Blinding and expectancy confounds in psychedelic randomized controlled trials

ABSTRACT

Introduction: There is increasing interest in the potential for psychedelic drugs such as psilocybin, LSD and ketamine to treat several mental health disorders, with a growing number of randomized controlled trials (RCTs) being conducted to investigate the therapeutic effectiveness of psychedelics.

Areas covered: We review previous literature on expectancy effects and blinding in the context of psychedelic RCTs – literature which strongly suggest that psychedelic RCTs might be confounded by de-blinding and expectancy. We conduct systematic reviews of psychedelic RCTs using Medline, PsychInfo and EMBASE (Jan 1990 – Nov 2020) and show that currently reported psychedelic RCTs have generally not reported pre-trial expectancy, nor the success of blinding procedures.

Expert opinion: While psychedelic RCTs have generally shown promising results, with large effect sizes reported, we argue that treatment effect sizes in psychedelic RCTs are likely over-estimated due to de-blinding of participants and high levels of response expectancy. We suggest that psychedelic RCTs should routinely measure de-blinding and expectancy. Careful attention should be paid to clinical trial design and the instructions given to participants to allow these confounds to be reduced, estimated and removed from effect size estimates. We urge caution in interpreting effect size estimates from extant psychedelic RCTs.

KEYWORDS:

• Psychedelics
• randomized controlled trials
• LSD
• ketamine
• psilocybin
• causation
• blinding
• masking
• placebo effect

Article highlights

• Increasing numbers of RCTs for psychedelic drugs such as ketamine, LSD and psilocybin are being reported for a variety of disorders including depression and anxiety.

• Psychedelic RCTs are likely confounded by de-blinding, which when combined with expectancy effects cause over-estimation of effect sizes.

• Extant psychedelic RCTs have generally not reported de-blinding and pre-trial expectancy.

• We suggest trial designs and statistical approaches that might help to ameliorate these confounds.

Declaration of interest

S Muthukumaraswamy has received research funding from F Hoffman La Roche Ltd and performed consulting work for Eleusis Health Solutions Ltd. Other author(s) have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Notes

1. In this paper we use a broad definition of psychedelics to include not only serotonergic psychedelics but non-serotonergic psychedelics such as ketamine. There is a long history of describing ketamine as a psychedelic (see as examples 1.Grinspoon L. Psychedelic drugs reconsidered/Lester Grinspoon & James B. Bakalar. New York: Basic Books; 1979. (Bakalar JB, editor. https://nla.gov.au/nla.cat-vn1097578)., 2.Grof S. My Ketamine Journeys, or Ketamine and the Enchantment of Other Worlds. In: Wolfson P, Hartelius G, editors. The Ketamine Papers: Science, Therapy, and Transformation: Multidisciplinary Association for Psychedelic Studies; 2016, 3.Krupitsky EM, Grinenko AY. Ketamine psychedelic therapy (KPT): a review of the results of ten years of research. J Psychoactive Drugs. 1997 Apr-Jun;29(2):165-83, 4.Vollenweider FX, Kometer M. The neurobiology of psychedelic drugs: implications for the treatment of mood disorders. Nat Rev Neurosci. 2010 Sep;11(9):642-51.)

2. From here we take the modern terminology of referring to blinding as masking. This terminology is being more widely adopted to distinguish masking from studies in the visually impaired. Further, although we refer to double-masking, this terminology is best avoided in detailed trial reporting. Following CONSORT guidelines, the specific groups being masked should be specified, for example, participants and outcome assessors.

3. It is inappropriate to apply regression approaches such as ANCOVA (or similar) to regress out blinding/confidence status from treatment outcome when there are between-group differences in the covariate. As such, some degree of balance must be achieved through experimental design 129.Miller GA, Chapman JP. Misunderstanding analysis of covariance. J Abnorm Psychol. 2001 Feb;110(1):40-8…

Additional information

Funding

This work was supported by the Health Research Council of New Zealand (18/193)