Ketamine for OCD: Stanford Trial Shows 3-Week Effect

TL;DR

• A Stanford research team presented unpublished results at APA 2026 from an NIH-funded randomized controlled trial of ketamine for OCD.
• 45 adults with moderate-to-severe OCD received either a single 40-minute IV infusion of ketamine or a comparison drug (midazolam — a different sedative used to control for "feeling something").
• The ketamine group's OCD symptoms dropped from "severe" to "mild" within 24 hours — and a substantial portion of that improvement was still measurable three weeks later. The midazolam group did not show the same effect.
• About 50% of ketamine patients met the "responder" threshold (a 35%+ drop in OCD severity) at one and two weeks. The comparison group was at 14%.
• This matters because OCD has been one of the most stubborn psychiatric conditions to treat. Standard SRI medications help many patients but leave a substantial minority without enough relief.
• Ketamine is not yet FDA-approved for OCD. Off-label use is legally prescribable by a physician but not a labeled indication.

This is part of a series of patient-facing posts translating what was discussed at the American Psychiatric Association's 2026 Annual Meeting.

Who presented this

The presenting researcher was Carolyn Rodriguez, MD, PhD — Director of the Stanford OCD Research Lab and Professor of Psychiatry at Stanford. She has been the leading voice on ketamine for OCD for over a decade. Her 2011 case report and 2013 randomized trial established the initial evidence base for this indication.

The data she presented at APA 2026 is from a larger NIH-funded follow-up study (R01 MH105461) that has not yet been published. Conference presentations of unpublished data are common when researchers want to share findings while the manuscript is in peer review. You're reading about this weeks or months before it appears in any journal.

Why OCD needs new treatments

Obsessive-compulsive disorder affects roughly 2-3% of adults in the U.S. It's characterized by intrusive, unwanted thoughts (obsessions) and repetitive behaviors that the person feels driven to perform (compulsions).

The standard treatments:

• High-dose SRIs (selective serotonin reuptake inhibitors — Prozac, Zoloft, Luvox, etc.) at doses higher than what's used for depression. Helpful for about 40-60% of patients, but full response often takes 10-12 weeks.
• CBT with exposure and response prevention (ERP) — a structured therapy where the patient progressively confronts triggers without performing compulsions. Gold-standard but availability is limited and adherence is hard.
• Antipsychotic augmentation when SRIs aren't enough — adds modest additional benefit.
• Deep brain stimulation (DBS) for the most severe, treatment-refractory cases — requires brain surgery and is reserved for a small population.

For the patients in between — too symptomatic to be helped by SRIs alone, but not severe enough to warrant DBS — options are limited. This is the gap Rodriguez has been working to fill.

What the trial looked like

Design:

• Double-blind, placebo-controlled (neither the patient nor the assessing physician knew who got what)
• 45 adults with OCD randomized
• Two arms:
  • Ketamine arm (n=30): 0.5 mg/kg IV over 40 minutes — the standard dose used in ketamine therapy for depression
  • Comparison arm (n=15): midazolam 0.045 mg/kg over 40 minutes — a different sedative used as an active control

Why midazolam as the comparison drug? Because researchers needed something that would also produce a feeling of being sedated, to control for the possibility that patients improved just because they expected to or because they felt something. This is a much stronger comparison than just using saline (saltwater).

Who was in the trial:

• Adults aged 18-65 with at least moderate OCD (Y-BOCS score ≥16 — a standard severity measure)
• Spending more than 8 hours per day on obsessions
• Not currently taking psychiatric medications
• Excluded if they had severe depression, psychosis, bipolar disorder, or current substance use disorder

This last point is important: this wasn't a "depression with OCD symptoms" study. These were people whose primary problem was OCD.

Treatment background: about 7 in 10 patients had previously failed at least one SRI. About 4 in 10 had completed CBT with ERP. These were not first-line patients — they were people for whom the standard playbook hadn't worked.

What happened

OCD severity is measured on the Y-BOCS (Yale-Brown Obsessive-Compulsive Scale), which runs from 0 to 40. Severe OCD typically scores 24 or higher; moderate is in the high teens.

Time point	Ketamine group	Comparison (midazolam) group
Before infusion	26 (severe)	26 (severe)
Day 1 after	14 (mild)	22 (still severe)
Week 1	17	24
Week 2	19	25
Week 3	21	25
Week 4	22	24

Translating into everyday language:

• Within 24 hours of a single 40-minute ketamine infusion, average OCD severity dropped from "severe" to "mild."
• The benefit faded somewhat over 4 weeks, but a meaningful portion was still present at Week 3.
• The comparison group barely moved.

The "responder" rate — the proportion of patients whose OCD severity dropped by at least 35% — was:

Time	Ketamine	Comparison
Day 1	55%	20%
Week 1	53%	14%
Week 2	50%	14%
Week 3	29%	7%
Week 4	27%	7%

The differences at Weeks 1, 2, and 3 were all statistically significant. The benefit is real, the comparison group ruled out simple expectation effects, and the methodology met modern standards for a controlled trial.

What patients have said

In Rodriguez's earlier studies, the qualitative reports from responders have been striking. Several quotes that have circulated in the field:

"I feel as if the weight of OCD has been lifted. I want to feel this way forever."

"I tried to have OCD thoughts but I couldn't."

"I don't have any intrusive thoughts. I was laughing when you couldn't find the key, which normally is a trigger for me. This is amazing, unbelievable. This is right out of a movie."

"I feel like someone is giving me an explanation for my OCD."

These aren't typical research-paper language. They're the language of someone discovering a different mental state is possible.

What was honest about the limitations

Rodriguez was direct about three things:

1. Blinding wasn't perfect

86.6% of patients in the ketamine group correctly guessed they had received ketamine (versus 73.3% of the midazolam group). The brief dissociative experience that comes with ketamine is hard to disguise.

Researchers debate whether this counts as a fatal flaw or a manageable limitation. Rodriguez's position — and one I share — is that the effect size is too large to be explained by expectation alone. A patient who expects to improve doesn't typically drop 12 points on a clinician-rated severity scale within 24 hours.

2. The effect fades

A single dose produced 1-3 weeks of meaningful improvement, but the benefit declined over 4 weeks. This isn't a cure; it's a window. The clinical question becomes: how do you maintain the benefit?

Several models are being studied:

• Repeated infusions on a maintenance schedule
• Bridging to oral medication for maintenance
• Combining ketamine sessions with CBT/ERP during the response window — when the brain is more receptive to learning new patterns

3. Side effects

Dissociation during the infusion (peaking at about 30 minutes and resolved by 2 hours). No serious adverse events. Standard ketamine side effect profile.

What this means in plain terms

If you have OCD and you've been frustrated with your current treatment:

• The standard playbook (SRI + CBT) is still the recommended starting point. That hasn't changed.
• For patients who haven't responded enough, ketamine therapy is becoming a real option — not approved for OCD specifically, but increasingly being offered off-label in interventional psychiatry programs.
• The effects are rapid (hours), and meaningful, but not permanent. Maintenance protocols are still being worked out.
• This is best done in a structured clinical program that includes physician supervision, dose adjustment, and ideally integration with therapy.

This is one of the conditions where the evidence base is moving fastest. Rodriguez's data, even before publication, will reshape how clinicians think about OCD treatment.

Where MDMA fits

Rodriguez also briefly touched on MDMA-assisted therapy for OCD — much earlier in the research timeline than ketamine. The hypothesis: that MDMA's effects on emotional processing might help patients engage more productively with exposure-response prevention work. A few small trials are running. No definitive data yet. But it's a thread to watch.

How to find ketamine for OCD

Ketamine treatment for OCD is not FDA-approved, but is increasingly offered off-label by interventional psychiatry programs. The practical landscape:

• Academic OCD programs at major medical centers sometimes offer ketamine as part of research or compassionate-use protocols
• Interventional psychiatry clinics in major metros may offer it cash-pay
• Telehealth programs like Isha Health typically treat depression and anxiety; OCD treatment depends on the specific program and presentation

The most useful conversation is with a physician who can review your specific OCD history, prior treatments, and severity. If you're in a state where Isha operates, an initial consultation can determine whether at-home oral ketamine is a reasonable next step for your situation.

Sources cited

• Rodriguez CI et al. American Journal of Psychiatry 2011 — First case report of ketamine for OCD
• Rodriguez CI et al. Neuropsychopharmacology 2013 — First randomized controlled trial of ketamine for OCD
• Rodriguez CI et al., unpublished — APA 2026 presentation, NIMH R01 MH105461
• Bloch MH et al. Biological Psychiatry 2012 — Open-label ketamine in OCD patients with comorbid MDD
• Berman RM et al. Biological Psychiatry 2000 — Foundational ketamine antidepressant trial
• Zarate CA et al. Archives of General Psychiatry 2006 — Ketamine for treatment-resistant depression
• Goodman WK et al. Archives of General Psychiatry 1989 — Y-BOCS development and validation
• NIMH R01 MH105461 — NIH grant supporting the Rodriguez ketamine-for-OCD RCT

---

Related Articles

• APA 2026 Annual Meeting Recap
• The 40-Year Antidepressant Plateau
• Vagus Nerve Stimulation for Depression
• How does ketamine work?

---

Considering ketamine therapy? Isha Health offers physician-led at-home treatment with an 88.8% improvement rate. Check appointment availability.