Table 2 Assessment of risk of bias of individual studies (using the RoB-2 for RCTs).

Dolder et al.²⁵

Low risk

Low risk

Low risk

Low risk

Low risk

Low risk

Explanation

The study used a double-blind, placebo-controlled, crossover design, with balanced order randomization.

Both groups received the intended intervention (LSD or placebo) under tightly controlled conditions.

The study reported minor data loss (2/24 participants due to technical issues), unlikely to influence outcomes significantly.

The outcomes were measured objectively using standardized tasks and tools, with blinding maintained throughout.

The study appears to have reported all relevant outcomes, and the analysis seems consistent with the planned methods.

Overall, the study exhibits a low risk of bias across the key domains. The randomization and blinding processes were robust, and the analysis appears comprehensive and consistent with the study design.

Pokorny et al.²²

Some Concerns

Low risk

Low risk

Low risk

Low risk

Low risk

Explanation

The study was described as double-blind and randomized.

The baseline characteristics of the participants were not reported in detail, making it difficult to assess if the randomization was successful in producing comparable groups.

The study is described as double-blind, and both participants and study personnel were likely unaware of the allocation, as psilocybin and placebo were administered in identical-looking capsules.

The study appears to have adhered to the intended interventions, with no deviations that could lead to bias. Participants were required to refrain from drug use, alcohol, and caffeine, and this was monitored.

One participant was excluded for not understanding the task, and data from another participant were missing due to technical issues. However, the number of missing data points seems minimal relative to the total sample size.

Outcomes were likely assessed by study personnel who were blinded to the intervention. The outcomes measured (empathy, moral decision-making) were measured by standardized tests

The same measures were used for all participants, reducing the risk of bias in measurement.

The study appears to have reported on all the outcomes that were planned and measured, with no evidence of selective reporting based on the results.

There is an area where the study lacks detailed information (e.g., randomization process) which leads to an overall judgment of “Some concerns” for the risk of bias. However, the study appears to have taken reasonable steps to minimize bias in most domains, suggesting that the results are likely reliable despite these concerns.

Uthaug et al.¹¹

Low Risk

High Risk

Low Risk

Some Concerns

Low Risk

High Risk

The study involved random assignment of participants to either the ayahuasca or placebo group, which was managed by the host organization. There is no mention of any significant issues in the randomization process, and the baseline characteristics of participants between the groups appear balanced.

Several factors contribute to a higher risk in this domain. The interventions were administered by facilitators of the retreat rather than the research team, and the procedures for preparing the ayahuasca and placebo were not standardized across all locations. Additionally, the research team had no control over the administration of doses or the environment in which the intervention took place. The study also reports that the contents and storage of the capsules were not well-documented, which introduces potential variability and risk.

The study does not report any significant missing data, and all participants seem to have been accounted for in the final analysis.

Several of the outcomes were based on self-report measures, which are subject to bias, especially in a study involving psychedelic substances where participants may have expectations about the effects. However, the study used validated questionnaires and tasks for their other outcomes.

The study appears to have reported on all pre-specified outcomes and used appropriate statistical analyses. There are no indications of selective reporting based on the provided information.

While the study has strengths in randomization and the comprehensiveness of outcome reporting, the significant deviations from intended interventions and the concerns over the measurement of outcomes (self-report) contribute to an overall high risk of bias. The lack of control over intervention standardization and potential influence of participant and facilitator biases are key issues that impact the reliability of the findings.