Reply to ‘Methodological concerns and interpretative limitations of assessing subjective effects of ketamine’

This response addresses methodological concerns raised regarding our meta-analysis of subjective effects and therapeutic outcomes following ketamine and psilocybin treatment. We discuss the potential influence of psychotherapy and differences in subjective-effect questionnaires across studies, acknowledging these as important limitations. We argue that these factors may partly explain observed treatment differences and highlight the need for future studies using standardized, multidimensional measures of subjective experience.

We appreciate the opportunity to respond to the Matters Arising article by Ramunas Janavicius, titled “Methodological Concerns and Interpretative Limitations of the Study”, and welcome the chance to offer further context and clarification in response to the points raised. In our study, we performed a meta-analysis and systematic review on the correlation between subjective effects (including dissociation) and therapeutic outcomes following ketamine or psilocybin treatment in patients with depression or substance use disorder (SUD)1. The main outcome of our exploratory and conceptual study was the modest role for subjective effects in mediating therapeutic outcomes, with R2-values ranging from 7 to 10% for ketamine and 24% for psilocybin.

We will address two main points raised by Janavicius: the use of psychotherapy in some studies, and the use of different questionnaires to quantify the subjective effects from ketamine and psilocybin. Janavicius notes that our analyses included studies in which pharmacological treatment was administered with or without adjunctive psychotherapy, and more specifically that the ketamine studies largely did not involve psychotherapy, whereas some form of psychological support or psychotherapy was standard practice in the psilocybin studies. He further describes that psychotherapy may act synergistically with the drug, potentially leading to stronger and more durable antidepressant outcomes.

We agree that this heterogeneity in study design constitutes an important methodological consideration, which we acknowledged and discussed in our original paper1. Importantly, if psychotherapy does indeed amplify subjective-experience-mediated effects, the asymmetric inclusion of psychotherapy across treatment arms, more common in psilocybin than in ketamine studies, would be expected to inflate the apparent mediation effect for psilocybin and deflate it for ketamine. Our finding of a modest role for subjective effects despite this design asymmetry therefore likely represents a conservative estimate, particularly for ketamine.

We included both study designs (without and with psychotherapy) because the total number of studies that report subjective effects was relatively small, and, as noted in our original paper, we approached the outcomes of our analysis as hypothesis generating and more exploratory than conclusory. Future studies should directly compare pharmacotherapy-only or minimally supported models with psychedelic-assisted psychotherapy, to determine whether the inclusion of psychotherapy amplifies, facilitates, or qualitatively alters the extent to which subjective experiences mediate clinical improvement. Regarding the different questionnaires used to capture subjective treatment effects, ketamine studies on treatment of depression generally used instruments designed to measure psychopathological constructs.

For example, the CADSS (Clinician-Administered Dissociative State Scale) was developed to assess trauma-related dissociative symptoms in patients with post-traumatic stress disorder, while the BPRS (Brief Psychiatric Rating Scale) was developed to rapidly assess symptom changes in schizophrenia and other psychotic disorders. In contrast, psilocybin studies used mystical experience questionnaires. We agree that this represents a methodological limitation, as these different approaches capture different therapeutic aspects.