Transdiagnostic Metacognitive Therapy Compared to Disorder-Specific Cognitive-Behavioral Therapy for Anxiety Disorders

Background Anxiety disorders are the most prevalent psychiatric disorders around the world. Effective treatment consists of pharmacotherapy or psychological treatment based on cognitive-behavioral therapy (CBT) and these treatment options are recommended in clinical guidelines, with CBT as the first-line treatment for anxiety disorders. However, only 50% of patients with anxiety disorders achieve remission status following CBT and 20% of patients drop out of CBT. Metacognitive therapy (MCT) represents an alternative treatment approach to CBT. The theoretical model of MCT emphasizes the role of dysfunctional metacognitions (rather than cognitions, as in CBT), particularly negative metacognitions, in the development and maintenance of anxiety disorders and other psychiatric disorders. Metacognitions refer to cognitions about cognition, for example, a belief such as "When I start worrying, I cannot stop". Several meta-analyses indicate that MCT may be superior to CBT for various psychiatric disorders. However, more studies with larger samples are required to draw firm conclusions about the effectiveness of MCT. An alternative approach to disorder-specific treatment is transdiagnostic treatment; that is, the application of a single, generic protocol for several disorders. There are advantages of transdiagnostic treatments in comparison to disorder-specific treatments in terms of therapist learnability (i.e., easier to learn one protocol than several) and dissemination into routine care. Despite the MCT model being described as applicable to a range of psychiatric disorders and MCT as a potentially transdiagnostic approach, at present there is only one sufficiently large study that compared transdiagnostic MCT (tMCT) to disorder-specific CBT. Purpose and aims The purpose of the present project is to investigate the effectiveness of tMCT compared to disorder-specific CBT in patients with anxiety disorders in psychiatric care and evaluate the cost-effectiveness. Aim 1 is to compare the short- and long-term effects of tMCT and CBT, from pre- to post-assessment and from post-assessment to 6- and 12-month follow-up assessments. Aim 2 is to examine possible mediators of change (metacognitions and cognitions). Aim 3 is to compare the cost-effectiveness of tMCT to CBT. Design and setting The project has a prospective, pragmatic, two-arm parallel-group randomized controlled superiority trial design and is conducted in psychiatric services in Stockholm, Sweden. Treatment is conducted in an individual format and face-to-face. Randomization and blinding Each participant is stratified individually on principal diagnosis prior to randomization and then randomly allocated with a 1:1 ratio to tMCT or CBT. A list of random numbers is generated for each diagnosis for each psychiatric unit by an individual independent of the project. Researchers, therapists, participants, and independent assessors are blinded to the allocation sequence. Assessors are also blinded to treatment condition at post-treatment assessment. Researchers are blinded to treatment allocation in the analysis phase at all assessment points. Therapist training and supervision Therapists are licensed psychologists or psychotherapists with prior training in CBT and employed in psychiatric services in Stockholm, Sweden. Only therapists who can show competence in MCT and CBT, respectively, are allowed to treat participants in the project. Procedure Patients are consecutively assessed for eligibility by project therapists. As part of routine clinical care, patients are assessed for principal and comorbid diagnoses. Patients meeting criteria for GAD, SAD, or PTSD are assessed whether they meet other inclusion but not exclusion criteria for participation in the project. Patients provide written informed consent to therapists. At pre-treatment, participants complete outcome measures. Participants are then randomly assigned to tMCT or CBT. Following the last session, and at 6-month and 12-month follow-up assessments, participants complete the same measures as at pre-treatment. In addition, at post-treatment principal and comorbid diagnoses are assessed by independent assessors. Data analysis Multilevel modeling is used to estimate between-group effects on outcome measures from pre- to post-assessment (following treatment completion; primary endpoint), and from post-assessment to 6- and 12-month follow-up assessments. To be comparable across diagnoses, scores on the primary outcome of disorder-specific measures are standardized by calculating z-scores. Missing data are estimated using maximum likelihood estimation. Data from all randomized participants are used in the multilevel models, following the principle of intention-to-treat. A detailed study protocol has been submitted for publication.