Cognitive Behavioral Treatment for Obsessive-Compulsive Disorder

August 1, 2006
Bradley C. Riemann, PhD

Patients with OCD generally respond best to cognitive-behavioral therapy (CBT). At the core of the behavior therapy program is a technique known as exposure and ritual prevention (ERP).

August 2006, Vol. XXIII, No. 9

Obsessive-compulsive disorder (OCD) is a common and debilitating condition. In many cases, it can come to dominate a person's behavior and cognitive processes, creating great anxiety. It typically affects all aspects of an individual's life, including school performance, occupational responsibilities, and family and social interactions.

In addition, our experience confirms that about 85% of persons with OCD report having some level of depression; Demal and associates1 found a similar rate, 79%, in their retrospective study. This is understandable, considering that persons with OCD tend to feel hopeless and helpless about controlling their obsessions and compulsions. Patients generally have insight into their illness and are aware that their actions and thoughts may be considered bizarre. As a result, they may hide their symptoms for fear of embarrassment. Some may not disclose their OCD symptoms even to their mental health provider for fear that they might be considered “crazy,” and as a result might lose their job or have their children taken away from them.

While medications can reduce symptoms of OCD by about one third,2 this is seldom sufficient to adequately control the disorder. Patients with OCD generally respond best to cognitive-behavioral therapy (CBT). While the term “CBT” is often used interchangeably with cognitive therapy and behavior therapy, there are distinctions between these interventions. The CBT programs used at my institution are composed of about 85% behavior therapy and 15% cognitive therapy. These percentages are adjusted slightly, according to an individual's profile and needs. Most patients we treat are already taking medication and have received whatever benefit medication can provide before they are admitted to one of our OCD programs.

Behavior therapy

At the core of the behavior therapy program is a technique known as exposure and ritual prevention (ERP).3 ERP is based on the process of habituation, in which individuals habituate to stimuli they don't like. We all experience this process in our daily lives. An example would be jumping into cold water. The water feels cold initially, but gradually feels warmer as our body becomes habituated to it. A person can become habituated to smells, sounds, physical sensations, and emotions. In persons with anxiety related to OCD, habituation is a decrease in anxiety that occurs with the simple passage of time. It is essential that nothing else, such as relaxation therapy, be added because other treatments could become compulsions in patients who have OCD. The elements of a behavior therapy program are outlined in Table 1.

To implement the ERP technique, the patient must be exposed to or placed in the feared situation. Simply doing this, however, does not guarantee that he or she is going to become habituated and experience relief from anxiety. For exposure to be effective, it must be prolonged, repetitive, and graduated.

• Exposure and ritual prevention is the key element

Based on the principle of habituation
Habituation: the decrease in anxiety experienced with the passage of time

Needs to be prolonged enough to lead to within-trial habituation (at least 50% reduction in anxiety)
Needs to be repetitive enough to lead to between-trial habituation (until situation causes minimal to no anxiety)
Needs to be graduated (increases compliance)

Replace the ritual with habituation as a way of controlling anxiety

Prolonged exposure

Continuing the water analogy, quickly jumping in and out of the water does not decrease your body's sensitivity to the cold. You must remain in the water long enough for your body to become habituated to it. The person who has a fear of contamination, for example, must touch the feared objects with all fingers of both hands, back and front. He should keep touching the object in this way until he reports his anxiety decreasing by at least half. You can use a scale of 0 (no anxiety) to 7 (greatest anxiety) to help in measuring this. The amount of time it takes a person to experience a 50% reduction in anxiety depends on many variables. However, if the exposure is done in a graduated fashion, this should be a matter of minutes, not hours. The time it takes for the patient's anxiety to halve is called an exposure trial. The quantitative reduction in anxiety-going from 4 to 2 on the scale, for example-is called within-trial habituation.

Repetitive exposure

For the patient's anxiety to permanently diminish, exposure to the stimulus or situation must be repeated until it is no longer bothersome. Most patients will reach this goal within 8 to 12 exposure trials (each trial being a single exposure) as long as the exposure is being done in a prolonged, gradual fashion.

Graduated exposure

Graduating the exposure to the stimulus helps improve compliance with the exposure program. Graduation should be done in a hierarchic manner. I find that patients exposed to a challenging but manageable range of anxiety-a level of 3 on the 7-point scale-tolerate treatment better than those who are plunged into treatment at a potentially overwhelming anxiety level of 4 or higher. As a result, we experience a lower dropout and refusal rate than has been reported in the literature for behavior therapy. With ERP, a “slow as you go” approach ultimately takes the same time as (or even less time than) a higherintensity approach because the patient experiences more rapid within-trial and between-trial habituation.

Imaginable exposure

In some cases, ERP can be supplemented with another technique called imaginable exposure. This is very helpful when the patient cannot physically repeat ERP often enough for it to be effective or cannot conduct an exposure in real life (eg, a person with obsessions about harming himself or others). You can ask the patient to imagine what it feels like to be in the feared situation. It is remarkable how anxious some OCD patients become just imagining their feared situations. Imaginal exposure can also play a key role in ensuring a graduated approach to the exposure. Many times imaginal scripts are used with endless loop audio tapes to provide prolonged, repetitive exposures.

Ritual prevention

The other essential component of ERP treatment is ritual prevention. This involves blocking the ritual that the patient typically performs before, during, and after exposure to feared objects/situations so that habituation can occur. The ritual is replaced with habituation as a way of controlling anxiety.3 In the case of a compulsive hand-washer, ritual prevention would involve not washing hands at all unless unwashed hands pose a health or safety hazard. Someone who wears contact lenses, for example, would be allowed to wash his hands just before inserting the lens, but not at any other time. He also would need to “recontaminate” his hands immediately after inserting his lenses. During ritual prevention, it is essential to take the patient beyond the normal limits (for instance, not washing hands after using the bathroom, or having him dry off after a shower with a towel that he had used previously and did not consider perfectly clean) to get him to ultimately arrive at normal behavior.

Treatment Steps

The steps involved in ERP are listed in Table 2. At the core of the treatment program is the hierarchy of exposure exercises that constitute the patient's treatment plan. It is important to be very specific when constructing the exercises; otherwise, the patient will simply do whatever is easiest. Take, for example, the patient who is afraid to use public telephones. His exercise plan should specify which public phones to use-for example, those near the bathrooms in restaurants, not just those near the front reservation desk.

The hierarchies used at my institution contain dozens of exposure exercises. These exercises are grouped according to the level of anxiety they generate for the patient, on the graduated scale of 0 to 7 that I described earlier. I start my patients on the exercises that provoke an anxiety level of about 3 on the scale.

The treatment schedule varies with the severity of OCD. Patients who score roughly between 18 and 23 on the Yale- Brown Obsessive Compulsive Scale (Y-BOCS)-the gold standard for OCD assessment-generally respond well to 1 to 2 hours of CBT a week. Persons who score roughly between 23 and 29 generally need 4 or 5 visits a week, for about 3 to 4 weeks. At the OCD residential program at Rogers Memorial Hospital, which serves those with a YBOCS score of about 30 or higher, patients get 50 hours of treatment a week, including both therapist-aided and self-exposure.

Persons who hoard deserve special comment, because their treatment plan differs slightly from that for others with OCD. Leaders in the hoarding field are beginning to consider that hoarders are different from other subtypes of OCD patients and may, in fact, not have OCD. Hoarders tend to have less insight into their illness and their thoughts related to their belongings appear less “unwanted.” As a result, they receive more cognitive therapy than other persons with OCD typically would. Also, if the patient hoards, therapy almost always needs to be done intensely and home visits become a must.

Patients who have severe, treatmentrefractory OCD that does not respond to intensive ERP either alone or in combination with medications are potential candidates for referral to a center that offers various forms of neurosurgery. One new, alternative type of neurosurgery is deep brain stimulation. This procedure is done at various US sites, including Massachusetts General Hospital, Brown University, and the University of Florida. About one third of patients appear to partially respond to neurosurgical strategies.2 In some cases, OCD symptoms can be reduced enough to allow the patient to benefit from ERP therapy.

Cognitive restructuring

Most patients with OCD benefit from the inclusion of cognitive therapy in their treatment program. As mentioned earlier, the general balance sought in our programs is 85% ERP and 15% cognitive restructuring.

During cognitive restructuring, the therapist tries to identify and correct errors in the patient's thinking that generate anxiety. Patients with OCD usually have 2 types of erroneous thinking. One is a probability overestimation error, in which they overestimate the probability of a bad event, such as becoming infected with HIV by touching a doorknob. The second is a catastrophe error, in which the patient magnifies the severity of a likely event, such as observing someone using the toilet and not washing his hands. Persons with OCD can best learn to identify and self-correct these errors in thought by going through a series of steps that we call “thought challenging.” The steps include “evidence identification” and completion of worksheets to correct both types of errors.

ERP in perspective

ERP treatment has been proved effective in clinical studies. Foa and colleagues4 performed a meta-analysis of 12 studies involving 330 patients with OCD who were treated with ERP therapy. In all, 83% of these patients were greatly improved.

Greist5 performed a different metaanalysis, examining behavioral therapy and Y-BOCS scores in 294 patients enrolled in 18 studies. Y-BOCS scores were compared before and after treatment with a serotonin reuptake inhibitor (SRI). The average decrease in scores on the Y-BOCS from pre- to post-therapy in persons treated with behavior therapy alone was 11.8, a decrease that is statistically significant and significantly greater than the average reduction on the Y-BOCS achieved with an SRI, which was 7.5.

Relapse rates are very low with ERP therapy. In reporting on 16 studies involving 376 patients, Foa and Kozak6 found that 76% had improved at followup. We have found that patients' completion of their exposure hierarchy is key to a low relapse rate.

ERP is an effective and robust treatment. Patients respond quickly, with many noting improvement after the first week. The only adverse effect is an increase in anxiety during treatment, which can be managed by graduating the exposure to the fearful stimulus appropriately. The disadvantages are that ERP is hard work and time-consuming for patients and therapists alike. However, the hours a day we ask patients to commit to treatment are hours they are already spending ritualizing. It can also be difficult to find therapists who are qualified in this kind of treatment. Based on its efficacy and low relapse rate, however, ERP is currently considered to be the first-line treatment for OCD.7


Initial evaluation (1 hour)
Confirm diagnosis
Identify problem areas (touching door knobs,
shaking hands, etc)
Assess for comorbid diagnoses
Educate patient and family about OCD
and treatment options

Detailed assessment (4 - 5 hours)*Patient completes Y-BOCS checklist and severity rating scale† Generate exposure exercises Patient rates each exercise on a scale of 0 to 7, depending on perceived difficulty Create exposure hierarchy Decide how intensively to deliver BT
Treatment (conducting the hierarchy)Mild to moderate cases can usually be managed in a weekly outpatient setting (little need for therapist-aided exposure) Moderate to severe cases typically need more intensive treatment (need more therapist-aided exposure) Multiple visits per week, each lasting several hours, for 3 weeks (for moderate to severe OCD) Homework for the patient to complete outside of office visits
OCD, obsessive-compulsive disorder; Y-BOCS, Yale-Brown Obsessive Compulsive Scale; BT, behavior therapy. *Detailed assessment is performed over several office visits. †Severity rating scale can be completed before, after, and every 2 weeks during treatment.

For more information on obsessive-compulsive disorder, visit

Dr Riemann is clinical director of the Obsessive-Compulsive Disorder Center at Rogers Memorial Hospital, Oconomowoc, Wis, and is on the scientific advisory board of the Obsessive Compulsive Foundation. He reports no conflicts of interest regarding the subject matter of this article.


1. Demal U, Lenz G, Mayrhofer A, et al. Obsessivecompulsive disorder and depression. A retrospective study on course and interaction. Psychopathology. 1993;26:145-150.
2. Piggott TA, Seay S. Biological treatments for OCD: literature review. In: Swinson RP, Antony MM, Richter MA, eds. Obsessive Compulsive Disorder. Theory, Research, and Treatment. New York: Guilford Press; 1998:298-326.
3. Meyer V. Modification of expectations in cases with obsessional rituals. Behav Res Ther. 1966;4:273-280.
4. Foa EB, Franklin ME, Perry KJ, Herbert JD. Cognitive biases in generalized social phobia. J Abnorm Psychol. 1996;105:433-439.
5. Greist JH. New developments in behaviour therapy for obsessive-compulsive disorder. Int Clin Psychopharmacol. 1996;11(suppl 5):63-73.
6. Foa EB, Kozak MJ. Psychological treatment for obsessive-compulsive disorders. In: Mavissakalian MR, Prien RF, eds. Long-Term Treatments of Anxiety Disorders. Washington, DC: American Psychiatric Press; 1996:285-309.
7. Foa EB, Franklin ME, Kozak MJ. Psychosocial treatments for OCD: Literature review. In: Swinson RP, Antony MM, Richter MA, eds. Obsessive Compulsive Disorder. Theory, Research, and Treatment. New York: Guilford Press; 1998.