OCD: When Standard Treatment Isn't Enough – Your Options

First, the direct answer. This page describes the adult treatment pathway. If you have obsessive-compulsive disorder and the standard treatments have not been enough (a proper course of exposure and response prevention, an SSRI at adequate dose and duration, or both), you are not out of options, and the remaining ladder is longer than most people are told. It is also less psychedelic than the headlines suggest: for OCD, unlike depression, the established next steps are mostly refinements of the standard treatments (higher doses, longer courses, different combinations) plus, at the far end, one neurosurgical option with real evidence behind it. This guide walks the whole ladder, with what the evidence and the European guidelines actually say at each rung, and a sober account of where ketamine and psilocybin research stands.

One reassurance before the list. Partial response is the norm in OCD care, not the exception, and the reference guidelines this page leans on: the UK's NICE guideline CG31 and Germany's S3-Leitlinie for obsessive-compulsive disorders; plan explicitly for it, with defined steps after each treatment that has not worked. Nothing on this page is a recommendation of one option over another; which step fits you is a decision for you and a clinician who knows your history.

What the options actually are

So much for the compressed version; here is the same ladder, rung by rung.

ERP: the first line, and the first thing to re-examine. Both reference guidelines put cognitive behavioural therapy including exposure and response prevention at the front of OCD treatment (NICE CG31; AWMF 038-017). ERP is a specific method (deliberately confronting the triggers of obsessions while not performing the compulsion, with a therapist trained in exactly this), and it is one of the most under-delivered treatments in psychiatry: general talking therapy, supportive counselling, or CBT that never included systematic exposure work is not ERP. If a course "didn't work", start by asking what was actually delivered, over how many sessions, and whether exposure exercises happened between them. A second course, with a therapist who specialises in OCD, in a different format, or at higher intensity, including inpatient or day-clinic ERP programmes in several European countries: is a standard guideline move, not an admission of failure.

SSRIs, done properly: higher doses, longer trials. SSRIs are the first-line medication for OCD in both guidelines, and OCD has its own pharmacological rules that non-specialists sometimes miss: response typically takes longer than in depression, up to 12 weeks, and doses often need to be higher than in depression, reached by gradual optimisation with the prescriber and held long enough at a level you tolerate (NICE CG31). Not everyone needs the maximum dose, and no dose change is a do-it-yourself step. A "failed" SSRI trial at a low dose for six weeks is not a failed trial by OCD standards. If one SSRI has genuinely not helped at an adequate dose and duration, or was not tolerated, switching to a different SSRI is one next rung; clomipramine (an older tricyclic with the strongest single-drug history in OCD) is an option after at least one ineffective or poorly tolerated SSRI, or earlier if you prefer it after previous benefit. It carries real cardiovascular, interaction and overdose risks: where cardiovascular risk is significant, NICE advises an ECG and blood-pressure assessment before starting, and it needs specialist monitoring throughout. Combining medication with ERP rather than choosing between them is itself a guideline option, and for many situations the recommended one; which first-line treatment to start with, ERP, medication or both, follows the degree of impairment and your preference, and severe impairment often warrants both from the outset.

Augmentation: the established specialist step most people haven't heard of. When SSRIs at adequate dose and duration have brought partial or no response, adding a low-dose antipsychotic (such as risperidone or aripiprazole) to the SSRI is an evidence-based augmentation strategy, and its place matters: it belongs in specialist review after the existing treatment has been properly optimised, not as a routine early step, and it comes with adverse-effect monitoring and a stop rule if it has not helped (NICE CG31; AWMF 038-017). It helps a meaningful minority (roughly one person in three in the trial literature), which is worth knowing before rather than after you conclude that medication has nothing left to offer. None of this list is a self-directed escalation plan; combinations and augmentation are prescribed and monitored by specialists. This is exactly the stage where a referral to a psychiatrist with OCD experience, rather than another round with a generalist, changes what is on the table.

Deep TMS: the non-invasive step most pathways skip. Before anything surgical, one qualified option deserves a mention: deep transcranial magnetic stimulation. A multicentre sham-controlled trial found benefit in OCD (Carmi et al. 2019), and the device has US regulatory clearance for OCD, but that clearance does not translate into routine European approval or reimbursement: NICE currently restricts TMS for OCD to research settings (NICE HTG548), recommendations vary by country, and availability is patchy, usually private. Worth asking a specialist about; not an established European pathway step.

Deep brain stimulation: real, rare, and last on the ladder. For the severest OCD, chronic, disabling, resistant to multiple adequate courses of ERP and medication, deep brain stimulation (DBS) exists: electrodes implanted in defined brain targets. Stimulation is adjustable and can be switched off, but switching it off does not reverse the surgery; implantation itself carries surgical, hardware and psychiatric risks that belong in the consent conversation. This is genuine neurosurgery, performed in a handful of specialist centres in Europe, on small numbers of carefully assessed patients, usually within research protocols or specialist commissioning. The controlled evidence for exactly this group: an individual-participant meta-analysis of nine randomised sham-controlled trials (91 participants) found a mean 5.1-point Y-BOCS advantage of active over sham stimulation (95% CI 2.0 to 8.1), with low certainty and substantial heterogeneity (Cohen et al. 2025, Molecular Psychiatry). DBS is nobody's next step after two SSRIs; it is a highly specialised option for carefully assessed refractory illness, and proof that the ladder does not end where most treatment histories stop.

Ketamine: early evidence. Ketamine's rapid effect in depression has made it a natural OCD candidate, but the evidence is thin: the controlled evidence comes from two very small crossover trials, and the routes should be kept distinct. The first, intravenous: 15 unmedicated adults (not treatment-resistant), rapid reductions in obsessions after a single infusion, with carryover effects limiting interpretation; in the first treatment period, 4 of 8 responded after ketamine versus 0 of 7 after saline at one week, a proof-of-concept figure, not a routine-care response rate (Rodriguez et al. 2013). The second, intramuscular, in treatment-resistant OCD with an active fentanyl control: 12 randomised, 10 completed, greater short-term symptom reduction after ketamine, with dissociative and cardiovascular effects and two discontinuations at the higher dose (Beaglehole et al. 2025). Together these add preliminary evidence, not confirmation of durable benefit or an established dosing course. No regulator has approved ketamine or esketamine for OCD; esketamine's EU label covers treatment-resistant depression together with an oral antidepressant, plus a separate short-term psychiatric-emergency indication, and neither is OCD, so treating coexisting depression does not establish an anti-OCD effect. Anyone offering ketamine as an established OCD treatment is ahead of the data. In Europe, a ketamine OCD trial at the Medical University of Vienna has an authorised CTIS record (2024-518212-40-00); whether it is enrolling right now is a question for the site, since registry status lags reality. The method-side summary of exactly this evidence, kept consistent with this page, is in ketamine for OCD: where the evidence stands. Where ketamine's evidence is genuinely strong, depression, is covered in our ketamine evidence review; and if you also have depression that has failed two antidepressants, that is a separate assessment worth making, mapped in depression treatment isn't working: your options.

Psilocybin (a US research story, for now. Psilocybin for OCD has a long research pedigree), the first modern study dates to 2006, and a currently active pipeline of small trials at Yale, Johns Hopkins, Arizona and Toronto, plus one study in Israel. In Europe the picture is sparse: a UK phase 1 feasibility study at Imperial College London has completed with 19 participants (PsilOCD, NCT06258031), and as of October 2026 we found no European psilocybin trial for OCD recruiting on either major registry, which is a statement about those registries on that date, not a guarantee that none exists. No regulator anywhere has approved psilocybin for any indication, so routine clinic treatment is not a lawful offer in the EU, and a clinic or retreat selling it is a red flag. The narrow, country-specific exceptions (Switzerland, for instance, operates a restricted individual-authorisation pathway for selected banned substances) are case-by-case medical decisions under national rules, not evidence of efficacy and not a travel option to plan around. What the psilocybin evidence does and does not show in its best-studied indication is in our psilocybin evidence review.

How to work out where you are

First: what have you actually tried? Write it down, in one document: every course of therapy, with whether it included real exposure and response prevention, the number of sessions, and whether between-session exposure work happened; every medicine, with name, highest dose reached, weeks at that dose, effect and side effects; any augmentation; any inpatient or day-clinic treatment. Old letters and prescriptions are the raw material. In OCD this document matters even more than usual, because under-dosed and under-delivered first-line treatment is so common: the difference between "SSRIs don't work for me" and "I had one SSRI at a starting dose for six weeks" determines the entire next conversation.

Second: where are you now? Clinicians grade OCD severity with instruments like the Y-BOCS interview; there is no self-test that replaces that, but a written account of your current obsessions, compulsions and the hours they consume anchors an appointment. Depression travels with OCD often enough that it is worth checking separately: our eligibility check runs validated depression screeners privately in your browser; it is an orientation tool, not a diagnosis, a treatment recommendation or trial enrolment. If your low mood has independently failed to respond to two antidepressants, a door may be open on that side (depression options guide). One more thing a page like this should say plainly: intrusive obsessions about harm are a symptom of OCD, not evidence of intent to act, and clinicians who treat OCD know the difference; acute risk of any kind still deserves individual assessment. If you are reading this for someone else, our guide for family members covers how to help without pushing, in OCD, that includes not participating in rituals, which is its own skill.

Trials as an option

For ketamine and psilocybin in OCD, trials are the only supervised route in Europe, and the pipeline is small. As of October 2026: the Medical University of Vienna's ketamine OCD trial has an authorised CTIS record (2024-518212-40-00, authorised October 2024; confirm current recruitment with the site); the UK's PsilOCD feasibility study has completed (NCT06258031); and the active psilocybin OCD trials are in the US, Canada and Israel. Registries change faster than any page can: check ClinicalTrials.gov and the EU's CTIS register directly, and read our trials guide for how phases, placebo and consent work. The same ground rules apply: trials are free, voluntary and ethics-supervised; screening is strict and most applicants are screened out; and some psychedelic protocols require supervised tapering of serotonergic medication, for OCD that can mean the SSRI doing part of the work: any proposed withdrawal for research is a decision to take with your treating clinician, never alone.

Frequently asked questions

What helps with OCD when nothing works?

Often part of the ladder has not yet been tried: the full guideline sequence (proper ERP, an SSRI at OCD-appropriate dose and duration, a second SSRI, clomipramine, ERP combined with medication, antipsychotic augmentation, intensive or inpatient ERP) is longer than most treatment histories. But some people remain symptomatic despite adequate treatment, and that is a recognised clinical reality, not a sign of insufficient effort; completing the ladder is no guarantee of recovery. In that situation a specialist review can check the diagnosis, how treatments were delivered, tolerability and coexisting conditions, and discuss further options, including, for the severest cases, deep brain stimulation in a specialist centre. The right next appointment is with a psychiatrist or therapist who treats OCD specifically.

Is ketamine a treatment for OCD?

Not an established one. The controlled evidence is two very small crossover trials (one IV against saline, one IM against an active fentanyl control) showing short-term effects, plus small open studies with mixed results; no regulator has approved ketamine or esketamine for OCD, so any such use is off-label and ahead of the data. A Vienna trial record exists in CTIS; confirm recruitment with the site. Esketamine's EU label covers treatment-resistant depression with an oral antidepressant, plus a short-term psychiatric-emergency indication: relevant only if you separately have depression needing its own assessment, and not evidence of an anti-OCD effect.

Can psilocybin cure OCD?

Nobody knows yet, and no regulator has approved it. The completed studies are small: the field's current trials, mostly phase 1 and 2, are running in the US, Canada and Israel; in Europe the UK feasibility study has completed (19 participants) and we found nothing recruiting on the major registries as of October 2026. Outside a clinical trial there is no routine lawful psilocybin route for OCD in the EU; the rare national exception pathways are restricted, individual medical decisions, not an access route to plan for.

Do I have to redo ERP if it didn't work the first time?

Nobody can oblige you, but before writing it off, check what was actually delivered. ERP means systematic, planned exposure with response prevention, and much of what is sold as CBT for OCD never includes it. A second course with an OCD specialist, at higher intensity or in a day-clinic format, is a standard guideline option with genuine response rates, and it is what OCD specialists themselves usually recommend before, and alongside, any medication escalation.

What if I also have depression?

Common, and worth treating as its own track. Depression alongside OCD can blunt ERP and needs its own assessment; and if your depression has separately failed two adequate antidepressant courses, the treatment-resistant depression pathway, including EU-approved esketamine, may be open on that side. Our depression options guide maps that ladder, and our eligibility check tells you which routes are realistic in your country.

Sources

1. NICE CG31 — Obsessive-compulsive disorder and body dysmorphic disorder: treatment
2. AWMF 038-017 — S3-Leitlinie Zwangsstörungen
3. Rodriguez CI et al. Randomized controlled crossover trial of ketamine in obsessive-compulsive disorder. Neuropsychopharmacology 2013
4. Cohen SE et al. Deep brain stimulation for obsessive-compulsive disorder: individual-participant data meta-analysis of randomised sham-controlled trials. Molecular Psychiatry 2025
5. Beaglehole B et al. Intramuscular ketamine vs fentanyl in treatment-resistant OCD: double-blind crossover trial. J Psychopharmacol 2025
6. Carmi L et al. Deep TMS for OCD: multicenter sham-controlled trial. Am J Psychiatry 2019
7. NICE HTG548 — transcranial magnetic stimulation for OCD (research-only recommendation)
8. EMA — Spravato product information (licensed indications)
9. MHRA — off-label or unlicensed use of medicines: prescribers' responsibilities
10. ClinicalTrials.gov — NCT06258031, PsilOCD feasibility study (Imperial College London)
11. CTIS — the EU Clinical Trials Information System
12. EMA — Spravato European Public Assessment Report

---

This guide is for general information only and is not medical advice, a diagnosis, or a recommendation of any treatment. It does not describe doses, preparation or session experiences, and it is not a guide to obtaining any substance. Discuss your options with a licensed clinician who knows your history. If you are in crisis, contact your local emergency number (112 in the EU) or a crisis line immediately.