Every U.S. state now writing a psychedelic-access framework, from Oregon and Colorado’s supervised-service model to trigger laws like West Virginia’s that wait on FDA approval, is solving a problem Canada already spent five years solving, badly, then differently. Canada built two successive federal pathways to get restricted psychedelics to patients outside formal clinical trials. The first one solved for legal permission and left supply broken. The second one fixed supply and, in doing so, made the door harder to walk through. That sequence, permission without supply, then supply with a gate, is the part worth studying closely, because it is not a Canada-specific quirk. It is what tends to happen when a government tries to open access to a substance that has no approved commercial product behind it.
The first pathway: permission without supply
Health Canada’s initial tool was Section 56 of the Controlled Drugs and Substances Act, which gives the federal Minister of Health broad discretion to exempt a person or class of people from the Act for medical, scientific, or public-interest reasons. In August 2020, Health Canada used it to grant a small number of terminally ill patients legal permission to possess and use psilocybin for end-of-life distress, the first exemptions of their kind, and later extended similar exemptions to clinicians seeking experiential training.
The mechanism had an obvious gap. Section 56 removed criminal liability for possession. It did not create a legal source of the drug. Patients who received an exemption to use psilocybin still had no lawful supplier to obtain it from, which meant a formally sanctioned medical treatment rested on procurement Health Canada had no way to standardize or oversee. Permission existed; supply did not.
The second pathway: supply without permission’s old ease
The fix was the Special Access Program, an existing Health Canada mechanism that lets physicians request non-marketed drugs for patients with serious or life-threatening conditions when conventional treatment has failed. Psychedelics had been excluded from it since 2013. Following sustained advocacy, Health Canada reversed that exclusion effective January 5, 2022, opening the SAP to psilocybin and MDMA.
The SAP solves the first pathway’s core defect. An approval creates a closed, legal supply chain: the drug must come from a specific Health Canada-licensed dealer, manufactured to controlled quality standards, and shipped to a clinical or hospital setting for administration under a practitioner’s supervision. It is a real fix to a real problem.
The trade nobody advertised
Fixing supply moved the gate rather than removing it, and the practical effect has been a program that grants access to a small fraction of those who seek it. Since the SAP reopened, the approval rate for psilocybin requests has run at roughly 16 percent, with only a few hundred approvals total by the middle of last year against a far larger volume of requests, a number that tells you plainly this is not functioning as broad-based access.
Three structural features explain why. The SAP shifted the burden of the application from the patient to the treating physician, who must document that conventional pharmacotherapy and psychotherapy have been tried and failed, a demanding evidentiary bar that also requires finding a physician willing to take it on, which itself has become a bottleneck in smaller communities. Health Canada has, in practice, made the SAP the mandatory first stop: applicants are routinely told a Section 56 exemption will only be considered after an SAP request has been attempted and denied, which converts what was once a direct, patient-driven route into a secondary fallback that few reach. And even an approved SAP authorization does not guarantee timely delivery, since the small number of licensed suppliers can face inventory constraints that stall a cleared request between paperwork and patient.
The net result: Canada now has a legally clean supply chain and a narrower door than the one it replaced. Patients who might once have pursued a Section 56 exemption directly are instead funneled through a harder, physician-gated process first, and denial rates suggest the funnel is doing exactly what a gate does.
Why this matters for the U.S. right now
American states are at the point Canada was in 2020, building permission mechanisms, Oregon and Colorado’s licensed-facilitator model operating wholly outside the FDA, trigger laws like West Virginia’s that activate only upon federal approval and rescheduling, without yet having built the supply-chain infrastructure Canada eventually layered on top of its own permission structure. Canada’s history is a preview of the tension coming: a workable legal supply chain and broad practical access appear to trade against each other, at least under the tools tried so far. A framework that only grants permission risks the same problem Section 56 had, sanctioned use with no controlled source. A framework that centralizes supply through a small number of licensed manufacturers and gatekeeping physicians, which is close to the FDA-approval-plus-DEA-rescheduling model several state trigger laws are betting on, risks recreating the SAP’s low throughput once the legal product actually exists.
The caveats
Canada and the U.S. sit inside genuinely different systems, single-payer federal health administration against fifty states and a distinct FDA-DEA architecture, so the exact mechanisms will not transplant identically. The SAP’s low approval rate may partly reflect its narrow eligibility criteria, restricted to serious or life-threatening conditions with documented treatment failure, rather than the supply-chain design alone, and a broader U.S. indication, treatment-resistant depression rather than end-of-life distress specifically, could produce different throughput dynamics. And Canada’s framework continues to evolve; a 2025 federal court ruling found Health Canada had not adequately justified reversing its position on practitioner training exemptions, which may still reopen parts of the Section 56 pathway.
The frame
The uncomfortable lesson is not that Canada got it wrong. It is that fixing the problem access advocates cared about most, an unregulated, unstandardized supply chain, created a different problem for the people the whole framework was meant to serve: most who ask still do not get in. As American states build their own access architecture, several of them explicitly betting on the same federal-approval-plus-controlled-supply model Canada eventually adopted, the relevant question is not whether that model can create a legal supply chain. Canada’s experience says it can. The question is whether it can do that without narrowing the door to a fraction of those standing in front of it, and so far, one country’s five-year experiment suggests the two goals may simply be in tension, not sequential problems to be solved one after the other.