Efficacy and Acceptability Comparisons of Cognitive Behavior Therapy, Drugs, and Their Combination for Panic Disorder in Adults: a Network Meta-analysis

Objective: To compare 22 oral drugs, cognitive behavior therapy (CBT), and their combination treatments for the acute treatment of adults with panic disorder in terms of remission rate and acceptability. Design: Systematic review and network meta-analysis Data sources: PubMed, Web of Science, Cochrane Central Register of Controlled Trials, and Embase databases from their inception up to May 26, 2019. Study selection: Randomized controlled clinical trials (RCTs) of any oral drugs, CBT, CBT combined with any drug, or placebo in the acute treatment of adults with panic disorder diagnosed according to standard operationalized criteria. The primary outcomes were efficacy (remission rate) and acceptability (treatment discontinuations due to any cause). Results: We identified 6585 reports that included 68 full-text RCTs involving 11101 patients. In terms of efficacy, 13 (68%) of 19 interventions were associated with higher remission rates than those of for placebo, with ORs ranging from 2.1 (95% credible interval [CrI] = 1.1 to 4.0) for sertraline to 13 (CrI = 4.5 to 44) for CBT combined with any drug. Regarding acceptability, alprazolam, imipramine, and etizolam were associated with lower dropout rates, with ORs ranging from 0.23 (CrI = 0.15 to 0.33) for alprazolam to 0.076 (CrI = 0.0021 to 0.77) for etizolam. Most of the differences between the other interventions were unclear. In head-to-head analyses, CBT combined with any drug was more effective than the other interventions, but it was no associated with an improvement in acceptability (OR = 0.12 to 0.219). Conclusions: CBT combined with any drug was more effective than the other interventions analyzed in this study. CBT alone did not differ significantly from other drugs alone. We found that most drugs are effective against panic disorder, but they exhibit different acceptability and tolerability profiles.