BACKGROUND: Treatments for Barrett's oesophagus, the precursor lesion of adenocarcinoma, are available but whether these therapies effectively prevent the development of adenocarcinoma, and in some cases eradicate the Barrett's oesophagus segment, remains unclear. OBJECTIVES: To summarise, quantify and compare the efficacy of pharmacological, surgical and endoscopic treatments for the eradication of dysplastic and non-dysplastic Barrett's oesophagus and prevention of these states from progression to adenocarcinoma. SEARCH STRATEGY: We searched CENTRAL (The Cochrane Library 2004, issue 4), MEDLINE (1966 to June 2008) and EMBASE (1980 to June 2008). SELECTION CRITERIA: Randomised controlled trials (RCTs) comparing medical, endoscopic or non-resectional surgical treatments for Barrett's oesophagus. The primary outcome measures were complete eradication of Barrett's and dysplasia at 12 months, and reduction in the number of patients progressing to cancer at five years or latest time point. DATA COLLECTION AND ANALYSIS: Three authors independently extracted data and assessed the quality of the trials included in the analysis. MAIN RESULTS: Sixteen studies, including 1074 patients, were included. The mean number of participants in the studies was small (n = 49; range 8 to 208). Most studies did not report on the primary outcomes. Medical and surgical interventions to reduce symptoms and sequelae of gastro-oesophageal reflux disease (GORD) did not induce significant eradication of Barrett's oesophagus or dysplasia. Endoscopic therapies (photodynamic therapy (PDT with aminolevulinic acid or porfimer sodium), argon plasma coagulation (APC) and radiofrequency ablation (RFA)) all induced regression of Barrett's oesophagus and dysplasia. The data for photodynamic therapy were heterogeneous with a mean eradication rate of 51% for Barrett's oesophagus and between 56% and 100% for dysplasia, depending on the treatment regimens. The variation in photodynamic therapy eradication rates for dysplasia was dependent on the drug, source and dose of light. Radiofrequency ablation resulted in eradication rates of 82% and 94% for Barrett's oesophagus and dysplasia respectively, compared to a sham treatment. Endoscopic treatments were generally well tolerated, however all were associated with some buried glands, particularly following argon plasma coagulation and photodynamic therapy, as well as photosensitivity and strictures induced by porfimer sodium based photodynamic therapy in particular. AUTHORS' CONCLUSIONS: Despite their failure to eradicate Barrett's oesophagus, the role of medical and surgical interventions to reduce the troubling symptoms and sequelae of GORD is not questioned. Whether therapies for GORD reduce the cancer risk is not yet known. Ablative therapies have an increasing role in the management of dysplasia within Barrett's and current data would favour the use of radiofrequency ablation compared with photodynamic therapy. Radiofrequency ablation has been shown to yield significantly fewer complications than photodynamic therapy and is very efficacious at eradicating both dysplasia and Barrett's itself. However, long-term follow-up data are still needed before radiofrequency ablation can be used in routine clinical care without the need for very careful post-treatment surveillance. More clinical trial data and in particular randomised controlled trials are required to assess whether or not the cancer risk is reduced in routine clinical practice.