Twenty-six studies (1,151 participants) were included in the review. Mean sample size was 50 (range 10 to 138).
Effects on cognitive performance (25 studies): Pooled results showed that the effect of cognitive remediation on overall cognition was medium and statistically significant (ES 0.41, 95% CI 0.29 to 0.52, p<0.001; 1,214 participants). Follow-up at eight months (six studies) showed that the average effect size for overall cognitive performance was 0.66, but not statistically significant. Other statistically significant medium-range effects were reported for attention/vigilance (ES 0.41, 95% CI 0.25 to 0.57, p<0.001; 659 participants), speed of processing (ES 0.48, 95% CI 0.28 to 0.69, p<0.001; 655 participants), verbal working memory (ES 0.52, 95% CI 0.33 to 0.72, p<0.001; 428 participants), verbal learning and memory (ES 0.39, 95% CI 0.20 to 0.58, p<0.001; 858 participants), reasoning/problem solving (ES 0.47, 95% CI 0.30 to 0.64, p<0.001; 564 participants) and social cognition (ES 0.54, 95% CI 0.22 to 0.88, p<0.001; 228 participants).
There was statistically significant heterogeneity in the analysis for verbal learning and memory (p<0.05). Moderator analysis concluded that significantly larger effect sizes for this outcome were associated with more hours of cognitive remediation (ES 0.57, p<0.05) and for drill and practice interventions (ES 0.48, p<0.05) compared with drill and practice plus strategy coaching.
Effects on symptoms and functioning (1,324 participants): Small pooled effect sizes were found for the effects of cognitive remediation on symptoms (ES 0.28, 95% CI 0.13 to 0.43, p<0.001; 709 participants) and functioning (ES 0.35, 95% CI 0.07 to 0.62, p<0.05; 615 participants).
Statistically significant heterogeneity was found for functioning (p<0.01). Moderator analyses concluded that larger effect sizes for improved psychosocial functioning were found in studies where adjunctive psychiatric rehabilitation was offered (ES 0.47, p<0.01), where drill and practice plus strategy coaching was provided compared to drill and practice alone (ES 0.62, p<0.05) and in studies that included older participants (ES 0.55, p<0.05).