OBJECTIVE: To determine effects of physical activity on depressive symptoms (primary objective), symptoms of schizophrenia, anthropometric measures, aerobic capacity, and quality of life (secondary objectives) in people with mental illness and explore between-study heterogeneity. DATA SOURCES: MEDLINE, Cochrane Controlled Trials Register, PsycINFO, CINAHL, Embase, and the Physiotherapy Evidence Database (PEDro) were searched from earliest record to 2013. STUDY SELECTION: Randomized controlled trials of adults with a DSM-IV-TR, ICD-10, or clinician-confirmed diagnosis of a mental illness other than dysthymia or eating disorders were selected. Interventions included exercise programs, exercise counseling, lifestyle interventions, tai chi, or physical yoga. Study methodological quality and intervention compliance with American College of Sports Medicine (ACSM) guidelines were also assessed. DATA EXTRACTION AND ANALYSIS: Two investigators extracted data. Data were pooled using random-effects meta-analysis. Meta-regression was used to examine sources of between-study heterogeneity. RESULTS: Thirty-nine eligible trials were identified. The primary meta-analysis found a large effect of physical activity on depressive symptoms (n = 20; standardized mean difference (SMD) = 0.80). The effect size in trial interventions that met ACSM guidelines for aerobic exercise did not differ significantly from those that did not meet these guidelines. The effect for trials with higher methodological quality was smaller than that observed for trials with lower methodological quality (SMD = 0.39 vs 1.35); however, the difference was not statistically significant. A large effect was found for schizophrenia symptoms (SMD = 1.0), a small effect was found for anthropometry (SMD = 0.24), and moderate effects were found for aerobic capacity (SMD = 0.63) and quality of life (SMD = 0.64). CONCLUSIONS: Physical activity reduced depressive symptoms in people with mental illness. Larger effects were seen in studies of poorer methodological quality. Physical activity reduced symptoms of schizophrenia and improved anthropometric measures, aerobic capacity, and quality of life among people with mental illness. TRIAL REGISTRATION: PROSPERO registration #CRD42012002012.