Nine RCTs (2,295 participants) were included in the meta-analysis. Eight open-label non-comparative studies met the inclusion criteria and were described in supplemental tables (available online). The overall quality of the augmentation studies was low. Quality was unclear for the monotherapy studies (only one of four studies could be assessed).
Augmentation studies (five RCTs, 912 participants): There was no statistically significant difference between second-generation antipsychotics and placebo for clinical response and remission. There was a significant increase in risk of all-cause discontinuation associated with second-generation antipsychotics (RR 1.43, 95% CI 1.04 to 1.96). There were no statistically significant differences between second-generation antipsychotics and placebo for change in Hamilton Anxiety Rating Scale score from baseline and with weight gain.
Monotherapy studies (four RCTs, 1,383 participants): A statistically significant difference in clinical response was reported in favour of second-generation antipsychotics in comparison with placebo (RR 1.31, 95% CI 1.20 to 1.44). Statistically significant differences in favour of second-generation antipsychotics were reported for remission (RR 1.44, 95% CI 1.23 to 1.68) and reduction in Hamilton Anxiety Rating Scale score (RR 3.66, 95% CI 5.13 to 2.19; one RCT). However, in comparison with placebo there was a statistically significant increase in risk of all-cause discontinuation (RR 1.30, 95% CI 1.09 to 1.54) and weight gain (2.2lb, 95% CI 1.16 to 3.24; one RCT). There was no evidence of significant statistical heterogeneity.
Further data reported in the review included a summary of data from open-label studies.