Ljuslin (ML): discussing results, writing the manuscript.

• Yvan Beaussant (YB): discussing results, writing the manuscript.

• Guido Schwarzer (GS): conducting meta‐analyses, writing the manuscript.

• Christopher Boehlke (CB): screening the search, data extraction, risk of bias and GRADE assessment, conducting meta‐analyses, conceiving, co‐ordinating, and writing the manuscript.

Sources of support

Internal sources

• None, Other

  No internal sources of support

External sources

• National Institute for Health and Care Research (NIHR), UK

  Cochrane Infrastructure funding to the Cochrane Pain, Palliative and Supportive Care Review Group (PaPaS)

Declarations of interest

• Sivan Schipper (SS): none known. SS is a specialist palliative care physician and manages palliative care patients. SS is a member of SÄPT (Swiss Medical Society for Psycholytic Therapies).

• Kabir Nigam (KN): is a specialist in psychiatry and manages palliative care patients. KN performed contract‐based consulting for Acuta Capital Partners, LLC, which was discontinued.

• Yasmin Schmid (YS): YS is a specialist in clinical pharmacology and toxicology. YS received a research grant from the Swiss National Science Foundation.

• Vanessa Piechotta (VP): none known.

• Michael Ljuslin (ML): ML is a specialist in internal medicine and palliative care and manages palliative care patients.

• Yvan Beaussant (YB): is a specialist palliative care physician and manages palliative care patients. YB received research grants from Sunstone Therapies LLC and Cy Biopharma, received consulting fees from Reunion Neuroscience USA Inc., and was a member of the advisory board of Reset Pharma, not receiving any payment.

• Guido Schwarzer (GS): received personal consulting fees for 'External statistical consultant of Roche Pharma AG, Grenzach‐Wyhlen, Germany'.

• Christopher Boehlke (CB): none known. CB is a specialist palliative care physician and manages palliative care patients. CB is a member of SÄPT (Swiss Medical Society for Psycholytic Therapies).

New