Six RCTs were included in the review (n=1,299 patients; range 83 to 369). There was a slight discrepancy between the number of patients reported in the forest plot and table; the figure from the forest plot has been cited in this abstract. The mean drop-out rate was 48% (range 27 to 77%). Four RCTs had unclear randomisation (two were at low risk of bias). Three RCTs had unclear allocation concealment (three were at low risk). Two RCTs had unclear blinding (four were at low risk). All the RCTs had unclear selective outcome reporting. Four RCTs were at high risk of bias for incomplete outcome data. Four RCTs were unclear for 'other' biases (two were at low risk of bias). Overall, trials were rated as being at moderate risk of bias.
selective serotonin re-uptake inhibitors (SSRIs) showed a statistically significantly greater improvement in response to treatment at six to eight months compared with placebo (OR 1.66, 95% CI 1.12 to 2.48; six RCTs), but there was evidence of statistical heterogeneity (I2=63.9%). Similar results were reported for patients with depression and no comorbidities (OR 2.13, 95% CI 1.11 to 4.08; I2=76.8%; three RCTs), but there was no statistical difference among patients with comorbidities (three RCTs) and no evidence of statistical heterogeneity.
There was no statistically significant difference in remission rates between patients receiving SSRIs and those receiving placebo (four RCTs). Subgroup analysis showed similar findings for patients with comorbidities (two RCTs), but in patients without comorbidities the remission rate was significantly higher in patients receiving SSRIs (OR 2.06, 95% CI 1.41 to 3.01; I2=0%; two RCTs).
There were no statistically significant differences between patients receiving SSRIs and those receiving placebo in the measure of overall acceptability (total drop-outs) and no evidence of statistical heterogeneity.
There was no evidence of publication bias according to funnel plots.
Secondary outcomes were reported in the review, but were somewhat limited by the poor reporting in the included trials.