The experience that researchers now call “dissociation” — the sense of floating outside yourself, of watching the world at a distance, of your sense of identity briefly becoming porous — was not supposed to be therapeutic. It was a side effect. The pharmaceutical target was pain. The drug was ketamine, approved in 1970 as a surgical anesthetic, and for the first thirty years of its clinical life the dissociative state it produced was the thing doctors tolerated in order to get the numbing they needed.
Then the psychiatrists started paying closer attention.
Today, dissociatives as a class sit at the most contested intersection in American medicine: a genuine pharmacological breakthrough — one that may be reshaping how we understand depression, addiction, and brain plasticity — operating inside a regulatory framework that can’t quite decide what to make of it. Understanding what ketamine actually does to the brain explains both why the clinical excitement is justified and why the regulatory concerns are, also, justified. The two things are not in tension. They are the same story.
What the Science Finally Shows
For decades, ketamine’s antidepressant effect was observed without being explained. Psychiatrists had known since the early 2000s that infused ketamine could lift severe, treatment-resistant depression in hours — where SSRIs and SNRIs take weeks if they work at all. The effect was real. The mechanism was mysterious.
A study published in Molecular Psychiatry in March 2026 clarified what’s happening. The research, examining patients with treatment-resistant depression given ketamine infusions, found that the drug’s rapid-onset antidepressant effect is driven by AMPA receptor activation — the same receptors that regulate fast excitatory signaling in the brain. Ketamine blocks NMDA receptors (its original anesthetic mechanism), but that block appears to create a rebound surge in AMPA receptor activity, which in turn drives a rapid process called synaptogenesis: the brain literally growing new synaptic connections, often within hours of a single infusion.
This is not what antidepressants do. SSRIs modulate serotonin reuptake over weeks. Ketamine triggers synaptic growth directly and quickly. Researchers have started calling it a “psychoplastogen” — a substance that induces neural plasticity — and the implications extend beyond depression. The same AMPA-mediated synaptogenesis that appears to reset depressive circuitry may also be relevant to addiction, trauma, and the kind of cognitive rigidity that characterizes severe OCD. This is the biological case for taking ketamine’s psychiatric applications seriously.
The dissociation, researchers now suspect, may be part of the mechanism rather than just a side effect. Several studies suggest that the ego-dissolution component of a ketamine session — the loosening of fixed self-referential patterns — may itself be therapeutically active, particularly in the context of psychotherapy integration. Whether the trip is the treatment or the AMPA surge is the treatment, or both, is still being worked out. What’s clear is that dissociation is not incidental noise. It is a pharmacological event with its own cognitive and psychological signature.
The Regulatory Problem Is a Biology Problem in Disguise
That psychological signature is also why the DEA classifies ketamine as Schedule III, why Spravato (esketamine, the FDA-approved nasal spray form) requires two hours of in-clinic monitoring after every dose, and why the FDA issued a formal warning on compounded ketamine products — particularly oral and at-home formulations. The drug that briefly makes your sense of self permeable cannot be sent home unsupervised without meaningful risk.
Spravato navigated this with the REMS program. Every Spravato-certified clinic is required to monitor patients on site for a minimum of two hours after each dose and ensure they can’t drive afterward. The protocol is tight because the side-effect profile is real: dissociation, sedation, elevated blood pressure, and in a small number of cases, emergent psychological distress. The REMS is not regulatory overcaution. It reflects a genuine, documented safety signal.
The protocol is tight because the side-effect profile is real: dissociation, sedation, elevated blood pressure, and in a small number of cases, emergent psychological distress.
The compounded ketamine market didn’t have a REMS. It had a mailbox.
Between 2020 and 2024, a wave of telehealth companies built businesses around mailing compounded ketamine — typically oral lozenges — to patients diagnosed remotely for depression. The argument was access: IV infusions require a clinic visit and often cost $500–$1,000 per session out of pocket; mailed lozenges could reach people who couldn’t afford the clinic model. A March 2026 STAT News investigation found that the access argument was not wrong — some patients in rural areas genuinely could not access clinic-based ketamine. But it also documented what the Reddit communities for at-home ketamine users had been saying for years: unsupervised use, inconsistent dosing, patients escalating their own doses without clinical oversight, and in some cases, the development of ketamine dependence.
The FDA’s warning on compounded ketamine formulations, issued in 2025 and reemphasized in 2026, was direct: the safety data that supports IV and intranasal ketamine in supervised clinical settings does not automatically transfer to oral formulations delivered at home. The bioavailability is different. The monitoring is absent. The emergent psychological states that the REMS was designed to catch in real time have nowhere to go.
The Telemedicine Extension Is a Stay of Execution, Not a Settlement
On December 31, 2025, the DEA issued its Fourth Temporary Extension of COVID-era telemedicine flexibilities, extending the ability of providers to prescribe controlled substances — including ketamine — via telehealth through December 31, 2026. That extension kept the telehealth ketamine industry functional for another year.
But “functional through December 2026” is not a policy. It’s a deadline. The DEA has signaled clearly that its final telemedicine rule, when issued, will impose more stringent requirements on prescribing Schedule II and III substances without in-person evaluation. The exact contours of the final rule are not yet public, but the direction is not ambiguous: the era of zero-guardrail ketamine telehealth is ending.
This creates a difficult reality for the patients who have genuinely benefited from telehealth-delivered ketamine. A 2026 survey of compounded ketamine users found that a significant proportion reported meaningful improvements in treatment-resistant depression, suicidal ideation, and anxiety — and that many had no realistic access to clinic-based care. The DEA crackdown, if it comes without building expanded clinic capacity first, will close the telehealth door without opening another one.
This is the harm reduction problem embedded in the regulatory problem: the answer to unsafe at-home use is not just “no at-home use.” It is a genuinely accessible, properly supervised alternative. The regulatory reckoning is justified. The access infrastructure to absorb it does not yet exist at scale.
What This Means for the Recovery Field
The growing use of ketamine in the SUD treatment context — not as a primary addiction treatment, but as a tool for the co-occurring depression and trauma that characterize many substance use disorders — adds another layer. Controlled research suggests ketamine may reduce cravings in alcohol use disorder and support engagement with therapy in PTSD treatment, both of which are common co-occurring conditions in people with opioid use disorder or stimulant dependence.
If the regulatory framework tightens to the point where only Spravato-certified clinics can administer ketamine, the population of people with SUD plus co-occurring severe depression will face a new access barrier: Spravato’s REMS requires stable clinical settings, regular monitoring, and a treatment-resistant depression diagnosis. People in early recovery often don’t have all of these in place.
The ideal outcome is a regulated framework that expands access to supervised ketamine while shutting down the unsupervised mail-order model. That framework doesn’t exist yet. The DEA’s telemedicine rule, expected in 2026 or 2027, will either build it or break it. Until then, the drug that works in hours when others take weeks is operating inside a system that is, in its way, still dissociated from its own evidence base.
The DEA’s telemedicine rule, expected in 2026 or 2027, will either build it or break it.
Filed Under
sciencepsychologytreatmentKetamineFDADEA