Effectiveness, Core Elements, and Moderators of Response of Cognitive Remediation for Schizophrenia: A Systematic Review and Meta-analysis of Randomized Clinical Trials

IMPORTANCE: Cognitive impairment is a core feature of schizophrenia, with negative consequences on functional outcomes. Although cognitive remediation (CR) is effective and mentioned in treatment guidance for schizophrenia, its active ingredients and ideal candidates are still debated. OBJECTIVE: To provide a comprehensive update on CR effectiveness for cognition and functioning in schizophrenia and analyze the core ingredients of efficacy and role of patient characteristics. DATA SOURCES: The reference list of the last comprehensive meta-analysis in 2011 was screened against eligibility criteria. Then, electronic databases (PubMed, Scopus, and PsycInfo) were systematically searched for articles published from January 2011 to February 2020. Reference lists of included articles and relevant reviews were hand searched, and Google Scholar was manually inspected. STUDY SELECTION: Eligible studies were randomized clinical trials comparing CR with any other control condition in patients diagnosed with schizophrenia spectrum disorders (with an unrestricted clinical status). Screening was performed by at least 2 independent reviewers. DATA EXTRACTION AND SYNTHESIS: The PRISMA guidelines were followed. Study data were independently extracted and pooled using random-effect models. Cohen d was used to measure outcomes. Trial methodological quality was evaluated with the Clinical Trials Assessment Measure. MAIN OUTCOMES AND MEASURES: Primary outcomes were changes in global cognition and overall functioning from baseline to after treatment, subsequently investigated through metaregressions, subgroup, and sensitivity analyses based on prespecified hypotheses, to identify potential moderators of response associated with treatment modality and patient characteristics. RESULTS: Of 1815 identified reports, 358 full texts were assessed and 194 reports on 130 studies were included. Based on 130 studies with 8851 participants, CR was effective on cognition (d, 0.29 [95% CI, 0.24-0.34]) and functioning (d, 0.22 [95% CI, 0.16-0.29]). An active and trained therapist (cognition: χ(2)(1), 4.14; P = .04; functioning: χ(2)(1), 4.26; P = .04), structured development of cognitive strategies (cognition: χ(2)(1), 9.34; P = .002; functioning: χ(2)(1), 8.12; P = .004), and integration with psychosocial rehabilitation (cognition: χ(2)(1), 5.66; functioning: χ(2)(1), 12.08) were crucial ingredients of efficacy. Patients with fewer years of education (global cognition: coefficient, −0.055 [95% CI, −0.103 to −0.006]; P = .03; global functioning: coefficient, −0.061 [95% CI, −0.112 to −0.011]; P = .02), lower premorbid IQ (global functioning: coefficient, −0.013 [−0.025 to −0.001]; P = .04), and higher baseline symptom severity (global cognition: coefficient, 0.006 [95% CI, 0.002 to 0.010]; P = .005) emerged as optimal candidates. CONCLUSIONS AND RELEVANCE: These findings show that CR is an evidence-based intervention that should be included consistently into clinical guidelines for the treatment of individuals with schizophrenia and implemented more widely in clinical practice.