Behavioral Therapy For Panic Disorder Research Articles

Predicting Dropout From Cognitive Behavioral Therapy for Panic Disorder Using Machine Learning Algorithms

Predicting Dropout From Cognitive Behavioral Therapy for Panic Disorder Using Machine Learning Algorithms

Feasibility of Guided Internet-Based Cognitive Behavioral Therapy for Panic Disorder and Social Anxiety Disorder in Japan: Pilot Single-Arm Trial.

Cognitive behavioral therapy (CBT) is effective in treating anxiety disorders. Accessibility to CBT has been limited in Japan due to the shortage of therapists. While an open-source e-learning system can be used to create a simple internet-based cognitive behavioral therapy (ICBT) program, the safety and outpatient acceptance of this treatment approach have not been explored in Japan. The aim of this study was to investigate whether outpatients with anxiety disorders could accept and successfully complete the ICBT program with guidance by CBT therapists when implementing therapeutic modules and CBT tasks. Due to being in the initial phase of a novel treatment in Japan, this study was intended for verification with a small sample size. In total, 6 adults, including 4 male participants and 2 female participants, were enrolled in a single-arm trial. The intervention involved guided ICBT comprising 12 sessions, including CBT text, comprehension confirmation tests, and explanatory videos about cognitive behavioral models, accessible through a website. The therapist guided the participants in accessing the ICBT program and answering their questions using a chat tool. The primary outcome was anxiety severity assessed using the State-Trait Anxiety Inventory-Trait. Secondary outcomes included the Panic Disorder Severity Scale, Liebowitz Social Anxiety Scale (LSAS), Beck Anxiety Inventory (BAI), Patient Health Questionnaire-9, Generalized Anxiety Disorder-7, and Working Alliance Inventory-Short Form (WAI-SF). Statistical analyses were performed using paired 2-tailed t tests to assess the changes in clinical symptoms. The total WAI-SF score at the final session was used to evaluate the therapeutic alliance. For statistical analyses, mean changes for total State-Trait Anxiety Inventory-Trait, BAI, Panic Disorder Severity Scale, LSAS, Patient Health Questionnaire-9, and Generalized Anxiety Disorder-7 scores were analyzed using the paired 2-tailed t test. The 2-sided significance level for hypothesis testing was set at 5%, and 2-sided 95% CIs were calculated. Most participants diligently engaged with the ICBT program. No adverse events were reported. The mean total scores for the primary outcome decreased by 11.0 (SD 9.6) points (95% CI -22.2 to 0.20; Hedges g=0.95), but it was not statistically significant. The mean total scores for the secondary outcomes that assess clinical symptoms decreased, with a significant reduction observed in the BAI of 15.7 (SD 12.1) points (95% CI -28.4 to -3.0; P=.03; Hedges g=1.24). The mean total scores for PDSS and LSAS decreased significantly, by 12.0 (SD 4.24) points (95% CI -50.1 to 26.1; P=.16; Hedges g=1.79) and 32.4 (SD 11.1) points (95% CI -59.7 to -4.3; P=.04; Hedges g=1.38), respectively. Of the participants, 67% (n=4) showed treatment response, and 50% (n=3) achieved remission after the intervention. The therapeutic alliance, measured using the WAI-SF, was moderate. Guided ICBT may be feasible for the treatment of outpatients with panic disorder and social anxiety disorder in Japan. University Hospital Medical Information Network Clinical Trials Registry UMIN0000038118; https://center6.umin.ac.jp/cgi-open-bin/ctr/ctr_view.cgi?recptno=R000043439.

Cognitive change before sudden gains in cognitive behavioural therapy for panic disorder.

Sudden gains occur in a range of disorders and treatments and are of clinical and theoretical significance if they can shed light on therapeutic change processes. This study investigated the relationship between sudden gains in panic symptoms and preceding cognitive change during cognitive behavioural therapy (CBT) for panic disorder. Participants with panic disorder completed in session measures of panic symptoms and catastrophic cognitions. Independent samples t-tests were used to compare the post-treatment score of those who met criteria for one or more sudden gain during treatment with those who did not, and to compare within-session cognitive change between pre-sudden gain sessions and the previous (control) session. Twenty-two (42%) of 53 participants experienced a sudden gain during treatment. Participants demonstrating a sudden gain showed more improvement in panic symptoms from pre- to post-treatment than those without a sudden gain. The within-session cognitive change score in the pre-gain session was significantly greater than in the control session. Sudden gains occurred in individual CBT for panic disorder and within-session cognitive change was associated with sudden gains. This is consistent with the cognitive model of panic disorder and highlights how sudden gains can help to identify key change processes.

Agreement in patient-therapist alliance ratings and its relation to dropout and outcome in a large sample of cognitive behavioral therapy for panic disorder

Objective The therapeutic alliance is related to treatment outcome but less is known about the agreement on alliance between patients and therapists and its relationship to outcomes. We examined the association of patient-therapist congruence of alliance perceptions, early and late in cognitive behavioral therapy for panic disorder in relation to symptom reduction and dropout. Method: Patients (n = 181) and their therapists provided alliance ratings early and late during 11-session treatment. Independent evaluators rated patients’ symptomatic levels post-treatment. Polynomial regression and response surface analysis were used to examine congruence as a predictor of outcome. Results: Early in therapy, stronger combined patient-therapist alliances, regardless of agreement, predicted lower symptom severity at the end of therapy and a lower likelihood of dropout. Late in treatment, the outcome was worse when therapist ratings of the alliance were higher than those of the patient. Conclusions: Therapist-patient agreement on the strength of the alliance is important for symptom improvement and dropout. The study highlights the importance of understanding the dyadic nature of the alliance and its impact on therapeutic change.

Maximizing the non-specific factors in brief cognitive behavioral therapy for panic disorder and agoraphobia: A multiple baseline case series documenting feasibility and initial efficacy.

There is evidence for non-specific factors impacting treatment outcomes, with pragmatic concerns regarding the need to popularize briefer formats of cognitive behavioral therapy (CBT). The need to have more culturally suitable and acceptable forms of CBT is also indicated. We evaluated the feasibility and efficacy of a brief 5 session CBT (bCBT) in participants (N = 4) with panic disorder (PD) and agoraphobia, using a non-concurrent multiple baseline design. In this case series, efforts were made to maximize non-specific factors of psychotherapy in bringing about treatment outcomes. Reliable and significant treatment effects were observed at post-intervention and follow-up assessments. The present study offers preliminary evidence of a bCBT protocol that comprises the efforts to maximize the non-specific factors in psychotherapy such as credibility, expectancy, and the therapeutic alliance in bringing treatment outcomes; however, further controlled evaluation is warranted. We also discuss the mechanisms contributing to these treatment outcomes in the present protocol.

Effectiveness of Self-Guided Virtual Reality-Based Cognitive Behavioral Therapy for Panic Disorder: Randomized Controlled Trial.

BackgroundVirtual reality (VR) is as effective a technique as traditional cognitive behavioral therapy (CBT) and a promising tool for treating panic disorder symptoms because VR exposure can be safer and has better acceptability than in vivo exposure and is more immersive than exposure through imagination. CBT techniques can be delivered more effectively using VR as well. So far, VR has required high-quality devices, but the development of mobile VR technology has improved user availability. At the same time, a well-structured form of VR can be reproduced and used anywhere. This means that VR can be used to provide a self-guided form of treatment and address the high treatment costs of evidence-based therapy and the lack of professional therapists. This study aimed to investigate the potential of self-guided VR as an alternative to high-cost treatment.ObjectiveThe main goal of this study was to offer data about the efficacy of a mobile app-based self-led VR CBT in the treatment of panic disorder.MethodsA total of 54 subjects with panic disorder were enrolled in this study and randomly assigned to either the VR treatment group or waitlist group. The VR treatment was designed to be total 12 sessions for 4 weeks. The VR treatment consists of 4 steps in which patients are gradually exposed to phobic stimuli while learning to cope with panic symptoms in each stage. The effectiveness of treatment was assessed through the Panic Disorder Severity Scale, Hamilton Rating Scale for Depression, Body Sensations Questionnaire, Albany Panic and Phobia Questionnaire, Anxiety Sensitivity Index, State-Trait Anxiety Inventory, Hospital Anxiety and Depression Scale, Korean Inventory of Social Avoidance and Distress Scale, Korean Inventory of Depressive Symptomatology, and Perceived Stress Scale. In addition, physiological changes using heart rate variability were evaluated.ResultsIn within-group analyses, the VR treatment group exhibited improvements in panic disorder symptoms, anxiety, and depression after 4 weeks, while the waitlist group did not show any significant improvement. Compared to the waitlist group, the VR treatment group showed significantly greater improvements in the Panic Disorder Severity Scale in both completer analysis and intention-to-treat analysis. Heart rate variability in the VR treatment group showed improvement in normalized high frequency from baseline to postassessment with no significant differences in any outcome measure between groups.ConclusionsThe self-guided, mobile app-based VR intervention was effective in the treatment of panic symptoms and restoring the autonomic nervous system demonstrating the validity of the use of VR for self-guided treatment. VR treatment can be a cost-effective therapeutic approach.Trial RegistrationClinicalTrials.gov NCT04985019; https://clinicaltrials.gov/ct2/show/NCT04985019

Internet-delivered exposure therapy versus internet-delivered cognitive behavioral therapy for panic disorder: A pilot randomized controlled trial

AimTo compare the efficacy and acceptability of internet-delivered exposure therapy for panic disorder, to multi-component internet-delivered cognitive behavioral therapy (iCBT) that included controlled breathing, cognitive restructuring and exposure. MethodsParticipants with panic disorder, with or without agoraphobia, were randomized to internet-delivered exposure therapy (n = 35) or iCBT (n = 34). Both programs were clinician guided, with six lessons delivered over eight weeks. Outcomes included panic disorder and agoraphobia symptom severity, as well as depression symptom severity, functional impairment and days out of role. ResultsParticipants in both conditions displayed a large reduction in panic disorder symptom severity (ds >1.30) from pre- to post-treatment. Participants in both conditions displayed medium to large reduction in agoraphobia and depression symptom severity, functional impairment and days out of role. Effects were maintained at three- and six-month follow-up. There was no significant difference between the interventions in clinical outcomes, adherence or treatment satisfaction. ConclusionsInternet-delivered exposure therapy appeared to be as acceptable and efficacious as more established iCBT, despite including less strategies. However, a fully powered replication is now needed to compare the two approaches.

Supplemental Material for Patient, Therapist, and Relational Antecedents of Hostile Resistance in Cognitive–Behavioral Therapy for Panic Disorder: A Qualitative Investigation

Supplemental Material for Patient, Therapist, and Relational Antecedents of Hostile Resistance in Cognitive–Behavioral Therapy for Panic Disorder: A Qualitative Investigation

공황장애 기독교인지행동치료 프로그램을 통한 변화과정 연구 : 사례연구를 중심으로

공황장애 기독교인지행동치료 프로그램을 통한 변화과정 연구 : 사례연구를 중심으로

Intensive one-week internet-delivered cognitive behavioral therapy for panic disorder and agoraphobia: A pilot study

This is the first pilot study to explore the feasibility, acceptability and preliminary efficacy of intensive cognitive behavioral therapy (CBT) for panic disorder and/or agoraphobia delivered via the internet. Ten participants who met DSM-5 criteria for panic disorder and/or agoraphobia (6 males; mean age = 43.40, SD = 15.25) completed The Intensive Panic Program: a six-lesson exposure-based CBT program, delivered online over seven days. Clinician support was provided via phone and email. All 10 participants completed the program (100% adherence) and high levels of satisfaction were reported. We found large and significant reductions in panic symptom severity at post-treatment (d = 1.40), which were maintained at two-month follow-up. We also found large reductions in agoraphobic avoidance (d = 0.92) and functional impairment (d = 1.04) at follow-up, and days out of role were halved. On average, 132 min (SD = 42, range: 47–183) of clinician time was spent per participant during the treatment week. The results provide promising preliminary evidence for the feasibility and acceptability of internet-delivered intensive CBT for panic disorder and/or agoraphobia. A larger, randomized control trial is now needed to evaluate the efficacy of this program compared to a control group and to explore long-term outcomes.Clinical trial registration numberACTRN12618001501235

A pilot study augmenting cognitive behavioral therapy for panic disorder with attention bias modification: Clinical and psychophysiological outcomes

Background and objectivesStudies show that attentional bias towards threat is a key maintenance factor for panic disorder (PD). Attentional bias may be an important mechanism of symptom reduction, and thus, a useful target for optimizing outcomes. The current study examined whether an attention bias modification (ABM) task enhanced CBT outcomes. Multiple methods for assessing PD were used, including physiological measurement and clinician-rated assessment. MethodsAdults with panic disorder (N = 24) received seven sessions of CBT with either ABM or sham attention tasks. Psychophysiological reaction to a loud tones startle paradigm was assessed before and after treatment. ResultsAcross both groups, panic symptom severity decreased with CBT. The ABM group showed greater reductions in PD symptoms compared to the placebo group. Notably, however, changes in attentional bias were not associated with symptom reductions across groups. No significant group differences on psychophysiological assessment were observed. LimitationsThis study is limited by the small sample size, which rendered our power to investigate mediators and moderators insufficient. More research is needed to validate the effect of attention bias modification on attentional bias to threat. ConclusionsThis pilot study suggests that future research should investigate attentional bias in conjunction with CBT. A larger sample would provide opportunity to further investigate the mechanisms through which ABM works, along with potential moderating factors and the use of psychophysiological measurements in panic disorder.

Internet-delivered cognitive behavioral therapy for panic disorder with or without agoraphobia: a systematic review and meta-analysis

ABSTRACT The current systematic review and meta-analysis examined the efficacy and effectiveness of internet-delivered cognitive behavioral therapy (iCBT) on panic disorder and agoraphobia symptom severity. Twenty-seven studies were identified. Results from nine randomised controlled trials (RCTs) showed that iCBT outperformed waiting list and information controls for panic (g = 1.22) and agoraphobia (g = .91) symptoms, but the quality of RCTs varied and heterogeneity was high. Results from three RCTs suggested iCBT may have similar outcomes to face-to-face CBT in reducing panic and agoraphobia symptoms. Within-group effect sizes between baseline and post-treatment were large for panic (n = 29, g = 1.16) and medium for agoraphobia symptom severity (n = 18, g = .73). Subgroup analyses of within-group pre/post treatment effect sizes showed larger within-group effect sizes for efficacy studies (n = 15) compared to effectiveness studies (n = 14) for panic severity (g = 1.38 vs. g = .98) but not agoraphobia severity. There was no impact of program length, inclusion or arousal reduction techniques, or degree of clinician support. Within-group effects of iCBT suggest the reduction in panic and agoraphobia symptom severity is maintained at 3–6 month follow-up (n = 12).

Development of a very brief scale for detecting and measuring panic disorder using two items from the Panic Disorder Severity Scale-Self Report

ObjectiveTo minimize the burden in detecting and monitoring Panic Disorder and Agoraphobia by developing a very brief scale with selected items from the Panic Disorder Severity Scale-Self Report (PDSS-SR), and to investigate the proposed scale's psychometric properties in a comorbid sample. MethodsA sample of 5103 patients from the Internet Psychiatry Clinic in Sweden, diagnosed and treated with Internet-based cognitive behavioral therapy for panic disorder (n = 1390), social anxiety disorder (n = 1313) or depression (n = 2400), responded to the PDSS-SR. Six criteria related to factor structure, sensitivity to change and clinical representativeness were used to select items. Psychometric analyses for the selected very brief scale were performed. ResultsItems 2 (distress during panic attacks) and 4 (agoraphobic avoidance), were selected to create the very brief PDSS-SR version. Correlations with the full scale were high at screening, pre and post, and for change (0.87–0.93). Categorical Omega was ⍵C = 0.74. With a cut-off of 3 points, the scale could detect panic disorder in a psychiatric sample with a sensitivity of 85% and a specificity of 66%. LimitationsLimitations include lack of healthy controls and lack of blinding on secondary outcome measures. ConclusionThe proposed 2-item PDSS-SR version is a good candidate for a very brief panic disorder questionnaire, both for detecting cases and for measuring change. This is especially useful in clinical settings when measuring more than one condition at a time.

The role of therapist and patient in-session behavior for treatment outcome in exposure-based cognitive behavioral therapy for panic disorder with agoraphobia.

There is a very limited amount of research on the relationship between therapist and patient in-session behavior and treatment outcome in cognitive behavioral therapy (CBT) for panic disorder with agoraphobia (PD/AG). Additionally, the findings tend to be inconclusive. This study investigates the association between therapist competence, adherence, patient interpersonal behavior, and therapeutic alliance and outcome in a low-control CBT setting by using comprehensive measures. Twenty-six patients with PD/AG received 12 sessions of exposure-based CBT. With regard to the outcome, treatments were classified either as problematic or nonproblematic by means of distinct criteria. Two raters evaluated the in-session behavior. Patient interpersonal behavior was significantly associated with outcome at follow-up (r = 0.49). At posttreatment, the correlation did not reach significance ( r = 0.34). Competence, adherence, and alliance were not outcome associated. The findings emphasize the need for therapists to pay particular attention to patients' interpersonal behavior during treatment.

Investigating effective treatment factors in brief cognitive behavioral therapy for panic disorder.

Numerous studies have provided evidence for the effectiveness of cognitive behavioral therapy (CBT) on panic disorders (PDs). There has also been growing attention on brief CBT with regard to delivering intensive treatment efficiently. This study investigated the essential parts of mindfulness-based brief CBT to optimize treatment benefits.A total of 37 patients were retrospectively enrolled in this study. They were recruited from the anxiety/panic/fear clinic of Seoul National University Hospital. The patients participated in group CBT once a week for a total of 4 sessions over a 4-week period, when they were assessed using the Panic Disorder Severity Scale (PDSS), Anxiety Sensitivity Index-Revised (ASI-R), Albany Panic and Phobia Questionnaire (APPQ), State-Trait Anxiety Inventory (STAI), Beck Anxiety Inventory (BAI), Beck Depression Inventory (BDI), and Yale-Brown Obsessive Compulsive Scale (Y-BOCS) before and after brief CBT. Twenty-nine patients completed the 1-month follow-up.There were significant reductions in PDSS (P < .001), ASI-R-fear of respiratory symptoms (P = .006), ASI-R-fear of publicly observable anxiety reaction (P = .002), ASI-R-fear of cardiovascular symptoms (P < .001), ASI-R-fear of cognitive dyscontrol (P = .001), ASI-R-Total (P < .001), APPQ-Agoraphobia (P = .003), APPQ-Total (P = .028), STAI-State anxiety (P < .001), STAI-Trait anxiety (P = .002), BAI (P = .003), and BDI (P < .001) scores. We also found significant associations between ASI-R-fear of cardiovascular symptoms, ASI-R-Total, and changes in PDSS scores. A stepwise multiple linear regression analysis indicated that anxiety sensitivity for fear of cardiovascular symptoms predicted an improvement in panic severity (β = 0.513, P = .004).Our findings suggested that behavioral aspects, especially physiological symptom control, needed to be considered in brief, intensive CBT for PD. The results also suggested that a mindfulness-based brief CBT approach might be particularly helpful for patients with PD who have severe cardiovascular symptoms.

Client resistance predicts outcomes in cognitive–behavioral therapy for panic disorder

Objective: Little is known about how therapy processes relate to outcome in cognitive–behavioral therapy (CBT) for panic disorder (PD). This study examined whether client resistance predicts CBT for PD outcomes beyond the effects of established pre-treatment predictors. A secondary aim was to assess the consistency of resistance over treatment. Method: Data were from 71 adults participating in up to 24 biweekly sessions of CBT in a randomized controlled trial. Panic severity was assessed before, during (at Weeks 1, 5, and 9), and at termination of treatment (Week 12) using the Panic Disorder Severity Scale. Trained coders reliably rated resistance in videos of Sessions 2 and 10 using the Client Resistance Code. Results: Resistance was found to be moderately consistent (r = .64). Although overall resistance was unrelated to outcomes, hierarchical linear modeling revealed that openly hostile resistance at Session 10 predicted significantly diminished symptom change (r = .28, CI95% = [.01, .51]), beyond the effects of pretreatment predictors. Hostile resistance at Session 2 predicted attrition (rrb = −.30, p = .001), even after established predictors were controlled. Conclusions: Although some forms of resistance may be benign, openly hostile resistance is an important therapy marker that warrants increased clinical and research attention.

Improvement in adult anxious and avoidant attachment during cognitive behavioral therapy for panic disorder

Objective: This study examined attachment within the framework of cognitive behavioral therapy (CBT) for panic disorder with agoraphobia (PDA) by measuring the changes in avoidant and anxious attachment in a session-by-session analysis. Method: Thirty-one patients with PDA were treated using CBT. Pre-session data on attachment style (ECR), avoidance behaviors (MI-alone: MI-al; MI-accompanied: MI-ac), anxiety sensitivity (ASI), emotion regulation (ERQ), and working alliance (WAI) were collected. Mixed model analyses were conducted to estimate relationship between changes in attachment, PDA symptoms, and related measures. Results: Variables improved during therapy. Changes in ASI were positively related to changes in avoidant and anxious attachment. Changes in MI-al, but not MI-ac, were related to changes in ECR. Changes in ASI and MI-al predicted changes in ECR-anxiety in the following session, but not vice versa. Similarly, changes in avoidance behaviors and ERQ-suppression, preceded changes in ECR-avoidance unidirectionally. Whereas WAI significantly improved, its variability was related only to simultaneous changes in ECR-anxiety. Conclusions: Overall, avoidant and anxious attachment improved during CBT for PDA. This change was related to and preceded by improved anxiety sensitivity, avoidance behaviors, and emotion regulation. These findings suggest that CBT for panic probably has downstream effects on attachment representations.Clinical or methodological significance of this article: This is the first report of changes in attachment via cognitive behavioral therapy for panic disorder. Data were examined using repeated measures session by session, allowing for examination of temporal precedence of changes. Results revealed that anxiety sensitivity, avoidance, and suppression preceded changes in attachment. Results are discussed in the framework of modern attachment theory models.

Anxiety sensitivity as a predictor of broad dimensions of psychopathology after cognitive behavioral therapy for panic disorder.

BackgroundPanic disorder (PD) is a common disease and presents with broad dimensions of psychopathology. Cognitive behavioral therapy (CBT) is known to improve these broad dimensions of psychopathology in addition to PD symptoms. However, little is known about the predictors of treatment response in comorbid psychiatric symptoms after CBT for PD. Recent studies suggest that anxiety sensitivity (AS) may be a key vulnerability for PD. This study aimed to examine AS as a predictor of broad dimensions of psychopathology after CBT for PD.Materials and methodsIn total, 118 patients with PD were treated with manualized group CBT. We used multiple regression analysis to examine the associations between 3 Anxiety Sensitivity Index (ASI) factors (physical concerns, mental incapacitation concerns, and social concerns) at baseline and the subscales of the Symptom Checklist-90 Revised (SCL-90-R) at endpoint.ResultsLow levels of social concerns at baseline predicted low levels on 5 SCL-90-R subscales after CBT: interpersonal sensitivity, depression, hostility, paranoid ideation, and psychosis. High levels of mental incapacitation concerns significantly predicted low levels on 3 SCL-90-R subscales after treatment: interpersonal sensitivity, hostility, and paranoid ideation. Physical concerns at baseline did not predict broad dimensions of psychopathology.ConclusionThis study suggested that the social concerns and mental incapacitation concerns subscales of the ASI at baseline predicted several dimensions of psychopathology after CBT for PD. To improve comorbid psychopathology, it may be useful to direct more attention to these ASI subscales.

Recovery of Percent Vital Capacity by Breathing Training in Patients With Panic Disorder and Impaired Diaphragmatic Breathing.

Slow diaphragmatic breathing is one of the therapeutic methods used in behavioral therapy for panic disorder. In practice, we have noticed that some of these patients could not perform diaphragmatic breathing and their percent vital capacity was initially reduced but could be recovered through breathing training. We conducted a comparative study with healthy controls to investigate the relationship between diaphragmatic breathing ability and percent vital capacity in patients with panic disorder. Our findings suggest that percent vital capacity in patients with impaired diaphragmatic breathing was significantly reduced compared with those with normal diaphragmatic breathing and that diaphragmatic breathing could be restored by breathing training. Percent vital capacity of the healthy controls was equivalent to that of the patients who had completed breathing training. This article provides preliminary findings regarding reduced vital capacity in relation to abnormal respiratory movements found in patients with panic disorder, potentially offering alternative perspectives for verifying the significance of breathing training for panic disorder.

The role of treatment delivery factors in exposure-based cognitive behavioral therapy for panic disorder with agoraphobia

Treatment delivery factors (i.e., therapist adherence, therapist competence, and therapeutic alliance) are considered to be important for cognitive behavioral therapy (CBT) for panic disorder and agoraphobia (PD/AG). In the current study, four independent raters conducted process evaluations based on 168 two-hour videotapes of 84 patients with PD/AG treated with exposure-based CBT. Two raters evaluated patients’ interpersonal behavior in Session 1. Two raters evaluated treatment delivery factors in Session 6, in which therapists provided the rationale for conducting exposure exercises. At the 6-month follow-up, therapists’ adherence (r=0.54) and therapeutic alliance (r=0.31) were significant predictors of changes in agoraphobic avoidance behavior; therapist competence was not associated with treatment outcomes. Patients’ interpersonal behavior in Session 1 was a significant predictor of the therapeutic alliance in Session 6 (r=0.17). The findings demonstrate that treatment delivery factors, particularly therapist adherence, are relevant to the long-term success of CBT for PD/AG.