Can pragmatic research, real-world data and digital technologies aid the development of psychedelic medicine?

Carhart-Harris and colleagues interpret their review as a call for a pluralistic research strategy that pairs confirmatory RCTs with pragmatic trials and large-scale real-world data to inform the safe, effective and cost-efficient rollout of psychedelic therapy. They contend that pragmatic designs can address questions that DB-RCTs typically constrain, such as variability in dose, number and timing of sessions, broader eligibility criteria, and the role of post-dosing psychotherapy. The authors also highlight single-case (n = 1) experimental designs and citizen-science initiatives as complementary methods that can generate high-frequency, idiographic data amenable to aggregation and Bayesian or multilevel analysis, enabling more nuanced modelling of treatment effects over time. In positioning these recommendations relative to earlier research, the paper stresses that DB-RCTs remain essential for regulatory approval and for convincing sceptical stakeholders, but argues that over-reliance on the traditional evidence hierarchy risks missing clinically relevant parameters early in development. The authors emphasise the potential of digital tools and biometric sensors to produce objective, scalable outcome measures and to support centralised registries that can feed prediction-of-response algorithms and safety surveillance. The authors acknowledge several practical and ethical constraints. They warn against relaxing contraindication-based screening designed to reduce adverse events, and they note that progressive policy changes will be needed to permit many of the proposed pragmatic approaches. Funding and incentives are identified as major barriers: a multi-site pragmatic programme would likely require payer or industry engagement, and mainstream institutional funding for psychedelic research remains limited. They also stress the risk of a hype cycle—rapid expansion of access prior to sufficient evidence—using the medical cannabis experience as a cautionary example. For future research and implementation, the paper recommends designing pragmatic trials and registries with careful attention to data quality and fitness for purpose, setting reasonable limits on dosing frequency to deter bad practice, and engaging diverse stakeholders—including the public, policymakers, clinicians and investors—to ensure sustainable development. The authors suggest that aggregated real-world data could help define ‘‘no-go’’ screening criteria, personalise treatment parameters, monitor safety across demographics and ultimately protect the long-term credibility and utility of psychedelic medicine.