Open Access

Angela Emmrich, Katja Beesdo-Baum, Andrew T. Gloster, Susanne Knappe, Michael Höfler, Volker Arolt, Jürgen Deckert, Alexander L. Gerlach, Alfons Hamm, Tilo Kircher, Thomas Lang, Jan Richter, Andreas Ströhle, Peter Zwanzger, Hans-Ulrich Wittchen

• Background: Controversy surrounds the questions whether co-occurring depression has negative effects on cognitivebehavioral therapy (CBT) outcomes in patients with panic disorder (PD) and agoraphobia (AG) and whether treatment for PD and AG (PD/AG) also reduces depressive symptomatology. Methods: Post-hoc analyses of randomized clinical trial data of 369 outpatients with primary PD/AG (DSM-IV-TR criteria) treated with a 12-session manualized CBT (n = 301) and a waitlist control group (n = 68). Patients with comorbid depression (DSM-IV-TR major depression, dysthymia, or both: 43.2% CBT, 42.7% controls) were compared to patients without depression regarding anxiety and depression outcomes (Clinical Global Impression Scale [CGI], Hamilton Anxiety Rating Scale [HAM-A], number of panic attacks, Mobility Inventory [MI], Panic and Agoraphobia Scale, Beck Depression Inventory) at post-treatment and follow-up (categorical). Further, the role of severity of depressive symptoms on anxiety/depression outcome measures was examined (dimensional). Results: Comorbid depression did not have a significant overall effect on anxiety outcomes at post-treatment and follow-up, except for slightly diminished post-treatment effect sizes for clinician-rated CGI (p = 0.03) and HAM-A (p = 0.008) when adjusting for baseline anxiety severity. In the dimensional model, higher baseline depression scores were associated with lower effect sizes at post-treatment (except for MI), but not at follow-up (except for HAM-A). Depressive symptoms improved irrespective of the presence of depression. Conclusions: Exposure-based CBT for primary PD/AG effectively reduces anxiety and depressive symptoms, irrespective of comorbid depression or depressive symptomatology.