Abstract Purpose To evaluate the effectiveness of peer-delivered interventions in improving clinical and psychosocial outcomes among individuals with severe mental illness (SMI) or depression. Methods Systematic review and meta-analysis of randomised controlled trials comparing a peer-delivered intervention to treatment as usual or treatment delivered by a health professional. Random effect meta-analyses were performed separately for SMI and depression interventions. Results Fourteen studies (10 SMI studies, 4 depression studies), all from high-income countries, met the inclusion criteria. For SMI, evidence from three high-quality superiority trials showed small positive effects favouring peer-delivered interventions for quality of life (SMD 0.24, 95 % CI 0.08–0.40, p = 0.003, I ² = 0 %, n = 639) and hope (SMD 0.24, 95 % CI 0.02–0.46, p = 0.03, I ² = 65 %, n = 967). Results of two SMI equivalence trials indicated that peers may be equivalent to health professionals in improving clinical symptoms (SMD −0.14, 95 % CI −0.57 to 0.29, p = 0.51, I ² = 0 %, n = 84) and quality of life (SMD −0.11, 95 % CI −0.42 to 0.20, p = 0.56, I ² = 0 %, n = 164). No effect of peer-delivered interventions for depression was observed on any outcome. Conclusions The limited evidence base suggests that peers may have a small additional impact on patient’s outcomes, in comparison to standard psychiatric care in high-income settings. Future research should explore the use and applicability of peer-delivered interventions in resource poor settings where standard care is likely to be of lower quality and coverage. The positive findings of equivalence trials demand further research in this area to consolidate the relative value of peer-delivered vs. professional-delivered interventions.