Combining Acceptance and Commitment Therapy With Exposure and Response Prevention to Enhance Treatment Engagement in Obsessive-Compulsive Disorder
Combining Acceptance and Commitment Therapy With Exposure and Response Prevention to Enhance Treatment Engagement in Obsessive-Compulsive Disorder
The aim of the study was to evaluate whether integrating Acceptance and Commitment Therapy (ACT) with Exposure and Response Prevention (ERP) increases the acceptability, tolerability, and adherence with ERP techniques relative to ERP without ACT. Fifty-eight adults with a DSM-IV diagnosis of Obsessive-Compulsive Disorder (OCD) received 16 twice-weekly sessions (2 hours per session) of either ERP with the inclusion of ACT techniques (ERP+ACT; n = 30) or ERP alone (n = 28). Assessments using interviews, self-report questionnaires, and behavioral observations were conducted at pre- and post-test, and at 6 month follow-up. Specific hypotheses were: 1) Patients receiving ERP+ACT will report greater treatment acceptability, and show higher quantity and quality of completed self-directed ERP assignments, relative to patients receiving standard ERP; 2) Both ERP and ERP+ACT will lead to clinically significant reductions in OCD symptoms from pre- to post-test and from pre-test to follow-up.
BACKGROUND AND SIGNIFICANCE.
The most effective psychological treatment for Obsessive-Compulsive Disorder (OCD) is Exposure and Response Prevention (ERP), which entails confronting obsessional triggers (i.e., exposure) and resisting urges to ritualize (response prevention). Although successful, these techniques are challenging and provoke high levels of anxiety. This may contribute to the fact that between 25% and 50% of patients with access to ERP refuse treatment, drop out prematurely, or do not adhere to the treatment instructions and show attenuated response. Given the effectiveness of ERP techniques, developing ways to make them more tolerable and increase patient adherence, while not compromising therapeutic integrity, is an important next step in OCD treatment research. Recent investigations have addressed whether adding medications, cognitive therapy, or motivational Interviewing to ERP improves outcome and adherence; yet to date, no consensus has emerged regarding the degree to which these combination treatments are more effective than ERP monotherapy. Acceptance and Commitment Therapy (ACT), which has recently been studied for OCD, offers new hope for addressing the tolerability and adherence issues with ERP. Specifically, ACT uses acceptance and mindfulness processes to foster a willingness to experience unwanted inner experiences (e.g., obsessional thoughts, anxiety). These acceptance and mindfulness processes are consistent with ERP, but they are hypothesized to increase engagement in exposure exercises beyond ERP alone. There is evidence for this claim from a randomized trial which demonstrated that ACT, without in-session exposure, had clinical response rates in the 55-65% range at posttreatment and three month follow-up using an intent to treat analysis. Additionally, drop-out and refusal rates were low at 12.2% and treatment acceptability at posttreatment was near the maximum on the scale 4.3 out of 5. In this study, for experimental reasons, ACT did not include in-session ERP. However, the most theoretically and practically consistent use of ACT for OCD is as a context from which to conduct exposure therapy or ERP. While ACT focuses on processes (e.g., acceptance, defusion, values) that are distinct from those involved in ERP (e.g., habituation, cognitive change), an important goal of both treatments is to broaden the patient's engagement with feared stimuli. In fact, ACT procedures have been shown to increase involvement in difficult activities, including participating in exposure therapy for anxiety disorders and willingness to experience unwanted obsessive thoughts. Thus, building ACT techniques into ERP is likely to help patients (a) engage in ERP tasks, (b) confront high levels of anxiety without using escape/avoidance strategies, and (c) resist rituals. The aim of the proposed study is to evaluate whether integrating ACT increases the acceptability, tolerability, and adherence with ERP techniques relative to ERP without ACT.
STUDY DESIGN AND HYPOTHESES.
Fifty-eight adults with a DSM-IV diagnosis of Obsessive-Compulsive Disorder (OCD) received 16 twice-weekly sessions (2 hours per session) of either ERP with the inclusion of ACT techniques (ERP+ACT; n = 30) or ERP alone (n = 28). Assessments using interviews, self-report questionnaires, and behavioral observations were conducted at pre- and post-test, and at 6 month follow-up. Specific hypotheses were: 1) Patients receiving ERP+ACT will report greater treatment acceptability, and show higher quantity and quality of completed self-directed ERP assignments, relative to patients receiving standard ERP; 2) Both ERP and ERP+ACT will lead to clinically significant reductions in OCD symptoms from pre- to post-test and from pre-test to follow-up.
METHODS
3a. Sample Size and Recruitment. The investigators aimed for an intent-to-treat sample of 60 adults across two sites: The Center for Clinical Research at Utah State University (USU) and the Anxiety and Stress Disorders Clinic at the University of North Carolina, Chapel Hill (UNC). Patients were recruited via advertisement of the study and via referrals from primary care and mental health providers familiar with the treatment centers. The USU Center (Twohig) and the UNC Clinic (Abramowitz) are well known research-oriented treatment clinics in their regions. USU is within a 1:30 minute drive of two million people and travel for treatment is common given the rural nature of the area. Also, no other known entity recruits participants with OCD in this area. The Raleigh-Durham-Chapel Hill area of NC is home to over one million people, and the UNC Clinic receives a steady stream of inquiries for treatment. Patient advocacy groups and associations that post research trials on the Internet were contacted to further publicize the study.
3b. Establishing Diagnoses. All diagnoses were established by trained interviewers using the Mini International Neuropsychiatric Interview (MINI).
3c. Randomization to Treatment Groups. Subsequent to screening, 30 individuals were randomly assigned (using a random number generator) to receive ERP+ACT and 28 to receive ERP monotherapy. With the exception of the inclusion of ACT techniques in the ERP+ACT condition, treatment was identical across the two groups (i.e., the number and duration of exposure sessions were the same).
3d. Treatments. ERP. ERP treatment was manualized based on Kozak and Foa's protocol and included 16 twice-weekly 120 minute sessions. Therapists were advanced graduate students who had received extensive training from experts in ERP and ACT for OCD. Sessions 1 and 2 included information-gathering, psychoeducation, presentation of the rationale for ERP, and introduction to self-monitoring of rituals. Session 3 was dedicated to developing the treatment plan (exposure hierarchy, response prevention plan). Sessions 4-15 included in-session prolonged and repeated gradual exposure therapy, informal discussion of mistaken cognitions, daily exposure homework assignments, and instructions to refrain from rituals (response prevention in session and between sessions), along with self monitoring of any rituals that were performed. Session 16 addressed discontinuation and relapse prevention.
ACT+ERP. This condition matched the ERP condition in terms of the number and duration of exposure sessions. Sessions 1 and 2 involved an abbreviated discussion of the ACT model of ERP focusing on acceptance of obsessions and anxiety, just noticing and not acting on obsessions, being present with one's internal experiences, and linking therapy to one's values. Session 3 involved the development of an exposure hierarchy and response prevention plan, and explaining the ACT-based approach to ERP which focuses on learning flexible responding in the presence of obsessions, anxiety, and urges to ritualize. The exposure exercises were procedurally similar to ERP but focused on the facilitation of ACT processes. For example, the therapist might teach just noticing of obsessions by exposing the participant to the feared stimulus and asking the participant to watch the obsessional thoughts pass by as if they were leaves floating on a stream. Homework exposure practice was linked to the patient's goals and values, and participants were instructed to practice ACT processes during these assignments. Session 16 covered an ACT-based model of relapse prevention focusing on following one's values in the presence of obsessive thoughts and compulsive urges.
3e. Standardization and Integrity Procedures. All assessors and therapists were trained to competency in their duties by Drs. Twohig and Abramowitz. All assessment and treatment sessions were videotaped and a random selection of 20% of tapes were reviewed by the investigators and scored for integrity of implementation. Drs. Twohig and Abramowitz also served as supervisors on the project. Cross-site data and video sharing of supervision occurred via the Internet in a HIPAA-compliant manner and weekly meetings occurred via Skype.
3f. Length of Study. The length of the study was 6 years. The investigators developed the ERP+ACT treatment manual during the first two months of the study period. Following this, the investigators completed recruitment and treatment of all patients between the 1st and 6th years. This left the final 1 year for follow-up data collection, data analysis, and for writing up and presenting the results.
3g. Statistical Methods. The investigators will use the following procedures to address each hypotheses: 1) Between-groups t-tests to examine differences on measures of treatment acceptability and the quantity and quality of adherence to ERP assignments at post-treatment. Corresponding effect sizes will be computed to examine the magnitude of this effect. 2) Within-group t-tests to determine the statistical significance of pre-posttest and pretest-follow-up differences on outcome measures. Corresponding within-group effect sizes will also be computed to examine the magnitude of these differences. Further, the methods described by Jacobson & Truax to examine the extent of clinically significant and reliable change from pre to posttest and at follow-up will be utilized.
3h. Power Analyses. Power calculations were based on the average within-group effect size of 1.41 reported in a meta-analysis of CBT for OCD. This effect size was derived from clinical interview measures (e.g., the Yale-Brown Obsessive Compulsive Scale). A sample size of 60 affords greater than 90% power to detect a similar effect size with a t-test in the present study if alpha is set at .05. There are no available studies comparing treatment adherence and quality of life between two active psychological treatments for OCD from which to calculate power estimates. The investigators will therefore examine effect sizes in addition to significance levels when comparing between ERP+ACT and ERP on these variables.
4. INTENDED USE OF THE RESULTS
The results from the present study will be published and disseminated through conference presentations. They will also be used as pilot data to apply for NIMH funding for a larger study using the R34 mechanism (treatment development).
Inclusion Criteria:
Exclusion Criteria: