Psychotherapies are the treatment of choice for panic disorder, but which should be considered as first-line treatment is yet to be substantiated by evidence.

To examine the most effective and accepted psychotherapy for the acute phase of panic disorder with or without agoraphobia via a network meta-analysis.

We conducted a systematic review and network meta-analysis of randomised controlled trials (RCTs) to examine the most effective and accepted psychotherapy for the acute phase of panic disorder. We searched MEDLINE, Embase, PsycInfo and CENTRAL, from inception to 1 Jan 2021 for RCTs. Cochrane and PRISMA guidelines were used. Pairwise and network meta-analyses were conducted using a random-effects model. Confidence in the evidence was assessed using Confidence in Network Meta-Analysis (CINeMA). The protocol was published in a peer-reviewed journal and in PROSPERO (CRD42020206258).

We included 136 RCTs in the systematic review. Taking into consideration efficacy (7352 participants), acceptability (6862 participants) and the CINeMA confidence in evidence appraisal, the best interventions in comparison with treatment as usual (TAU) were cognitive–behavioural therapy (CBT) (for efficacy: standardised mean differences s.m.d. = −0.67, 95% CI −0.95 to −0.39; CINeMA: moderate; for acceptability: relative risk RR = 1.21, 95% CI −0.94 to 1.56; CINeMA: moderate) and short-term psychodynamic therapy (for efficacy: s.m.d. = −0.61, 95% CI −1.15 to −0.07; CINeMA: low; for acceptability: RR = 0.92, 95% CI 0.54–1.54; CINeMA: moderate). After removing RCTs at high risk of bias only CBT remained more efficacious than TAU.

CBT and short-term psychodynamic therapy are reasonable first-line choices. Studies with high risk of bias tend to inflate the overall efficacy of treatments. Results from this systematic review and network meta-analysis should inform clinicians and guidelines.

Digital interventions for anxiety disorders are a promising solution to address barriers to evidence-based treatment access. Precise and powerful estimates of digital intervention effectiveness for anxiety disorders are necessary for further adoption in practice. The present systematic review and meta-analysis examined the effectiveness of digital interventions across all anxiety disorders and specific to each disorder v. wait-list and care-as-usual controls.

A systematic search of bibliographic databases identified 15 030 abstracts from inception to 1 January 2020. Forty-seven randomized controlled trials (53 comparisons; 4958 participants) contributed to the meta-analysis. Subgroup analyses were conducted by an anxiety disorder, risk of bias, treatment support, recruitment, location and treatment adherence.

A large, pooled effect size of g = 0.80 [95% Confidence Interval: 0.68–0.93] was found in favor of digital interventions. Moderate to large pooled effect sizes favoring digital interventions were found for generalized anxiety disorder (g = 0.62), mixed anxiety samples (g = 0.68), panic disorder with or without agoraphobia (g = 1.08) and social anxiety disorder (g = 0.76) subgroups. No subgroups were significantly different or related to the pooled effect size. Notably, the effects of guided interventions (g = 0.84) and unguided interventions (g = 0.64) were not significantly different. Supplemental analysis comparing digital and face-to-face interventions (9 comparisons; 683 participants) found no significant difference in effect [g = 0.14 favoring digital interventions; Confidence Interval: −0.01 to 0.30].

The precise and powerful estimates found further justify the application of digital interventions for anxiety disorders in place of wait-list or usual care.