Abstract
Benzodiazepines are commonly used for anxiety disorders, but there is a lack of systematic reviews relating to their safety and efficacy. A Cochrane Review is in preparation comparing beizodiazepines with placebo and preliminary data indicates that benzodiazepines have a significantly higher response rate and a significantly lower drop-out rate, but at the cost of more side effects. One of the issues is that there is a moderate to high rate of heterogeneity, with I-squared values estimated at between 20% (for side effects)to 91% (for HAM-A). We will be presenting sensitivity analyses of these results, looking at the year of publication, diagnostic criteria, dose effects, and baseline severity.