Systematic review and meta-analysis of the placebo effect in panic disorder: Implications for research and clinical practice.

Abstract

This review aimed to measure the degree of placebo response in panic disorder. We searched major databases up to 31 January 2021, for randomized pharmacotherapy trials published in English. A total of 43 studies met inclusion criteria to be in the analysis (with 174 separate outcome measurements). Changes in outcome measures from baseline in the placebo group were used to estimate modified Cohen's d effect size. A total of 43 trials (2392 subjects, 174 outcomes using 27 rating scales) were included in the meta-analysis. Overall placebo effect size was 0.57 (95% confidence interval = [0.50, 0.64]), heterogeneity (I2: 96.3%). Higher placebo effect size was observed among clinician-rated scales compared to patient reports (0.75 vs 0.35) and among general symptom and anxiety scales compared to panic symptoms and depression scales (0.92 and 0.64 vs 0.56 and 0.54, respectively). There was an upward trend in effect size over the publication period (r = 0.02, p = 0.002) that was only significant among clinician-rated scales (r = 0.02, p = 0.011). There was no significant publication bias, Egger's test (p = 0.08). We observed a substantial placebo effect size in panic disorder. This effect was more prominent for some aspects of panic disorder psychopathology than for others and was correlated with the source of the assessment and publication year. This finding has implications both for research design, to address the heterogeneity and diversity in placebo responses, and for clinical practice to ensure optimal quality of care. PROSPERO, CRD42019125979.