Cognitive-Behavioral Therapy for Obsessive-Compulsive Disorder: 2021 Update

Over the years, CBT has come to represent a blending of techniques to modify cognitions and behavior. In traditional cognitive therapy, faulty thoughts and beliefs are targeted (2–4). In behaviorally focused treatment, termed “exposure and response prevention” (ERP), patients engage in exposure to stimuli that trigger obsessions and are encouraged through various techniques to refrain from performing compulsions (5).

The American Psychiatric Association (APA) guidelines (6) support the use of cognitive therapy and ERP but note that there is more evidence for the efficacy of ERP. A meta-analysis (7) found that cognitive therapy and ERP produce very similar effect sizes, although a larger amount of data exist in support of ERP. This finding led to the recommendation that ERP be considered as the first-line psychotherapeutic treatment for OCD. In practice, we recommend a combination of cognitive and behavioral techniques.

A discussion of the superiority of CBT monotherapy compared with progressive muscle relaxation, anxiety management training, and systematic relaxation was provided in our 2015 review (1). Head-to-head comparator trials of CBT versus medication are few. A recent meta-analysis (8) found there was insufficient evidence to determine the superiority of CBT-ERP compared with medication. Although comparison data are few, it is important to note that ERP may reduce risk for relapse once medication is discontinued (1).

The APA practice guidelines (6) state that modest evidence exists to support the use of CBT in augmentation of selective serotonin reuptake inhibitors (SSRIs) (1). CBT in combination with medication has been found to have greater efficacy than medication or CBT alone in several studies (1). APA treatment guidelines (6) recommend combination therapy for co-occurring disorders that are responsive to serotonin reuptake inhibitors, for severe OCD symptoms, and for patients who wish to limit the duration of pharmacotherapy. A recent meta-analysis (8) concurred with the recommendation to provide concurrent CBT and medication in severe cases of OCD.

Recently, acceptance and commitment therapy (ACT) has been studied as an alternative or complement to CBT-ERP. A full discussion of ACT is beyond the scope of this article. In brief, rather than repeated exposures to anxiety-provoking stimuli, ACT focuses on strategies to encourage engagement in life activities despite the presence of obsessions. ACT encourages patients to engage in behavior consistent with their life values and to willingly approach feared situations in order to achieve important life goals. Multiple studies of ACT (9–14) have shown low drop-out rates, positive ratings by participants, and decreases in OCD symptoms and depression.

Because patients are often reluctant to participate in treatment of OCD, motivational interviewing has been studied to enhance engagement. Research examining motivational interviewing for OCD has been limited to small trials. Evidence suggests this technique can be useful to encourage patients to enter psychotherapy and to maintain participation in psychotherapy (15), although the duration of benefit is unclear (16).

The following case example is of a fictional character incorporating elements of personal histories from multiple patients. This case is not based on an individual.

Ms. H, a 46-year-old woman, presented for psychiatric evaluation of possible OCD. Ms. H had been hospitalized 2 months prior for a severe skin infection. Ms. H reported a history of anxiety beginning in childhood. When she was a child, she worried that someone might break into their home, and she repeatedly asked her parents to check that the doors and windows were locked. Following the skin infection and hospitalization 2 months ago, Ms. H experienced a significant increase in anxiety. She complained of a persistent fear of contracting a second skin infection or some other life-threatening illness. Ms. H’s fear of contracting an illness had become so severe that she rarely ventured out of her apartment, because she feared coming in contact with germs. She developed strict rituals for cleaning utensils and washing her hands. Food could be eaten from only freshly opened packages. Ms. H reported that these rituals and rules provided her with reassurance that she was less likely to become sick. Ms. H entered treatment at the urging of her sister. During the interview, the clinician noticed that Ms. H’s hands were red and abraded. The clinician also noted that Ms. H frequently asked her sister if she thought there were germs in the office. Ms. H’s sister would calm her by repeatedly stating “I’m sure you won’t get sick.”

Ms. H elected to engage in a trial of CBT. During the first several sessions, the clinician developed rapport with Ms. H and helped her to identify the thoughts, emotions, and behaviors that developed and maintained the OCD symptoms. Avoidance of “contaminated objects,” seeking reassurance from family members, washing her hands, and eating from only prepackaged food were identified as the primary compulsions. Mental compulsions of “cleanliness checklists” were also identified. Ms. H’s score on the Yale-Brown Obsessive Compulsive Scale (YBOCS) prior to treatment was 25 (indicating moderate-to-severe illness).

An exposure hierarchy was constructed (Table 1). For the initial exposure session, Ms. H placed her hand on a countertop. The initial subjective units of distress rating during the exposure was 4 out of 10. With encouragement from the clinician, Ms. H engaged in this exposure for 15 minutes and refrained from washing her hands afterward. The clinician was careful to provide encouragement (e.g., “You can do this, it’s important to you”) rather than reassurance (e.g., “I cleaned the counter this morning. You won’t get sick”). At the end of the exposure session, the distress rating was 2 of 10. Ms. H was instructed to continue this exposure exercise at home. The family was instructed to provide encouragement rather than reassurance at home.

During later exposure sessions, the clinician recognized that Ms. H appeared somewhat distracted during more difficult exposures. On discussing this with Ms. H, it was discovered Ms. H had been using mental compulsions of “mental checklists” during these exposure activities in an attempt to reduce her anxiety. Ms. H agreed the “mental checklists” were simply maintaining the OCD and decided to mentally repeat the phrase, “I want to do scary things so I can beat OCD,” whenever the urge to engage in mental compulsions arose.

After completing several exposure sessions, Ms. H and the clinician began cognitive restructuring. Ms. H was introduced to the thought record technique. Initially, the clinician elicited thoughts from Ms. H immediately following their exposure sessions. After Ms. H understood the technique, she was instructed to complete thought records at home. An example of Ms. H’s thought record is shown in Table 2. Ms. H recorded her initial thoughts when confronting feared situations and a more helpful alternative thought. Ms. H was provided psychoeducation regarding the disadvantages of using reassurance as an alternative thought (e.g., “I’m sure I can wash my hands enough to get rid of any germs”) and was instructed to instead accept and embrace the fears (e.g., “There is always a possibility I could get sick. It’s going to be tough to completely avoid the risk if I want to live my life.”). Ms. H reported that her ability to cope with the feared thoughts and situations gradually improved with the use of the thought record. She noted that “reassurance thoughts” seemed to work her “into a frenzy,” because she found it difficult to definitively conclude that nothing bad would happen to her in the future. In contrast, the thoughts that were more accepting of danger provided her with a feeling of calm, because she was no longer expending mental energy on such a difficult task.

On further discussion with the clinician, it was identified that Ms. H’s ultimate fear was that she would bring home germs that would get her family members sick. She had two teenage sons and thought how horrible it would be if they had to spend a significant portion of their young lives in the hospital. Worse yet, if they were to die in their teenage years, they would be robbed of the majority of their lives. Because an in vivo exposure could not readily be designed, the clinician assisted Ms. H in creating an imaginal exposure script for this feared event. Ms. H recorded a detailed story of her sons contracting a mysterious illness and how Ms. H may have been responsible for transmitting this illness to her children. Ms. H then listened to this recording for the remainder of the sessions. Ms. H also was instructed to listen to the recording for 1 hour every day between sessions. With Ms. H’s consent, her husband and sons were instructed not to provide any reassurance (e.g., “No, I don’t feel like I have a fever today.”) to Ms. H between sessions. Instead, family members would respond by saying, “It sounds like you are looking for reassurance. We agreed that giving you reassurance doesn’t help. How about we go for a walk or watch television instead?”

Ms. H engaged in 20 sessions of CBT. She was adherent to the between-session homework assignments, and her family members refrained from accommodating behaviors. At the end of the 20th session of treatment, her YBOCS score was 5 (indicating remission). Family members reported that Ms. H appeared happier. Ms. H reported she indeed felt better and had started attending college part-time. Ms. H agreed to continue the daily exposure sessions at home. Ms. H gradually reduced the frequency of sessions with her clinician to once every 3 months and was able to maintain remission of OCD.

Assessment of OCD begins with accurate diagnosis. The DSM-5 diagnostic criteria for OCD is covered in the this issue of Focus. Once the diagnosis of OCD has been made, the clinician can begin to determine the underlying behavioral contingencies and maladaptive cognitions in order to develop a treatment plan.

To the outside observer, the connection between the obsessions and compulsions may appear to be an unreasonable appraisal of the true likelihood of danger. This unrealistic appraisal is often the case, but need not be present for the diagnosis of OCD. The degree to which the appraisal of threat is recognized by the patient as unreasonable is referred to as the degree of insight possessed by the patient (which is one of the newly included DSM-5 specifiers).

Regardless of the degree of insight, distorted cognitions are held by patients with OCD and are a second focus of treatment planning. Common cognitions may be “All countertops contain germs. If I touch them, I or my family will get sick.” A major model of the cognitive process in OCD posits that there are five dysfunctional assumptions held by patients with OCD: thoughts are the same as actions, directly causing harm is equivalent to failing to prevent harm, responsibility is not attenuated by outside variables, desire for the feared consequence to occur is equivalent to failure to perform a compulsion, and thoughts must be controlled at all times (3).

After conceptualizing the above stimuli and associated cognitions, the provider is encouraged to begin ERP concurrently while continuing to provide psychoeducation about OCD (17). Cognitive restructuring can be used after ERP is initiated, and techniques for this portion of therapy are presented later in this article.

Historically, ERP was believed to reduce OCD symptoms via habituation. That is, through repeated exposure to a stimulus, the stimulus would create less fear and desire to perform a compulsion as the body physiologically adapted to the stimulus. A newer theory, inhibitory learning, suggests that rather than physiologically habituating to the stimulus, the patient learns new nonthreat associations that compete with the previously conditioned threat associations (18, 19). The inhibitory learning theory would posit that rather than “forgetting” old information, patients learn new information that competes with the previously held beliefs. Over time, the new information outweighs the old information and the fear and the desire to perform compulsions is reduced. The exact mechanisms by which CBT-ERP operates remain unclear, and we hope that future research will provide further insights.

OCD is estimated to affect approximately 1%−3% of children and adolescents, with approximately 50% diagnosed before age 15. According to research, 80% of adult OCD cases have their onset during childhood. The mean age of onset of pediatric OCD ranges from 7.5 to 12.5 years, with an average of 2.5 years between age at onset and treatment. A higher genetic risk of developing pediatric-onset OCD (increased 10 times among first-degree relatives) compared with adult onset (increased two times) is seen. Some etiological evidence indicates a neurodevelopmental perspective, suggesting that early-onset OCD may reflect a specific subtype versus adult-onset OCD (1). Westwell-Roper and Stewart (48) provide an updated review of the etiology and pathogenesis (including pathophysiology, genetics, immune, and environmental factors) for pediatric OCD. Additionally, a section on pediatric acute-onset neuropsychiatric syndrome (PANS) is included.

PANS refers to a subgroup of children who experience OCD symptoms because of an infectious trigger, environmental factors, or other possible triggers (e.g., metabolic disorders) that affect the immune system and result in inflammation of the child’s brain. Because of this inflammation, a sudden onset of severe OCD symptoms or severely restricted food intake occurs that does not follow the typical, more gradual onset of OCD symptoms seen in pediatric OCD. In addition, neuropsychiatric symptoms, such as anxiety, mood disruptions, irritability, frequent urination, academic regressions, and more may be present. The interested reader is referred to Chang et al. (49) for a consensus statement (derived from the First PANS Consensus Conference) on clinically distinct PANS criteria and for expert recommendations on evaluating PANS. These experts outlined the core components of a thorough diagnostic evaluation of suspected PANS cases.

Children with OCD have obsessions and compulsions similar to adults; however, the content differs, as it is more developmentally appropriate. The most common obsessions for children and/or teens are contamination, sexual and somatic obsessions, and excessive scruples or guilt; and the most common compulsions are washing, repeating, checking, and ordering. In addition, young people with OCD may manifest compulsions without distinct or clearly defined obsessions. Early intervention in pediatric OCD is important, because without proper treatment, there is greater likelihood of developing adult mental health issues and lower overall functioning (1).

DSM-5 diagnostic criteria (50) specify that young persons are not required to have insight into the excessive or unreasonable nature of their symptoms. Compared with adults, children tend to have less insight in seeing their obsessions and compulsions as irrational, although a range of insight has been seen. Studies in adult-onset OCD suggest an association between insight and both OCD symptom severity and outcomes. Among youths, poor insight predicted greater FA and severity of symptoms and was associated with diminished CBT response (1). Selles et al. (51), with data from six international pediatric OCD programs, provide support for the finding that lower levels of insight are associated with increased OCD severity (i.e., increased distress and avoidance and decreased symptom resistance). However, it is possible that OCD severity contributes to poorer insight, suggesting the need for more systematic research to understand the nature of these relationships.

Prior to a diagnosis of OCD, a structured, systematic assessment, including family, developmental, and medical history; diagnostic interviews; and a functional analysis with information on the severity and duration of OCD symptoms, is obtained. Common psychological assessment tools for pediatric OCD include the Children’s Yale-Brown Obsessive-Compulsive Scale (CY-BOCS) (52), the child version of the Obsessive–Compulsive Inventory (53), and the Children's Florida Obsessive–Compulsive Inventory (54). Johnco and Storch (55) provide a review of clinical and self-report assessment measures.

Pediatric OCD typically has a chronic and fluctuating course, is associated with impaired social and family functioning and academic difficulties, and influences child development, resulting in reduced quality of life. Thus, a review of the functional consequences of OCD symptoms in home, school, and social environments is needed. Storch and colleagues (56) found that quality of life is inversely related to OCD symptom severity and impairment. Storch and other colleagues (57) found that greater rates of peer victimization among youths with OCD were positively related to loneliness, child-reported depression, parent-reported internalizing and externalizing symptoms of children, and clinician-rated OCD severity. In a pilot study, Negreiros et al. (58) evaluated academic skills among youths with OCD compared with healthy control children. The authors identified significantly higher deficits in mathematical skills among the children with OCD, with no differences in reading or spelling, and without association with symptom severity. These preliminary findings suggest the importance of considering psychoeducational assessments if there is a history of academic difficulties.

Similar to adult OCD, CBT is effective and the treatment of choice for pediatric OCD, even for children as young as age 3. Studies demonstrating positive benefit from CBT for pediatric OCD often include components of psychoeducation, cognitive strategies, and ERP. In a thorough review of evidence-based treatment for pediatric OCD, Grebe et al. (59) conclude that meta-analytic results suggest CBT as the first line of treatment for youths with OCD, including multicomponent programs containing ERP (60–63).

According to clinical guidelines (64), combined CBT and SSRI treatment is recommended for moderate-to-severe cases of pediatric OCD, whereas CBT alone is recommended for children with mild-to-moderate OCD severity. However, in a thorough meta-analysis, Ost et al. (61) (N=42 studies) found large treatment response rates for combination (SSRI and CBT) treatment, although this treatment was not more effective than CBT alone, irrespective of initial severity levels. In comparisons of treatment response rates, CBT alone had higher response rates (70%) compared with SSRIs alone (49%), and both had higher rates than placebo (29%) and the waitlist control condition (13%). With respect to remission rates, Ost et al. found that CBT alone (53%) was more effective than SSRI treatment alone (24%). A second meta-analysis (65) also found no differences between combined treatment and CBT alone. A recent meta-analysis by Uhre and colleagues (66) found comparable effects on remission with CBT and pharmacologic agents. Inconsistencies across these meta-analyses may be accounted for by differences between studies, including methodological differences (e.g., in comparison groups, such as placebo or waitlist). As noted by Grebe et al. (59), further research is needed to examine the comparative efficacy of combined treatment versus CBT or SSRIs alone, particularly whether there are identifiable variables that may help determine when CBT plus SSRI would be beneficial.

There is ample evidence (1) that CBT is an efficacious treatment for pediatric OCD, and that the full CBT protocol is more effective than simply giving instructions on CBT, relaxation strategies, or a waitlist control condition. Further research (61, 66, 67) has supported and extended the finding that CBT is the treatment of choice, with respect to its efficacy, safety, and response durability, for children and adolescents with OCD.

Torp and colleagues’ (68) findings, in the Nordic Long-Term Obsessive-Compulsive Disorder Treatment Study (NordLOTS), suggest that manual-guided CBT can be applied effectively in community child and adolescent mental health clinics. Evidence of the robustness and durability of CBT effects for patients who respond to treatment has been seen in multiple follow-up studies. With this same sample, Højgaard et al. (69) found that at 12 months posttreatment, 78% of participants were in remission (i.e., had CY-BOCS total scores ≤10), whereas 92% had responded to the treatment (i.e., had CY-BOCS total scores≤15). Further treatment gains were found in a 3-year follow-up of the NordLOTS (70).

Skarphedinsson and colleagues (71) randomly assigned the initial nonresponders of the NordLOTS sample to receive extended treatment with continued CBT or pharmacotherapy with sertraline. They found no difference at posttreatment for the CBT or medication groups. Melin and colleagues (70) studied the long-term durability of this sample by following patients for 3 years posttreatment. They found a significant decrease in CY-BOCS total score from baseline to the 3-year follow-up, with a mean decrease of 5.9 points from posttreatment to the 3-year follow-up. Ninety percent of participants were rated as responders and 73% as in remission after 3 years of follow-up. Overall, these results suggest that evidence-based treatment for pediatric OCD has long-term positive effects.

Kircanski and Peris (72), in a pilot RCT, found that the greater the variability of distress during ERP, and when more than one OCD symptom is targeted, the better the treatment outcomes—supporting ERP as the core ingredient of CBT. Despite the effectiveness of ERP, it is underused by clinicians. Common barriers to the use of ERP are therapist-related factors (e.g., level of training and negative beliefs about safety, patient tolerance, and ethicality) and OCD symptoms (hoarding and “taboo” obsessions), suggesting the need to target these areas when training therapists (73). In an effort to address the lack of clinicians effectively trained in pediatric OCD treatment, Piacentini et al. (74) developed competency standards recommended for clinicians delivering CBT for pediatric OCD.

Grebe et al. (59) discussed how therapists can use treatment exercises with children to gain further information about their obsessions and rituals. For example, asking the child “What are you thinking?” when triggered during ERP may help identify the fear and connect it to the feeling of distress. Other considerations in children are a present orientation that makes future symptom relief less motivating, low frustration tolerance, lower attention sustainability, and poor coping abilities. Child-specific CBT that uses age-appropriate metaphors when providing psychoeducation and therapy (e.g., false alarms, bossing back OCD, naming OCD, use of fear thermometers) and incentive programs help address these challenges. Tompkins et al. (75) describes evidence-based CBT treatment adapted for pediatric OCD, including diagnoses, case conceptualization, and treatment planning.

In summary, CBT with an emphasis on ERP is a robust and effective treatment of OCD with or without medication. Some individuals benefit from combined treatment. To speed treatment response or to treat treatment-resistant patients, it may be necessary to consider combined treatment with medication or the use of techniques to enhance the efficacy of CBT-ERP. FA and EE are important topics for the clinician to be aware of in pediatric and adult populations, and evidence suggests that attention to these topics results in improvement of OCD symptoms. It is clear from the ongoing clinical research that family responses (EE and FA) to people with OCD, whether pediatric or adult, clearly impact the course of the disorder as well as response to ERP. The development of a robust family informed ERP package that could be tested for efficacy and longitudinal outcome would be a great contribution to the field.