Psychedelic Trials Reveal Gaps Between Controlled Efficacy and Real-World Readiness

Narrow inclusion criteria may limit how well psychedelic trial results translate to patients with the comorbidities seen in routine practice, according to Brian Barnett, MD, of Cleveland Clinic. Barnett noted that functional unblinding had been reported with benzodiazepines, stimulants, antidepressants, and antipsychotics, and that in the only trial to measure expectancy directly, which compared psilocybin with escitalopram, expectancy predicted response to escitalopram but not psilocybin.¹ In multidose LSD trials for generalized anxiety disorder, efficacy did not track with the rising functional unblinding seen at higher doses.² He observed that preparation predicted outcomes while integration did not, and that classic psychedelics may need repeating every 3-6 months compared with every 1-4 weeks for ketamine and esketamine. Barnett advised substantial caution for patients with a personal or first-degree family history of psychosis or mania and described readiness in training, facilities, and reimbursement as uneven should a classic psychedelic gain US Food and Drug Administration approval.

Psychiatric Times: You've been a principal investigator on trials of psilocybin, LSD, and ketamine. What has running these studies taught you that clinicians reading only the published results might miss?

Brian Barnett, MD: It is important to look closely at a trial’s inclusion and exclusion criteria. Many trials enroll a narrow subset of patients with the condition being studied, excluding people with psychiatric or medical comorbidities that are common in everyday practice. Trials designed to reflect routine clinical care are an exception, but they remain relatively uncommon in psychedelic research. We should therefore be cautious about assuming that the benefits and safety profile observed in a carefully selected trial population will fully translate to the broader, more complex population we treat in our clinics every day.

PT: Blinding is a persistent problem in psychedelic research because participants can usually tell whether they received the active drug. How do you and other investigators address functional unblinding and expectancy effects, and how should clinicians factor this in when they interpret trial results?

Barnett: While there has been a lot of focus on functional unblinding in psychedelic trials, it should be emphasized that this problem is not unique to psychedelics. High rates of functional unblinding have been reported for all the strongly psychoactive drugs we use in psychiatry, including benzodiazepines and stimulants, as well as drugs that are less psychoactive such as antidepressants and antipsychotics. Based on our experiences with those drugs in clinical practice, there is little doubt that they are effective treatments for many patients, so the results seen in clinical trials cannot be written off as simply patients being aware they got active drug instead of placebo. Therefore, we shouldn’t overemphasize the functional unblinding issue with psychedelics and doubt the therapeutic efficacy they are showing in clinical trials. Regarding expectancy (eg, how strongly a patient expects a treatment will help their condition), it has been argued by some that we should doubt the results of psychedelic trials because functional unblinding may increase expectancy in those receiving active drug in the trials. However, the one trial that has actually measured this, which compared psilocybin versus the SSRI escitalopram, found expectancy did not predict antidepressant efficacy of psilocybin, though it did predict the efficacy of escitalopram. We already knew that expectancy predicts therapeutic efficacy of SSRIs, as well as transcranial magnetic stimulation, another important depression treatment. Therefore, the reality of our field based on the current state of knowledge is actually the opposite of what some critics are claiming- expectancy does not appear to be related to outcomes for psilocybin, while it is related to current standard of care treatments.

With that said, I think psychedelics have made psychiatry more aware of the issue of functional unblinding and expectancy made us think harder about how to address it in clinical trial methodologies. Now the expectation is that these issues will be measured and more routinely included in analyses of clinical trial results to help us untangle these possible relationships. Psychedelic trials are also employing multiple doses to help determine whether being aware that you got a higher psychoactive dose rather than a lower non-psychoactive dose predicts outcomes. For example, multiple trials of LSD for generalized anxiety disorder have shown that that as functional unblinding increases with higher doses of LSD, this does not track with efficacy of each dose, reassuring us that the observed efficacy is inherent to the drug (once a certain dosage threshold has been crossed), rather than solely being explained by unblinding.

PT: How much of the therapeutic effect do you attribute to the drug itself and how much to the psychological support around it? What does the evidence suggest about how much preparation and integration therapy is actually needed?

Barnett: Many industry sponsored trials now include no psychotherapy and just enough support to ensure patients are prepared for the psychedelic experience and can have that experience safely in a monitored environment. These trials are still showing strong efficacy signals, so this indicates the effects of the drugs themselves are primarily driving the results we see. We have not yet had a trial of psychedelic with psychotherapy versus a psychedelic with minimal psychological support. Having trials that assess that will be essential to see what psychotherapy adds to efficacy and safety. Typically in psychiatry, combining psychotherapy with pharmacological treatments creates synergistic benefits not found in either treatment alone. I would be very surprised if adding psychotherapy to psychedelics does not improve outcomes. However, it’s an important empirical question the field needs to address, and I expect we will have an answer in the next few years. A recent development of interest is the finding that the amount of preparation patients receive prior to their psychedelic experience predicts outcomes, while the amount of integration (eg, where patients meet with a provider to make sense of their psychedelic experience and determine how to translate it into durable behavioral changes) does not. This needs to be replicated in many more trials, but it suggests that we need to expend most of our efforts in making patients feel safe and comfortable when preparing for their psychedelic experience to ensure they get the most therapeutic benefit out of it. Finally, all clinical trials do include more support overall than patients would typically get in clinical practice. As a result, this is another reason why results in clinical practice are never as strong as they are in clinical trials, so that should not be overlooked.

PT: Psilocybin and LSD work mainly through serotonin 2A receptor agonism, while ketamine acts on the glutamate system. In practice, do you see these as distinct treatments suited to different patients, or as variations on a shared mechanism of rapid neuroplasticity?

Barnett: While classic psychedelics like psilocybin and LSD act pharmacologically differently that non-classic psychedelics like esketamine and ketamine, they appear to overlap in their psychoactive effects and in downstream pharmacological consequences in the brain. All these drugs increase neuroplasticity. Neuroplasticity is the brain’s ability to change how its cells connect and communicate in response to experience. In depression, the hope is that enhancing this capacity could make it easier to move away from entrenched patterns of thinking and behavior. An important distinction between these drugs is that classic psychedelics appear to create a window of neuroplasticity post treatment that is longer than that of ketamine, which may explain why the therapeutic effects of psychedelics appear to last longer. With that said, if classic psychedelics become FDA approved, they will likely need to be recurrent treatments for patients who respond. However, they appear to only be needed every 3 to 6 months depending on the particular condition being treated, as opposed to esketamine and ketamine, which need to be repeated every 1 to 4 weeks.

We still do not know whether it would be helpful for patients to be treated with both kinds of psychedelics as opposed to one or the other. Without head-to-head trials we also don’t know whether classic versus non-classic psychedelics will work better for a particular condition or patient. I suspect those trials will occur in the next few years. Based on what we know now though, the benefits of classic psychedelics appear to last longer than esketamine and ketamine. However, esketamine and ketamine still offer some benefits since they can essentially be combined with all psychiatric oral medications, while patients treated with classic psychedelics will need to be tapered off many drugs such as antipsychotics and lithium prior to treatment, which creates challenges for patients and providers. Some patients, particularly those with severe depression, may not be able to safely come off these medications to allow for treatment with classic psychedelics, so esketamine and ketamine may be the only pharmacologic interventional psychiatry treatment option for their conditions.

PT: Given your training in addiction psychiatry, how do you weigh the misuse potential of psychedelics against their therapeutic promise? Are there patients you would steer away from these treatments, such as those with a personal or family history of psychosis or bipolar disorder?

Barnett: I would not automatically exclude someone simply because they have a history of addiction. I would assess the specific substance and their current stability. There are also increasing data showing that, when used in strongly supported settings, both ketamine and classic psychedelic can treat some types of addictions.

Classic psychedelics such as psilocybin and LSD have a quite low addiction liability, but that does not mean they are harmless. People most at risk for developing compulsive use are typically people who misuse multiple types of substances, indicating a desire to escape their normal mental status using any pharmacological means available. I would not recommend these patients for treatment with psychedelics until we have clinical trial data in this population with long term follow up. Ketamine is different: misuse and addiction are meaningful concerns and we are seeing ketamine use increase significantly within the US. The addictive potential is something I am sure to discuss during informed consent discussions with patients we are treating with ketamine or esketamine. However, when these drugs are administered in a supervised healthcare setting, the proportion of patients going onto misuse them outside the clinical is quite small, which I think speaks to the importance of appropriate screening, informed consent and thorough medical monitoring. While all these drugs have misuse potential, it is manageable and should not deter us from taking full advantage of their therapeutic potential within psychiatry.

For classic psychedelics, I would generally steer patients with a personal history of psychosis or mania away from treatment outside a carefully designed research study. We know these drugs can induce psychotic and manic episodes in some people with psychotic conditions and bipolar disorder, though we need more clinical trial data in these particular populations. Bipolar II disorder, as opposed to Bipolar I disorder, deserves some nuance. A small, closely monitored open-label study produced encouraging findings without an observed increase in mania or hypomania, but 15 participants cannot establish safety for routine practice. Those findings also should not be generalized to bipolar I disorder. We also have growing data that people with personality disorders are more likely to experience adverse psychological effects from psychedelics, so that is another group of patients I would not recommend psychedelic treatments for at this time.

A first-degree family history of psychosis or bipolar disorder would also warrant substantial caution. The exact risk is difficult to quantify because these patients have often been excluded from trials. However, naturalistic use data suggest that psychedelics can precipitate first episode psychosis or mania in individuals with these family histories, likely due to genetic predispositions that are interacting with the psychedelic’s pharmacological effects.

PT: Several companies are developing non-hallucinogenic analogs and shorter-acting compounds meant to deliver the benefit without a long, intense session. Do you think the subjective experience is necessary for the therapeutic effect?

Barnett: Shorter-acting psychedelics and non-hallucinogenic psychedelics analogs are different propositions. Shorter acting psychedelics tend to be intensely psychoactive. Reducing duration could make psychedelic treatment more practical and scalable since we could use the same treatment room to treat multiple patients each day instead of one patient with a longer acting psychedelic like psilocybin or LSD.

We don’t know yet whether the psychedelic experience is essential for therapeutic benefit, and I think it is important to remain open to either answer. A meaningful psychedelic experience could contribute to clinical improvement through changes in perspective or behavior. However, an association between the intensity of an experience and subsequent improvement does not establish that the experience caused the benefit. It might instead be that the psychedelic experience is a marker of how strongly the drug is affecting the brain, with the therapeutic benefit arising partly or entirely from other effects of the drug. Both could also contribute. We need studies designed to separate those possibilities before we can say whether the subjective experience is necessary. Animal studies suggest that some plasticity-promoting and potential therapeutic effects can occur with non-hallucinogenic analogs of psychedelics. That is an important proof of concept, but it does not establish equivalent antidepressant benefit in humans. Clinical trials are underway to see if these animal findings translate into humans. If we could develop treatments from psychedelics that lack hallucinogenic effects and provide comparable therapeutic benefit that would be a major advance for psychiatry given how logistically difficult it is to deliver psychedelic treatments.

PT: From a forensic psychiatry perspective, what informed consent, liability, and documentation issues should clinicians consider as state-level legalization expands and patients show up having used psychedelics outside a clinical setting?

Barnett: The first distinction is between state authorization and federal approval or legality. Those are separate questions. Oregon, for example, has a regulated psilocybin services system while acknowledging that psilocybin remains federally classified as a Schedule I substance. Clinicians need to understand the particular jurisdiction and its licensure requirements for administering state approved psychedelics, whether their licensing board permits participation in these programs, and whether their malpractice coverage would apply, rather than assuming that they can administer these treatments legally and in a manner covered by their malpractice insurance. Of note, a medical license alone is insufficient to administer psilocybin in either Oregon or Colorado’s programs.

For anyone delivering these treatments, I would want informed consent to address uncertain benefit for patients, alternatives, the possibility of acute distress during treatment, possible persistent psychiatric symptoms after treatment, physical risks (particularly cardiac risks), and the follow-up plan. Consent should occur before intoxication and explicitly address the use of therapeutic touch during sessions (e.g. where the provider can hold the patient’s hand or place a hand on their shoulder if the patient is distressed), recording the session, confidentiality, and professional boundaries.

When a patient comes in after using psychedelics elsewhere, my first responsibility is to assess and treat them without judgment. I would document what they believe they took and how much, when they took it, other substances and medications consumed around administration time, whether they found the psychedelic helpful or not, and any adverse effects including mania, psychosis, and suicidality. I would also be cautious in treating memories or interpretations arising during a psychedelic experience as established facts, since psychedelics can induce false memories. You can take a patient’s distress seriously without affirming an unverified explanation. From a forensic perspective, that distinction is particularly important.

PT: If psilocybin or another classic psychedelic gains FDA approval, what will general psychiatrists need to know, and how prepared is the field in terms of training, facilities, and reimbursement models to deliver these treatments safely?

Barnett: General psychiatrists will need to understand who was actually enrolled in trials conducted for FDA approval, the approved indication, contraindications and precautions, medication interactions, expected benefits, and the limits of the durability data. They will also need to recognize potential complications of psychedelic treatments and know when to refer patient for higher level of psychiatric care after treatment. They will also need to know how to counsel patients on whether a psychedelic treatment might be helpful or harmful for them based on their psychiatric and medical conditions. I would want general psychiatrists to approach these treatments as another potential tool within psychiatry’s armamentarium. Even when someone improves substantially, they still need a plan for residual symptoms, recurrence of symptoms, and what happens next if the treatment does not work.

My assessment is that readiness with our healthcare system is uneven at best. Specialty interventional psychiatry programs have relevant experience, including lessons from ketamine and esketamine, but delivering a prolonged psychedelic session creates additional staffing and facility demands. Training will be needed by everyone involved in the administration of psychedelics and should cover appropriate patient selection, management of acute distress, professional boundaries, emergency procedures, and follow-up in addition to the pharmacology of these drugs. Adequate reimbursement for treatment delivery will be crucial. A psychedelic treatment course may involve relatively few doses while still requiring substantial clinician and administrative staff time and space. We will need payment models that support the necessary care around optimal administration, with the details guided by the eventual product label and any safety monitoring program required by the FDA. There is much work to be done to get these treatments to as many patients who could benefit from them as possible should they become FDA approved. Setting up the necessary infrastructure and getting enough clinicians trained for treatment delivery will almost certainly take years if not a decade or longer.

Dr Barnett is a psychiatrist certified in addiction and forensic psychiatry, practicing at the Cleveland Clinic.

References

1. Szigeti B, Weiss B, Rosas FE, et al. Assessing expectancy and suggestibility in a trial of escitalopram v. psilocybin for depression. Psychol Med. 2024;54(8):1717-1724.

2. Robison R, Barrow R, Conant C, et al. Single treatment with MM120 (lysergide) in generalized anxiety disorder: a randomized clinical trial. JAMA. 2025;334(15):1358-1372.