Pharmaceutical psilocybin and community psilocybin are becoming two different drugs. Treating them as one policy question is the industry's biggest blind spot.

A patented, FDA-track product moving through Lilly-scale pharma capital. A licensed-facilitator service model built entirely outside the FDA. A capped state clinical pilot. A federal access program that fixed its supply chain by narrowing who gets through. These are not four routes to the same destination. They are becoming different products, for different patients, at different prices, and the field's habit of discussing psychedelic access as a single question is obscuring that rather than clarifying it.

Ask most people following this sector whether psychedelic medicine is arriving, and they will point to some mix of the same evidence: a Compass Pathways NDA moving through rolling FDA review, Oregon’s licensed service centers now years into operation, a Massachusetts pilot program working its way through the legislature, Eli Lilly agreeing to pay up to $3.8 billion for a late-stage 5-MeO-DMT program. Treated together, this reads as a single, accelerating story. It is not one story. It is at least three, and this desk’s own reporting across the past several months is the evidence that they are diverging, not converging.

The products are already different

Compass’s COMP360, and now AtaiBeckley’s BPL-003 under Lilly’s balance sheet, are patented, standardized pharmaceutical products moving through the FDA’s approval architecture, with fixed dosing, fixed formulations, and, eventually, insurance-billable procedure codes. Oregon and Colorado’s service-center model is not a drug product at all in the pharmaceutical sense. It is a supervised personal-use service, delivered by a licensed facilitator, using psilocybin sourced from a state-regulated but not federally standardized supply, entirely outside FDA jurisdiction. Massachusetts’s pilot is neither of those things; it is a capped, state-permitted clinical research exercise, three clinics, five years, explicitly structured to generate data rather than to scale access. Canada’s Special Access Program is a fourth model again, a federal exception-request pathway that, this desk documented directly, traded an unregulated supply chain for a legally clean one and, in doing so, cut its own approval rate to roughly one in six applicants. Four genuinely different mechanisms, not four maturity stages of one mechanism.

The populations will not overlap the way the industry assumes

A pharmaceutical psilocybin product moving through Medicare’s physician fee schedule, the same reimbursement infrastructure this desk covered in detail for its 2027 proposed changes, is built to reach patients through psychiatric diagnosis, insurance coverage, and eventually a covered procedure code. An Oregon service-center session costs on the order of one to several thousand dollars, paid entirely out of pocket, no diagnosis required. Those are not the same patient population separated by a few years of regulatory catch-up. They are different populations separated by ability to pay, by whether a formal psychiatric diagnosis is the gateway or irrelevant, and by whether the experience is delivered as a medical procedure or a supervised personal one. A pharmaceutical approval does essentially nothing to make an Oregon-style session more accessible, and an expanding service-center market does nothing to get a patented product into a Medicare beneficiary’s covered treatment plan. Treating “psychedelic access” as one improving number obscures that these are two different access problems with two different solutions, advancing on entirely separate tracks.

Canada already ran the experiment that shows why this matters

The most important evidence for this argument is not hypothetical. It already happened, in a system this desk examined directly. Canada tried to fix its own single-track supply-chain problem by moving toward something closer to the pharmaceutical model’s logic, a controlled, licensed, verifiable supply, and the practical result was an approval rate around 16 percent. Building a legally clean pharmaceutical-style pathway did not expand who could get psilocybin. It narrowed the door, because a clean supply chain and broad access turned out to be in tension, not in sequence. Every American state currently choosing between an Oregon-style service model and a pharmaceutical-approval-contingent trigger law is choosing between two different versions of Canada’s dilemma, not picking a faster or slower path to the same outcome.

Why the industry keeps missing this

Coverage of this sector, including a fair amount of this desk’s own early framing, has tended to treat every regulatory action, a voucher, a guidance document, a state bill, an acquisition, as another data point on a single progress bar toward “psychedelic medicine arriving.” That framing serves headlines better than it serves clarity. A $3.8 billion pharma acquisition and a three-clinic state pilot are not two speeds on the same road. They are different roads, built for different destinations, and conflating them produces bad judgment in exactly the places judgment matters. An investor pricing Compass’s approval odds is answering a different question than a state legislator deciding whether to fund a service-center model, and a patient asking when psilocybin therapy will be available to them is asking a question that does not have one answer, because which version of “available” they mean determines which of at least four separate systems they are actually asking about.

What treating this as one question actually costs

The practical harm is not abstract. Sponsors and investors who read pharmaceutical approval as the leading indicator for the whole sector will misjudge how much a Compass or Lilly outcome actually predicts about service-center access, because the two are not on the same curve. State policymakers who assume a future FDA approval will eventually solve their state’s access question are making the same category error Canada’s regulators made, assuming supply-chain legitimacy and broad access move together when the evidence says they trade off against each other. And patients and advocates deserve a clearer answer than “psychedelic therapy is coming,” because what is actually coming is at least two, and plausibly four, different things, arriving on different timelines, reaching different people, at different prices, through entirely different institutions.

What people building each system actually say

This desk asked people working inside each of these tracks whether they read the divergence the same way.

Ron Brooks, founder and principal consultant at RRB LLC, spent fifteen years inside regulated laboratories before turning to cannabis and psychedelic regulatory consulting, including a term as Scientific and Laboratory Director at BelCosta Labs, a California cannabis testing laboratory, from January 2019 to August 2021. His read tracks the cannabis parallel this piece leans on directly: “Pharmaceutical psilocybin and the state facilitator models are already two different products serving two different patients: one is a drug moving through FDA’s evidentiary machinery, the other is a regulated service. Nothing in either pathway forces convergence. The biggest misconception is that access gets decided in the statute. In practice, it gets decided in the infrastructure underneath: licensing, product testing, quality control, and enforcement.” Brooks states no financial interest in any psychedelic or cannabis company.

None of that infrastructure question means anything to a patient if the treatment isn’t covered. Alan Wiederhold, a healthcare executive with nearly thirty years in employer-sponsored benefits and stop-loss insurance and president of accident and health at Amynta Risk Solutions, argues the piece’s divergence has a second axis entirely separate from testing and licensing: who actually pays. “The danger is assuming that FDA approval automatically equals access. In employer-sponsored healthcare, coverage, reimbursement, medical necessity, network design, cost-sharing, and navigation all determine whether a treatment is actually reachable for the patient. A therapy can be clinically promising and still functionally inaccessible.” He points to an existing example of exactly the dynamic this piece traces for psilocybin: “We already see this with ketamine. IV ketamine itself is inexpensive, yet many patients cannot access it through insurance coverage. Meanwhile, branded alternatives may be more reimbursable but far more expensive. That creates an access paradox: the lower-cost intervention may be harder to obtain, while the higher-cost patented pathway becomes the reimbursed default.” Wiederhold discloses that he helped create Clarity Corridor, a plan-integrated ketamine stabilization program built with Amynta Risk Solutions, and that he is a co-founder of Neuma. Neuma also markets a commercial supplement for post-psychedelic-therapy recovery, a positioning the desk notes here that Wiederhold did not include in his own disclosure.

Dr. Qiao Yufei, a family physician at MediWay Medical Clinic, sees the divergence less as a policy failure and more as two tools serving different patients: “I don’t see pharmaceutical psilocybin and community-based psilocybin services as rivals. A patient with severe depression, multiple medical conditions, or complex medication use may need a tightly regulated medical pathway. Another person may value a structured service model centered on preparation, support, and reflection. Confusion starts when people assume every patient fits a single route instead of matching care to clinical needs, safety, and personal circumstances.” Qiao states no involvement in psychedelic drug development, investment, licensed facilitation, or commercial psychedelic services.

Not everyone reads the trajectory as permanent separation. Agnes Horry, founder of The Psychedelic Digest, an independent media platform covering psychedelic science, policy, business, and culture, expects the opposite of this piece’s central claim over the longer run: “From where I sit, I think we’re heading towards a hybrid system rather than one model replacing the other. Pharmaceutical pathways and facilitator-led models are solving different problems today, but they don’t have to compete forever. The future isn’t about choosing one pathway over another. It’s about building a system where different models serve different needs while maintaining safety, quality, and informed choice.” That is a direct complication of the argument above, not a confirmation of it, and it deserves to be read as such rather than smoothed over. Horry is also a self-published author on microdosing and a content creator in this space, disclosures worth weighing alongside her view.

The frame

None of this argues that any of these four tracks is the wrong one, or that convergence is impossible eventually. It argues that the field, this desk included at times, has been describing genuine structural divergence as if it were staggered progress on a single path, and that the difference matters for anyone making a decision based on the description. A patented product moving through Lilly’s balance sheet and an Oregon facilitator’s supervised session are not the same medicine at different stages of the same rollout. They are becoming two different medicines, for two different kinds of patients, built by two different kinds of capital, and the sooner coverage of this sector says so plainly, the sooner the people actually making decisions, investors, legislators, and patients among them, can ask the right question instead of the convenient one.