Background: Patients with medically unexplained symptoms (MUS) are difficult to treat and cause high health-care costs. Psychological interventions might be a beneficial option for treating patients with MUS, but evidence is inconsistent. This meta-analysis compares the effectiveness of psychological interventions for MUS - delivered either by psychotherapists (PTs) or by general practitioners (GPs) - with that of usual care. Method: We conducted a systematic review and meta-analysis on randomised controlled trials of psychological interventions for MUS. Physical symptoms were the primary outcome, and physical functioning and psychological symptoms were the secondary outcomes. We pooled between-group effect sizes (ESs) after the treatment and at the follow-up in random-effects meta-regressions and stratified meta-analyses. We repeated these analyses with the intervention provider, intervention dose, MUS severity and methodological quality as predictors of relative intervention effects. Results: A total of 3,225 patients in 20 studies were analysed. After the treatment, small and significant ESs were found for all 3 outcome domains (ES range: 0.13-0.19, all p < 0.05). Psychological interventions were more beneficial for physical symptoms when delivered by PTs than by GPs (p = 0.02). There was no difference between PTs and GPs in terms of physical functioning and psychological symptoms. Conclusion: Psychological interventions are effective for patients with MUS, but the effects are small and most likely of short duration. Interventions that are delivered by PTs appear to have larger effects on unexplained physical symptoms than those delivered by GPs. Whether this superiority is due to a larger number of sessions of PT interventions remains unclear from our findings.