The efficacy of physical exercise on body mass index in schizophrenia patients: A systematic review and meta-analysis on exercise prescription, intensity, and delivery.

Patients with schizophrenia exhibit a significantly elevated prevalence of overweight and obesity, which contribute to serious cardiometabolic disease and premature death. Although physical exercise is considered a promising add-on treatment, its specific impact on BMI and the most effective components of the intervention remain unclear. We conducted a systematic review and meta-analysis of randomised controlled trials to determine whether physical exercise reduces BMI in this population, and to examine how exercise prescription, delivery mode, and exercise dose (intensity and duration) influence outcomes. PubMed, Embase, Web of Science, and Cochrane Library were searched from inception to December 2025. Data were pooled using a random-effects model, and subgroup analyses were conducted for exercise prescription (aerobic, anaerobic, concurrent), mode (individual, group), intensity (low-to-moderate, moderate-to-vigorous), and duration (<20 weeks, ≥20 weeks). Nine trials with 356 participants were included. Physical Exercise interventions yielded a statistically significant, small-to-moderate reduction in BMI. Subgroup analyses showed a significant effect for aerobic exercise; however, no clear effect was observed for anaerobic or concurrent training. Meanwhile, higher-dose regimens of moderate-to-vigorous intensity and interventions lasting ≥20 weeks appeared to significantly influence BMI. Importantly, individual-based interventions demonstrated meaningful effects, whereas group-based interventions tended to have less pronounced results. Therefore, structured physical exercise, particularly moderate-to-vigorous aerobic training for ≥20 weeks, appears to be an effective intervention for reducing BMI in patients with schizophrenia. The findings indicate an interesting pattern: individually delivered protocols yielded statistical efficacy, whereas group formats showed descriptively lower dropout rates, suggesting potentially better feasibility for implementation.