OCD Treatment Options Compared: ERP, Medication and What Comes Next

The two first-line treatments for obsessive-compulsive disorder (OCD) are cognitive behavioral therapy that includes exposure and response prevention (ERP) and a serotonin reuptake inhibitor medication. UK guidance treats an SSRI and more intensive CBT with ERP as comparably effective for mild to moderate impairment, and recommends combining them when impairment is severe. In the US, five medicines carry FDA approval for OCD: fluoxetine, fluvoxamine, sertraline, paroxetine and clomipramine.

OCD medication works differently from depression treatment in two practical ways: benefit can take up to 12 weeks to appear, and higher doses tend to work better. When the first plan helps only partly, adding ERP has stronger trial support than adding an antipsychotic.

OCD treatments side by side

ERP: the psychotherapy with the strongest OCD evidence

ERP is a structured form of CBT. With a therapist, you build a ladder of situations that trigger obsessions, then practice staying in contact with each trigger while not performing the compulsion, whether that is washing, checking, asking for reassurance or a mental ritual. Over repeated practice, the urge and the distress usually fall on their own. A separate guide covers what an ERP course looks like week by week.

UK guidance sets out two levels of intensity. Low-intensity CBT including ERP, up to 10 therapist hours per patient, is offered first when impairment is mild or when the person prefers a lighter approach. More intensive CBT including ERP, more than 10 therapist hours per patient, is offered as an alternative to an SSRI for mild impairment that did not respond and for moderate impairment.

The most-cited head-to-head trial gives a sense of scale. Adults were randomized to intensive ERP for 4 weeks followed by eight weekly maintenance sessions, clomipramine, the combination, or placebo. At week 12, response rates among participants who started treatment were 62% for ERP, 42% for clomipramine, 70% for the combination and 8% for placebo; ERP alone did not differ from the combination, and both beat clomipramine alone. That trial used an intensive format at expert centers, so results in routine care may be smaller.

ERP is not the same as general talk therapy or relaxation training. If a therapist has never asked you to practice resisting a compulsion, you have probably not had ERP, even if you have had CBT. That distinction matters when deciding whether therapy "has already been tried."

SSRIs and clomipramine: the FDA-approved medications

Each of the five approved OCD medicines has its own label:

• Fluoxetine is approved for acute and maintenance treatment of OCD and for OCD in children and adolescents aged 7 to 17.
• Fluvoxamine's only FDA indication is OCD; its approval rested on trials in adults and one in patients aged 8 to 17.
• Sertraline is approved for OCD in adults and in children and adolescents aged 6 to 17; OCD is its only pediatric indication.
• Paroxetine is approved for OCD in adults and is not approved for pediatric use.
• Clomipramine, a tricyclic antidepressant, is approved for OCD; its studies included adults and patients aged 10 to 17.

Every antidepressant, including these five, carries a boxed warning that antidepressants increased suicidal thoughts and behaviors in children, adolescents and young adults in short-term studies, with close monitoring advised for everyone who starts one. UK guidance adds that people should be told about possible early worsening of anxiety and about discontinuation symptoms if doses are missed or stopped. If you are thinking about suicide or are in crisis, call or text 988 (the 988 Suicide & Crisis Lifeline) or go to the nearest emergency room.

Some medicines that help depression or anxiety are not standard OCD treatments. UK guidance says SNRIs such as venlafaxine, tricyclics other than clomipramine, and MAOIs should not normally be used for OCD without another condition that calls for them. For a broader comparison of serotonin medicines, see fluoxetine vs sertraline.

Why OCD medication trials are longer and doses higher

Two features surprise people who have taken an antidepressant for depression before.

Time

UK guidance warns that OCD medication commonly has a delayed onset of up to 12 weeks, although depressive symptoms improve more quickly. Judging an SSRI as a failure at four weeks is usually too early for OCD. If it is effective, the same guidance recommends continuing for at least 12 months to prevent relapse and allow further improvement, then reviewing the need to continue.

Dose

A meta-analysis of 9 fixed-dose trials with 2,268 adults found that higher SSRI doses were more effective for OCD than low or medium doses, a pattern unlike major depression. Higher doses also led to more people stopping because of side effects. The authors found too little data to apply this to children. Your prescriber sets the dose and pace of any increase, weighing response against side effects; label maximums differ by drug.

In practice, an "adequate trial" for OCD means a dose in the effective range for long enough. Before concluding a medicine did not work, it is worth asking your prescriber whether both conditions were met. The guide to signs your treatment isn't working covers how to raise this.

Therapy, medication or both

For many adults the choice is genuinely open. UK guidance offers either an SSRI or more intensive ERP for moderate impairment, because they appear comparably effective, and recommends the combination for severe impairment. A meta-analysis of 52 trials found that adding psychotherapy to antidepressant medication worked better than medication alone in OCD, panic disorder and major depression, with effects lasting up to two years.

The broader question of how to weigh medication, therapy or both is covered separately.

When the first treatment helps only partly

Partial response is common. The evidence suggests a fairly consistent sequence, though the order should fit your history and preferences.

1. Check the basics: was the dose adequate, was the trial long enough, and were doses missed?
2. Add or intensify ERP. In a randomized trial of adults still at least moderately ill after 12 or more weeks on an SRI, adding 17 sessions of ERP improved symptoms more than adding risperidone or placebo. Response rates were 80% with ERP, 23% with risperidone and 15% with placebo, and risperidone did not differ significantly from placebo. The authors concluded ERP should be offered before antipsychotics.
3. Switch to a different SSRI or to clomipramine. UK guidance offers either after an inadequate response, and considers clomipramine after at least one SSRI has failed or was not tolerated.
4. Consider augmentation. In a meta-analysis of 12 placebo-controlled trials, about one-third of SRI-resistant patients benefited when an antipsychotic was added. Only risperidone showed significant efficacy; quetiapine and olanzapine did not. No antipsychotic is FDA-approved for OCD, so this is off-label use.
5. Other options after several full trials include further CBT or cognitive therapy, and combining clomipramine with citalopram, which is also off-label.

Any change to a dose or a medicine is a decision for your prescriber. Stopping an SSRI or clomipramine abruptly can cause discontinuation symptoms, and UK guidance advises reducing gradually. If you are weighing a different approach, a psychiatric medication second opinion can help.

Deep TMS and other options for treatment-resistant OCD

In August 2018 the FDA permitted marketing of the BrainsWay deep transcranial magnetic stimulation system for OCD through its de novo pathway, for patients who had not responded to traditional treatments. In the pivotal trial, 38% of people who received active treatment responded, defined as a reduction of more than 30% on the Yale-Brown Obsessive Compulsive Scale, compared with 11% of those who received sham treatment. The published trial delivered daily sessions for 6 weeks after individualized symptom provocation.

Deep TMS is a general option offered at specialized centers, usually alongside ongoing medication. It is not a first-line treatment, and the evidence base is smaller than for ERP or SSRIs. For people with severe, long-standing OCD, intensive outpatient, residential or specialty programs exist; the guide to levels of psychiatric care explains how they differ.

Special situations that change the plan

Children and teenagers

Fewer medicines are approved for young people. Sertraline is approved for OCD from age 6, fluoxetine is approved from age 7, fluvoxamine's approval included a trial in ages 8 to 17, and clomipramine's included ages 10 to 17; paroxetine is not approved for pediatric use. The boxed warning on suicidal thoughts applies most directly to this age group, so prescribers usually schedule closer follow-up after a start or a dose change. For moderate to severe impairment, UK guidance names CBT including ERP, adapted to the child’s developmental age and involving the family, as the treatment of choice. Parents weighing options can start with child anxiety treatment options, which covers how pediatric CBT is organized.

Pregnancy or planning a pregnancy

Decisions about SSRIs or clomipramine during pregnancy weigh the possible risks of the medicine against the risks of untreated symptoms, and the balance differs from person to person. This is a conversation to have with your prescriber before you stop anything; the guide to psychiatric medications during pregnancy sets out the questions to ask.

OCD with depression or another condition

When depression is significant, an SSRI can address both conditions, since several of the OCD medicines are also approved for major depressive disorder. Some people start with medication and add ERP once their mood improves; the order is worth discussing with both your prescriber and your therapist. If tics, ADHD or an eating disorder are present, the plan usually needs a clinician who can see the whole picture.

Family involvement

Relatives often get drawn into rituals, for example by answering the same reassurance question many times or checking locks on someone's behalf. UK guidance says families or carers should be involved as appropriate at every stage of OCD care, and especially for children and young people. Any change in how relatives respond to rituals works best when it is planned with the therapist and the person who has OCD, rather than imposed suddenly.

What the evidence does not show

• UK guidance lists several SSRIs as initial options without ranking one as best for OCD; choice usually rests on side effects, interactions and past response.
• Supplements and dietary changes are not established OCD treatments, and none is FDA-approved for OCD.
• The higher-dose finding comes from adult trials and does not automatically apply to children.
• Trial results are averages from selected study groups, so how well they apply to you, especially if you also have depression, ADHD or a tic disorder, is a question for your clinician.

Questions to ask your clinician

• Is the therapy you are recommending ERP specifically, and how much practice between sessions should I expect?
• What dose would count as an adequate trial for OCD with this medicine, and how long before we judge it?
• If I improve only partly, would you add ERP, switch, or augment, and why in that order?
• What side effects should prompt me to call before my next appointment?
• How will we measure progress? Many clinicians use the Yale-Brown Obsessive Compulsive Scale.

How this practice approaches OCD care

Dr. Lewis's team includes psychiatrists, psychiatric nurse practitioners, psychotherapists and a health and wellness coach. Ongoing care at the practice includes psychiatric evaluation and medication management and psychotherapy, including CBT and DBT. You can read the practice's overview of OCD treatment and integrative psychiatry care or its medication management page, and start with the Patient Interest Form. To check whether ERP is available with a specific therapist, ask on the free discovery call. For the full set of decision guides, see Choosing Psychiatric Care.

Frequently asked questions

How long does it take for an SSRI to work for OCD?

UK guidance notes that OCD medication commonly takes up to 12 weeks to show its effect, longer than many people expect from depression treatment. Your prescriber will judge the trial on both time and dose.

Is ERP better than medication for OCD?

Both are first-line. In one well-known trial, intensive ERP outperformed clomipramine alone and did about as well as the combination. UK guidance treats an SSRI and more intensive ERP as comparably effective for moderate impairment and recommends combining them when OCD is severe.

Can I do ERP while taking medication?

Yes. Combining the two is recommended for severe OCD, and adding ERP to an SSRI improved results more than adding an antipsychotic in a randomized trial.

Are antipsychotics approved for OCD?

No. The FDA-approved OCD medications are fluoxetine, fluvoxamine, sertraline, paroxetine and clomipramine. Antipsychotics such as risperidone are sometimes added off-label when an SSRI has not been enough.

Why might my OCD dose be higher than a typical depression dose?

A meta-analysis of adult trials found higher SSRI doses were more effective for OCD than lower doses, unlike depression, though they also caused more side-effect dropouts. Your prescriber sets the dose.

What if I can't find an ERP-trained therapist right away?

UK guidance treats an SSRI as a reasonable first choice for moderate OCD, so starting medication while you look is one option to discuss with a prescriber. ERP can be added later; in one trial, adding it to an SSRI helped more than adding an antipsychotic.

Sources

1. Obsessive-compulsive disorder and body dysmorphic disorder: treatment (CG31), Recommendations. National Institute for Health and Care Excellence, 2005. nice.org.uk
2. PROZAC (fluoxetine capsules) prescribing information. DailyMed / U.S. FDA label, 2026. dailymed.nlm.nih.gov
3. Fluvoxamine maleate tablets prescribing information. DailyMed / U.S. FDA label, 2025. dailymed.nlm.nih.gov
4. ZOLOFT (sertraline) prescribing information. DailyMed / U.S. FDA label, 2023. dailymed.nlm.nih.gov
5. PAXIL (paroxetine) prescribing information. DailyMed / U.S. FDA label, 2026. dailymed.nlm.nih.gov
6. Anafranil (clomipramine hydrochloride) capsules prescribing information. DailyMed / U.S. FDA label, 2024. dailymed.nlm.nih.gov
7. Randomized, placebo-controlled trial of exposure and ritual prevention, clomipramine, and their combination in the treatment of obsessive-compulsive disorder. American Journal of Psychiatry (PubMed 15625214), 2005. pubmed.ncbi.nlm.nih.gov
8. Meta-analysis of the dose-response relationship of SSRI in obsessive-compulsive disorder. Molecular Psychiatry (PubMed 19468281), 2010. pubmed.ncbi.nlm.nih.gov
9. Antipsychotic augmentation of serotonin reuptake inhibitors in treatment-resistant obsessive-compulsive disorder: a meta-analysis of double-blind, randomized, placebo-controlled trials. International Journal of Neuropsychopharmacology (PubMed 22932229), 2013. pubmed.ncbi.nlm.nih.gov
10. Cognitive-behavioral therapy vs risperidone for augmenting serotonin reuptake inhibitors in obsessive-compulsive disorder: a randomized clinical trial. JAMA Psychiatry (PubMed 24026523), 2013. pubmed.ncbi.nlm.nih.gov
11. FDA permits marketing of transcranial magnetic stimulation for treatment of obsessive compulsive disorder. U.S. Food and Drug Administration, 2018. fda.gov
12. Efficacy and Safety of Deep Transcranial Magnetic Stimulation for Obsessive-Compulsive Disorder: A Prospective Multicenter Randomized Double-Blind Placebo-Controlled Trial. American Journal of Psychiatry (PubMed 31109199), 2019. pubmed.ncbi.nlm.nih.gov
13. Adding psychotherapy to antidepressant medication in depression and anxiety disorders: a meta-analysis. World Psychiatry (PubMed 24497254), 2014. pubmed.ncbi.nlm.nih.gov

Educational information, not medical, tax or legal advice. Decisions about diagnosis, medication and treatment belong with a clinician who knows your history.