Network meta-analysis (NMA) is a statistical technique for making direct and indirect comparisons between different treatment and control groups. Despite its many advantages, NMA may be misleading when evaluating networks that are disconnected, inconsistent or of low reliability and validity. We review how well the analysis of trials of adjunctive psychosocial treatment in bipolar disorder is served by NMA. We conclude with recommendations for future treatment trials in bipolar disorder and guidelines for NMAs.
