Chronic pain in childhood is an international public health problem. We conducted a systematic review and meta-analysis to provide a summary of the published evidence of pharmacological, physical, and psychological therapies for children with chronic pain conditions. We searched CENTRAL, MEDLINE, EMBASE, and PsycINFO from inception to April 2020, clinical trial registries, and other sources for randomised controlled trials or comparative observational trials. We extracted critical outcomes of pain intensity, quality of life, physical-, role-, and emotional functioning, sleep, and adverse events. We assessed studies for risk of bias and certainty of the evidence using GRADE. We included 34 pharmacological (4091 participants), 25 physical therapy (1470 participants), and 63 psychological trials (5025 participants). Participants reported a range of chronic pain conditions. Most studies were assessed to have unclear or high risk of bias across multiple domains. Pharmacological, physical, and psychological therapies showed some benefit for reducing pain, post-treatment but only physical and psychological therapies improved physical functioning. We found no benefit of any treatment modality for health-related quality of life, role functioning, emotional functioning, or sleep. Adverse events were poorly reported particularly for psychological and physical interventions. The largest evidence-base for the management of chronic pain in children supports the use of psychological therapies, followed by pharmacological and physical therapies. However, we rated most outcomes as low or very low-certainty, meaning further evidence is likely to change our confidence in the estimates of effects. This protocol was registered on PROSPERO (CRD42020172451).