Neuroscientist and entrepreneur
Expert Review of Clinical Pharmacology / 2021

Letter to the editor regarding: blinding and expectancy confounds in psychedelic randomized controlled trials

Dear Editors,

Unblinding is an important confounding in pharmacological research potentially affecting rigor and objectivity of the scientific data underlying evidence-based medicine. Therefore, I want to commend the work by Muthukumaraswamy et al. [1] reporting on this problem in clinical trials with psychedelics (Ketamine, LSD, psilocybin and ayahuasca), which are rapidly becoming a focus of increased attention and hope, but also concern.

The authors presented an in-depth and carefully argued overview of unblinding (also called de-blinding, unmasking or de-masking) including promising alternatives and careful procedures to improve research practices. The two blinding indexes, the distinction between malicious and benign unmasking and the nine different study designs, four of which they do not recommend, are welcome. These were complemented by straightforward statistical estimates of average treatment effect avoiding inflation of effect-sizes because of unblinding, considering four inducers of healing: active treatment, placebo effect, therapeutic alliance and spontaneous or ‘natural’ healing. Furthermore, the two psychometric measures to evaluate expectancy can add important information at low cost. All of these expert suggestions should be noted by researchers not only when designing studies with psychedelics, because unblinding and underreporting became pervasive in pharmacology [2,3]. This can bias large scale pharmaceutical practices with consequences for regulatory approvals, potentially affecting millions of patients.

However, we must critically examine if the proposed research program is pragmatic, if it can improve standards of care and if it can solve conceptual challenges in psychedelic research and psychopharmacology in general. First, two of the suggested designs involve more than two conditions, thus potentially increasing the number of patients necessary for adequate statistical power. Unfortunately, sample size estimates were not presented. Yet, a larger number of patients may increase costs and given these are not currently demanded by regulatory authorities, it is unlikely that the pharmaceutical industry would invest in such designs, while the academic sector may face financial obstacles for implementation. To complicate matters further, ethical considerations may hinder these proposals, not limited to the issue of patient deception, rightly pointed out by the authors. There are more serious concerns, as recently happened in the first Phase 3 trial with MDMA-assisted therapy for the treatment of PTSD: the only three serious adverse events occurred in the placebo group, with one patient having two episodes of suicidal behavior and another having one episode of suicidal ideation leading to self hospitalization [4]. Regarding active comparators, evidence from MDMA research, unfortunately missing in the paper, is useful: low dose MDMA can be stressful and trying for patients, propitiating drop-outs [5], thus these designs may violate clinical equipoise. Therefore, the proposed research program may not be pragmatically and ethically implemented. Second, it is dubious if the suggested practices will benefit patients. While research aims to isolate pharmacological effects from other variables in order to regulate and authorize drug marketing, in clinical practice patients always expect something out of the treatments they search for, and therapeutic alliance and placebo effects are always part of the process. Finally, we need to examine if it actually makes sense to try to disentangle the contributions of these inductors of healing as relatively independent variables, or if they are so intertwined as to make the goal of such research unachievable in principle.

The question as to whether psychopharmacology can be reduced to objective and quantitative measures alone bears on epistemic considerations regarding what is psychiatry, what are mental disorders and mental health. The etymology of psychopharmaceuticals comes from psȳkhē, from the Greek ‘breath, life, or soul,’ similarly included in the 1956 ‘psychedelic’ (‘mind manifesting’) neologism by psychiatrist Humphry Osmond. Then is it logical to consider that people should not become consciously aware of changes in their feelings, emotions, sensations and thoughts when the treatment is exactly aiming at people’s feelings, emotions and thoughts? Is it coherent to strive for pure objectivity if the symptoms of the disorders and the effects of the drugs are mostly subjectively felt? If we expect to find drugs which minimize or eliminate chronic symptoms including intense fear, anxiety or sadness, how could patients’ feelings change without them becoming consciously aware of it? And if they do not perceive it, thus preserving masking, wouldn’t it actually mean that the drug did also not exert any useful effect? In other words, if a drug has therapeutic effect in psychiatry, would not this most often strongly correlate with unmasking, and if masking is maintained, could it be indicative of minor subtle changes or no subjective changes whatsoever, and therefore the drug has limited utility?

These are definitely not novel questions. But psychiatry is not a discipline from the so-called ‘natural sciences.’ Rather, it is situated between biology and the social sciences, as all psychiatric disorders involve active relationships with significant others, family members, community values and cultural norms, which cannot be reduced to the brain only. Actually, the brain is an open complex system constantly in development and changing according to environmental interactions. So whatever drugs can do, ultimately it is about how patients actively engage with their feelings during the drug’s effects and also while constantly and actively interacting in society. This is a unique conceptual situation, even if mental distress and social issues are secondary concerns in all fields of medicine. In psychiatry subjectivities are the primary concern, and not recognizing it is a blind spot highlighting the need for a specific epistemology of psychiatry [6]. The challenge then is how to avoid excessive reductionism aiming at the best mental health-care possible, for the benefit of patients. A critical examination of the double-blind method reveals it does not adequately fit an epistemology of psychiatry, with far-reaching consequences: psychiatric drug development, mostly based on double-blind randomized clinical trials (RCT), resulted in the highest rate of post-marketing safety events among 222 drugs approved in the USA by the FDA between 2001 and 2010, staining 80% of approved drugs for psychiatric disorders more than 10 years after approval. That is twice the rate in neurology and almost quadruple than in cancer or cardiovascular diseases [7]. And for some psychiatric medications, especially serotonergic antidepressants, statistical calculations regarding efficacy has been a very contentious topic eg [8].

As Muthukumaraswamy et al. rightly reference, the standardization of the double-blind RCT as the gold-standard pharmaceutical practice and regulatory demand occurred concomitantly with the first wave of psychiatric research with psychedelics in the 1950’s and 60’s. These two fields were not, however, “parallel histories’’. Some of the very pioneers in psychedelic research, Osmond and Hoffer, discussed limitations of the new blinding methodology in 1961 [9] and also conducted one of the first blind trials in psychiatry [10]. These convergences matter because the difficulty to blind psychedelic trials was the main reason for distrust regarding clinical results obtained with LSD in alcoholics. At the time, the role of compassionate psychotherapeutic or psychoanalytically oriented approaches were criticized as not objective enough, and some patients were abusively restrained in the name of ‘scientific objectivity’ [9]. Thus it was the epistemic debate which ultimately led psychedelics to be prematurely discarded from psychiatry [9,11]. Furthermore, inspired by psychedelics, Osmond and Hoffer elaborated one of the earliest hypotheses centering on the role of neurochemicals in the etiology and possible treatment of psychiatric illnesses, during the early days of biological psychiatry [9,10]. What then were unconventional ideas in a predominantly psychodynamic field became the prevailing biological psychiatry messaging half a century later, which propagates a misleading narrative that “mental disorders are brain diseases cured by scientifically designed medications’’, which may be detrimental to patients [12].

With the approval of esketamine for the treatment of depression in the USA and Brazil, and MDMA and psilocybin having been granted ‘breakthrough therapy status’ from the FDA for PTSD and depression, the discussion of how to build an appropriate database of rigorous scientific evidence to inform drug policy is necessary, including to avoid repeating the mistake of the 1960’s. Here it is worth emphasizing that placing RCTs at the top of a rigid hierarchy may be misguided, because instead of constituting an utmost level of unbiased and indispensable scientific proof, RCTs are relevant sources of data which must be better complemented by other sources of information, each with its own limitations. This is not specific to psychedelics, with debates about RCT limitations and relevance for cannabinoid medicines ongoing. Furthermore, dozens of currently approved medications have not been developed through the double-blind research pipeline [13].

Interpreting blinded RCT data as the ultimate scientific information about psychedelics can configure epistemic injustices which drive loss of important information [14]. Considering ancient and contemporary medicinal uses of psychedelics by many populations across the planet, with their own epistemologies and ontologies, it is time to reflect on the goal of clinical research and ponder if a purely objective approach is desirable and achievable. Beyond stand-alone drug treatments, what may be welcome is a paradigmatic shift in psychiatric research and development [15], carefully considering the importance of patients actively engaging with contents of non-ordinary states of consciousness through deep therapeutic alliance constituting benign unmasking.


1. Muthukumaraswamy S, Forsyth A, Lumley T. Blinding and expectancy confounds in psychedelic randomised controlled trials. Expert Rev Clin Pharmacol. 2021. DOI:10.1080/17512433.2021. 1933434 2 E. E. SCHENBERG

•• Expert review on how unblinding in psychedelic clinical trials biases estimation of average treatment effects.

2.Bello S, Moustgaard H, Hróbjartsson A. Unreported formal assessment of unblinding occurred in 4 of 10 randomized clinical trials, unreported loss of blinding in 1 of 10 trials. J Clin Epidemiol. 2017;81:42–50.

3. Bello S, Moustgaard H, Hróbjartsson A. The risk of unblinding was infrequently and incompletely reported in 300 randomized clinical trial publications. J Clin Epidemiol. 2014;67(10):1059–1069.

4. Mitchell JM, Bogenschutz M, Lilienstein A, et al. MDMA-assisted therapy for severe PTSD: a randomized, double-blind, placebo-controlled phase 3 study. Nat Med. 2021;27(6):1025–1033.

5. Oehen P, Traber R, Widmer V, et al. A randomized, controlled pilot study of MDMA (±3,4-Methylenedioxymethamphetamine)-assisted psychotherapy for treatment of resistant, chronic Post-Traumatic Stress Disorder (PTSD). J Psychopharmacol. 2013;27(1):40–52.

6. Berrios, G., & Marková, I. Toward a new epistemology of psychiatry. In: Kirmayer L, Lemelson R, Cummings C, editors. Re-visioning psychiatry: cultural phenomenology, critical neuroscience, and global mental health. Cambridge: Cambridge University Press; 2015. p. 41-64. DOI:10.1017/CBO9781139424745.005.

7. Downing NS, Shah ND, Aminawung JA, et al. Postmarket safety events among novel therapeutics approved by the US food and drug administration between 2001 and 2010. JAMA. 2017;317 (18):1854–1863.

8. Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. Lancet. 2018;391 (10128):1357–1366.

9. Dyck E. Psychedelic Psychiatry : LSD from Clinic to Campus / Erika Dyck. Baltimore, Md: Johns Hopkins University Press; 2008. Print.

10.Mills JA. Hallucinogens as hard science: the adrenochrome hypothesis for the biogenesis of schizophrenia. Hist Psychol. 2010;13 (2):178–195.

11. Oram M. Efficacy and enlightenment: LSD psychotherapy and the drug amendments of 1962. J Hist Med Allied Sci. 2014;69 (2):221–250.

12.Dumas-Mallet E, Gonon F. Messaging in biological psychiatry: misrepresentations, their causes, and potential consequences. Harv Rev Psychiatry. 2020;28(6):395–403.

13. Nutt D, Bazire S, Phillips LD, et al. So near yet so far: why won’t the UK prescribe medical cannabis? BMJ Open. 2020;10(9): e038687.

14. Schenberg EE, Gerber K. Overcoming epistemic injustices in the biomedical study of Ayahuasca: towards ethical and sustainable regulation. Transcult Psychiatry. 2021 (in press).

15. Schenberg EE. Psychedelic-assisted psychotherapy: a paradigm shift in psychiatric research and development. Front Pharmacol. 2018;9:733.

“This is an Accepted Manuscript of an article published by Taylor & Francis in Expert Review of Clinical Pharmacology on July 6th, 2021, available at http://wwww.tandfonline. com/10.1080/17512433.2021.1951473.”
Who is blind in psychedelic research?

Eduardo Schenberg @ 2017 - All rights reserved.

Developed by Thiago Saccól.