BACKGROUND: Obesity is associated with an increased incidence of at least 13 types of cancer. Although bariatric surgery has been associated with a reduced risk of hormonal cancers, data for non-hormonal cancers are scarce. The aim of this study was to evaluate the effect of bariatric surgery on the incidence of non-hormonal cancers. METHODS: Cochrane, Embase, PubMed, Scopus, and Web of Science were searched for articles between 1984 and 2018, following the PRISMA system. Meta-analysis was conducted using a random-effect model with subgroup analysis by procedure and cancer type. RESULTS: From 2526 studies screened, 15 were included. There were a total of 18 583 477 patients, 947 787 in the bariatric group and 17 635 690 in the control group. In comparison to the non-surgical group, the bariatric group had a lower incidence of cancer (OR .65 (95 per cent c.i. 0.53 to 0.80); P < 0.002). In the subgroup analysis, Roux-en-Y gastric bypass and sleeve gastrectomy were associated with decreased risk of developing cancer, while no difference was observed with adjustable gastric banding. When evaluated by cancer type, liver (OR 0.417 (95 per cent c.i. 0.323 to 0.538)), colorectal (OR 0.64 (95 per cent c.i. 0.49 to 0.84)), kidney and urinary tract cancer (OR 0.77 (95 per cent c.i. 0.72 to 0.83)), oesophageal (OR 0.60 (95 per cent c.i. 0.43 to 0.85)), and lung cancer (OR 0.796 (95 per cent c.i. 0.45 to 0.80)) also presented a lower cancer incidence in the bariatric group. CONCLUSION: Bariatric surgery is related to an almost 50 per cent reduction in the risk of non-hormonal cancers.