Each dot is one trial: how many points more the drug cut depression than the control group, on the same 60-point scale (MADRS). Further left = bigger benefit. Hover or tab to a dot for details.

Phase 2 (smaller, earlier) Phase 3 (large, pivotal) Esketamine, already FDA-approved Red arrow: Compass, phase 2 → phase 3

The blinding problem

Share of participants who correctly guess whether they got the real drug.

Read this before comparing dots

Trials measured at different times (day 8 to week 6), in different patients (treatment-resistant vs general depression), against different controls (placebo, a 1 mg dose, niacin). Small phase 2 trials usually shrink in phase 3, as Compass did. A March 2026 JAMA Psychiatry analysis found psychedelic therapy no better than antidepressants when patients in both groups knew what they were taking.

Sources: Goodwin et al. 2022, NEJM; Compass Pathways COMP005 (June 2025) and COMP006 (Feb 2026) releases; Raison et al. 2023, JAMA; Definium (MindMed) Emerge release, 2026; GH Research GH001 phase 2b; Popova et al. 2019, Am J Psychiatry (esketamine). Blinding: Williams, Barnett & Szigeti 2026, JAMA Psychiatry (press figures, approximate). 95% confidence intervals drawn where published.