Seven controlled studies with 340 patients (108 dropped out) were included. Four studies provided follow-up data (277 patients with 62 drop-outs).
All of the studies satisfied Rosenthal's chi-squared test for heterogeneity.
The mean effect size for change in psychiatric symptoms from pre- to post-treatment was 0.65 (95% confidence interval, CI: 0.56, 0.71). All 7 studies reported a statistically-significant decrease in positive symptoms at post-treatment, while 5 of the 7 reported a statistically-significant decrease for CT relative to the control condition at post-treatment.
Simple regression analyses revealed no statistically-significant relationship between the number of sessions of CT treatment and change in psychotic symptoms (correlation, R=0.28; d.f.=1,4, p=0.64). In 3 studies, the treatment intensity varied across participants and, as such, these studies were excluded from the analysis.
Four studies assessed the outcome past 6-months' post-treatment. In all 4 studies, the patients continued to improve statistically significantly over the follow-up periods; their combined mean effect size at follow-up was 0.93.
The studies included strong experimental designs, well-articulated treatment interventions (which were generally standardised in the form of a treatment manual), and comprehensive assessment batteries. Although the assessment raters did not provide the treatment in most of the included studies, they were not blinded to the treatment allocation. The use of independent raters was not mentioned in 2 studies and, therefore, was assumed not to have been employed. The overall drop-out rate was 12.4%.