Background
The efficacy of combining psychotherapy and benzodiazepines for panic disorder is unclear, despite widespread use.
Objectives
To examine the efficacy of the combination compared with either treatment alone.
Search strategy
Randomised trials comparing the combination of psychotherapy and benzodiazepine with either therapy alone for panic disorder were identified. The Cochrane Depression, Anxiety and Neurosis Group Studies and References Registers were searched. References of relevant trials and other reviews were checked. Experts in the field were contacted. Additional unpublished data were sought from authors of the original trials.
Selection criteria
Two authors independently checked the records retrieved by the searches to identify randomised trials comparing the combined therapy versus either of the monotherapies, among adults with panic disorder.
Data collection and analysis
Two authors independently checked eligibility, assessed quality and extracted data from the eligible trials using a standardised data extraction form. The primary outcome was "response" based on global judgement. Random-effects meta-analyses were conducted, combining data from included trials.
Main results
Three trials met eligibility criteria. A 16-week behaviour therapy intervention was used in two trials, and a 12-week cognitive-behaviour therapy intervention in the third. Duration of follow-up varied, ranging from 0 to 12 months.
Two trials (total 166 participants) provided data comparing combination with psychotherapy alone (both using behaviour therapy). No statistically significant differences were observed in response during the intervention (relative risk (RR) for combination 1.25, 95% CI 0.78 to 2.03, P = 0.35), at the end of the intervention (RR 0.78, 0.45 to 1.35, P = 0.37), or at the last follow-up time point, although the follow-up data suggested that the combination might be inferior to behaviour therapy alone (RR 0.62, 0.36 to 1.07, P = 0.08).
One trial (77 participants) compared combination with a benzodiazepine alone. No differences were found in response during the intervention (RR 1.57, 0.83 to 2.98, P = 0.17). Although the combination appeared to be superior to the benzodiazepine alone at the end of treatment (RR 3.39, 1.03 to 11.21, P = 0.05) the finding was only borderline statistically significant, and no significant differences were observed at the 7-month follow-up (RR 2.31, 0.79 to 6.74, P = 0.12).
Authors' conclusions
The review established the paucity of high quality evidence investigating the efficacy of psychotherapy combined with benzodiazepines for panic disorder. Currently, there is inadequate evidence to assess the clinical effects of psychotherapy combined with benzodiazepines for patients who are diagnosed with panic disorder.