Treating obsessive-compulsive disorder

Published: March, 2009

Options include medication, psychotherapy, surgery, and deep brain stimulation.

Obsessive-compulsive disorder (OCD), which affects 2% to 3% of people worldwide, often causes suffering for years before it is treated correctly — both because of delays in diagnosis and because patients may be reluctant to seek help. One review estimated that, on average, patients with OCD take more than nine years to be diagnosed correctly, and 17 years to receive appropriate care.

Although OCD tends to be a chronic condition, with symptoms that flare up and subside over a patient's lifetime, effective help is available. Only about 10% of patients recover completely, but 50% improve with treatment.

Challenges in diagnosis

As the name implies, OCD is characterized by two hallmark symptoms. Obsessions are recurring and disturbing thoughts, impulses, or images that cause significant anxiety or distress. Compulsions are feelings of being driven to repeat behaviors, usually following rigid rules (such as washing hands multiple times after each meal). When these symptoms interfere with work, social activities, and personal relationships, it is time to consider treatment.

It may be difficult to distinguish OCD from other psychiatric disorders with similar symptoms. In its updated guidelines, the American Psychiatric Association (APA) provides sample screening questions to better identify patients with OCD, as well as suggestions for differentiating OCD from other disorders. For example, obsessions in OCD typically involve an object or person other than the self, such as a fear of becoming contaminated or acting aggressively toward someone else, whereas ruminations in depression usually involve self-criticism or guilt about the past — and they are not usually accompanied by compulsive rituals. Obsessions in OCD usually are clearly defined, while those in generalized anxiety disorder may be vaguely preoccupied, for example, with nagging worries about bad outcomes.

Initial treatments

For initial treatment of OCD, the APA recommends cognitive behavioral therapy, drug therapy with selective serotonin reuptake inhibitors (SSRIs), or a combination of the two.

Behavioral treatment. The most effective behavioral treatment for OCD is exposure and response prevention. In this therapy, patients encounter the source of their obsession repeatedly and learn ways to stop performing associated rituals until they are able to resist these compulsions. For a patient who avoids using silverware because it might be contaminated with germs, a clinician might direct the patient to pick up a fork and imagine the microorganisms — but to delay washing his hands.

Behavioral treatment alone may be an option for patients with mild symptoms of OCD or for those who don't want to take medications. It may take three to five months of weekly sessions to achieve results. The goal is to gradually extinguish a conditioned behavior pattern. Little evidence supports the use of cognitive therapies unless they include a behavioral component.

SSRIs. Drug treatment may be tried first if behavioral therapy isn't available or convenient, or if the patient's symptoms are severe. Although the FDA has approved the tricyclic antidepressant clomipramine (Anafranil) for treatment of OCD, this medication may cause anticholinergic side effects such as dry mouth, blurred vision, constipation, delayed urination, and a rapid heartbeat. The APA therefore recommends starting with one of the SSRIs because their side effects may be better tolerated.

All of the SSRIs are equally effective, although individual patients may respond better to one than another, and it may take some trial and error to determine which one is best. Generally 40% to 60% of patients with OCD will experience at least a partial reduction in symptoms after treatment with an SSRI. However, many continue to have residual symptoms.

To treat OCD, SSRI doses are usually higher than those used for depression. It also takes longer for these medications to alleviate symptoms of OCD. While patients with major depression might take two to six weeks to respond to an SSRI, patients with OCD typically take 10 to 12 weeks to respond.

The most common side effects of SSRIs are gastrointestinal distress, restlessness, insomnia, and sexual dysfunction (such as reduced libido, erectile dysfunction, and inability to reach orgasm). Drug choice may also be swayed by a patient's health profile and use of other medications. Paroxetine (Paxil), for example, is the SSRI that is most likely to cause weight gain and anticholinergic side effects; as such, the APA recommends against it as a first choice for patients who are obese, have type 2 diabetes, or suffer from urinary hesitancy or constipation.

Maintenance therapy. Many patients successfully treated for OCD will benefit from continuing medication indefinitely. A few medication discontinuation trials have been conducted in OCD patients, and most have found high relapse rates after SSRI withdrawal. It's possible that lower doses can be used during maintenance treatment, but this is not clear. One way to reduce relapse is to combine drug treatment with exposure and response prevention therapy, so that when the drugs are withdrawn patients are better able to cope with environmental triggers.

When to consider a change. As a general rule, the APA recommends that clinicians and patients give the initial treatment enough time to work before considering a change. If 13 to 20 weekly sessions of behavioral therapy — or 10 to 12 weeks of drug treatment — have not sufficiently alleviated symptoms, consider a new strategy.

Additional treatment strategies

For patients whose symptoms have only been partially relieved by a first treatment, augmenting that treatment may be more effective than switching to a new one. Time makes this strategy a prudent one. Switching to another drug as monotherapy may take another 10 to 12 weeks to show results. Augmenting an SSRI with some other medication, on the other hand, can produce effects within four weeks.

Augmentation options. One option is augmenting an SSRI with an antipsychotic. Drug choices include first- or second-generation antipsychotics, but the evidence is stronger for the newer drugs. Studies indicate that 40% to 55% of patients with OCD, after failing to respond to a first treatment, do improve when an antipsychotic is added to an SSRI — although residual symptoms may remain. If one antipsychotic doesn't work, the APA recommends trying another.

Bear in mind that an antipsychotic used to augment OCD treatment should be prescribed at the lower end of the dosing range. At high doses — or when prescribed alone — antipsychotics may worsen OCD symptoms.

Another option is to augment an SSRI with clomipramine. However, several SSRIs are metabolized by the same cytochrome P450 enzymes as clomipramine, and therefore may interact in a way that can cause heart problems in some patients. Before prescribing clomipramine with an SSRI, the APA recommends considering a screening electrocardiogram in patients who are older than 40 or who have heart disease. It may also make sense to avoid prescribing fluvoxamine (Luvox), fluoxetine (Prozac), and paroxetine, SSRIs that increase blood levels of clomipramine.

Switching to a new drug. If treatment with an SSRI does not work, consider switching to another SSRI or another type of drug. The APA estimates that 50% of patients with OCD who do not respond to one SSRI will respond to another one. However, the response rate may decrease as a third or fourth SSRI is tried. Other less well-studied options include switching to a non-SSRI antidepressant, such as venlafaxine (Effexor) or mirtazapine (Remeron).

Neurosurgery or brain stimulation

Roughly 10% of patients with OCD will get worse in spite of treatment. Patients who suffer severe and incapacitating symptoms despite multiple medication trials may be eligible for brain surgery or deep brain stimulation. (Electroconvulsive therapy and transcranial magnetic stimulation have not proven effective in treatment of OCD.)

Both surgery and deep brain stimulation remain investigational, partly because researchers are still trying to identify the proper brain targets. These options are usually held in reserve for patients with the most treatment-resistant OCD. Typically patients who opt for these strategies have debilitating symptoms and have tried other treatments for 10 years without success.

Ablation. Neurosurgery for OCD involves the destruction (ablation) of small amounts of brain tissue. Procedures include anterior capsulotomy, limbic leucotomy, cingulotomy, and gamma-knife radiosurgery. These approaches differ in the precise brain area targeted and the amount of tissue destroyed. Studies report that 35% to 50% of patients with OCD who undergo neurosurgery improve. Risks include seizures, personality changes, and more transient side effects associated with surgery and anesthesia.

Deep brain stimulation. In this technique, a surgeon implants electrodes in the brain and connects them to a small electrical generator in the chest. Deep brain stimulation does not permanently destroy neural tissue, as surgery does; instead, it uses electricity to modulate the transmission of brain signals.

It's not clear why this technique works, and there is no consensus about the right targets — although researchers are working to clarify both issues. In 2008, an international collaboration of four institutions reported results of deep brain stimulation of the best-studied brain area — the junction of the ventral capsule and ventral striatum — in 26 patients. Although most patients continued to have residual symptoms, their scores on clinical instruments such as the Yale-Brown Obsessive Compulsive Scale indicated that on average, OCD intensity declined from severe to moderately severe. As the surgeons performed more operations and better refined the brain target, more patients improved: one-third of the first group of patients improved, compared with 70% in both the second and third groups.

As researchers learn more about the brain basis of OCD, they hope to target brain regions more precisely, to achieve better results.

Jenike MA. "Clinical Practice: Obsessive-Compulsive Disorder," New England Journal of Medicine (Jan. 15, 2004): Vol. 350, No. 3, pp. 259–65.

Koran LM, et al. "Practice Guideline for the Treatment of Patients with Obsessive-Compulsive Disorder," American Journal of Psychiatry (July 2007): Vol. 164, No. 7, Suppl., pp. 5–53.

For more references, please see

As a service to our readers, Harvard Health Publishing provides access to our library of archived content. Please note the date of last review or update on all articles. No content on this site, regardless of date, should ever be used as a substitute for direct medical advice from your doctor or other qualified clinician.