Systematic Review and Network Meta-analysis of Cognitive and/or Behavioral Therapies (CBT) for Tinnitus

Objective:To evaluate the efficacy of cognitive and/or behavioral therapies in improving health-related quality of life (HRQOL), depression, and anxiety associated with tinnitus.Data Sources:EMBASE, MEDLINE, PubMed, PsycINFO, and the Cochrane Registry were used to identify English studies from database inception until February 2018.Study Selection:Randomized controlled trials (RCTs) comparing cognitive and/or behavioral therapies to one another or to waitlist controls for the treatment of tinnitus were included.Data Extraction:Quality and risk were assessed using GRADE and Cochrane's Risk of Bias tool respectively.Data Synthesis:Pairwise meta-analysis (12 RCTs: 1,144 patients) compared psychological interventions to waitlist controls. Outcomes were measured using standardized mean differences (SMDs) and 95% confidence intervals (CI). I-2 and subgroup analyses were used to assess heterogeneity. Network meta-analysis (NMA) (19 RCTS: 1,543 patients) compared psychological therapies head-to-head. Treatment effects were presented by network diagrams, interval plots, and ranking diagrams indicating SMDs with 95% CI. Direct and indirect results were further assessed by inconsistency plots.Conclusions:Results are consistent with previously published guidelines indicating that CBT is an effective therapy for tinnitus. While guided self-administered forms of CBT had larger effect sizes (SMD: 3.44; 95% CI: -0.022, 7.09; I-2: 99%) on tinnitus HRQOL, only face-to-face CBT was shown to make statistically significant improvements (SMD: 0.75; 95% CI: 0.53, 0.97; I-2: 0%). Guided self-administered CBT had the highest likelihood of being ranked first in improving tinnitus HRQOL (75%), depression (83%), and anxiety (87%), though statistically insignificant. This NMA is the first of its kind in this therapeutic area and provides new insights on the effects of different forms of cognitive and/or behavioral therapies for tinnitus.