Author
IPSER Jonathan C.; KARIUKI Catherine M.; STEIN Dan J.;
Pharmacotherapy for social anxiety disorder: a systematic review.
Journal citation/publication details
Expert Reviews Neurotherapeutics, 8(2), February 2008, pp.235-257.
Summary
This systematic review and meta-analysis of 51 randomised controlled trials (RCT) revealed that pharmacotherapy for social anxiety disorder (SAD) is both effective and well tolerated. The efficacy and tolerability of venlafaxine suggest that it should be considered as a first-line agent along with the SSRIs. The results are discussed in terms of the current consensus on the treatment of SAD by different drug classes. An expert commentary on diagnosis and treatment and a five-year view, which looks at research in progress, are also included.
Context
Social anxiety disorder, or social phobia, is a highly prevalent condition associated with marked functional impairment that mostly develops in the teenage years. Both cognitive behavioural therapy and pharmacotherapy have been shown to be effective in managing the disorder. This systematic review focuses on the effects of medication in improving treatment response and symptom severity in SAD in order to determine whether particular classes of medication are more effective than others and identify any factors that may predict treatment response.
Methods
What sources were searched?
Searches were conducted on the Cochrane Collaboration Depression, Anxiety and Neurosis Controlled Trials Register (CCDAN-CTR), Medline (via PubMed) and PsycINFO. Additional studies were identified through scanning the reference lists of relevant articles and by contacting key researchers and pharmaceutical companies funding research in the field.
What search terms/strategies were used?
The keywords ‘social phobia’, ‘social anxiety’ and ‘performance anxiety’ were used in the CCDAN-CTR database. The Medline search query was based on a strategy developed by Robinson and Dickersin 2002; further details are available from the authors. Databases were searched from the date of inception to 2007.
What criteria were used to decide on which studies to include?
Randomised controlled trials of pharmacotherapy in adult participants aged 18 years or older who had been diagnosed with SAD according to DSM III or ICD-9 criteria were eligible for inclusion. Drug comparison trials were eligible if there was also a placebo treatment group. There were no language limitations and both published and unpublished studies were included. Treatment response outcomes had to include the Clinical Global Impressions Improvement (CGI-I) scale, or a closely related measure, and the Liebowitz Social Anxiety Scale (LSAS).
Who decided on their relevance and quality?
Two reviewers working independently assessed articles for inclusion. Disagreements were resolved by discussion with a third reviewer. Study quality was determined by one of the reviewers using the 23-point CCDAN Quality Rating Scale.
How many studies were included and where were they from?
Initially, the number of articles identified from PubMed, PsycINFO and CCDAN-TR was 758, 140, and 37, respectively. A total of 51 trials in 9,914 participants was included in the review after the exclusion of ineligible studies; reasons for exclusion are included in a flowchart of the search process, Figure 1.
How were the study findings combined?
Data was extracted onto standardised forms by two reviewers and exported into RevMan software for meta-analysis. Analyses of outcome data were intention-to-treat. Response to treatment was expressed as relative risk (RR) and number needed to benefit (NNTB). Effect size estimates were calculated using Hedge’s g and a random effects model. Further details of the statistical methods used are available in the text.
Findings of the review
Of the 51 RCTs included in the review, 32 were short term studies of 14 weeks duration or less, 11 were maintenance trials and eight studies had a relapse component. Most (36) were funded by pharmaceutical companies. There were 26 trials of SSRIs, five of the SNRI venlafaxine, three of the MAOI phenelzine, nine of RIMAs and 14 studies of other medications.
Patients who received active medication for an average of twelve weeks were significantly more likely to respond to treatment, as measured on the CGI scale, than those receiving placebo; 55.2% responded to treatment on average. Superior response to treatment was seen for SSRIs, venlafaxine, phenelzine, brofaromine, and moclobemide. All of the medications, with the exception of moclobemide and phenelzine, reduced symptom severity on LSAS subscales. Drug treatment was also beneficial in the longer term as evidenced from the results of maintenance and relapse studies. Medication was also more effective than placebo in reducing symptoms of depression, and disability at work, socially and in the family. SSRIs were consistently effective across these domains. The number of dropouts due to adverse events varied substantially across the classes of medication with the number needed to harm ranging from 87.2 for moclobemide to 9.3 for SSRIs; the overall figure was 14.4.
Authors' conclusions
‘This review provides evidence of the overall short-term efficacy of medication in treating SAD, with more than half of the participants responding to treatment’.
Implications for policy or practice
None are discussed.