RESUMO
BACKGROUND: Several systematic reviews and meta-analyses evaluated the associations between dietary factors and the incidence of gastric cancer (GC). OBJECTIVES: To evaluate the strength and validity of existing evidence, we conducted an umbrella review of published systematic reviews and meta-analyses that investigated the association between diets and GC incidence. METHODS: We searched the PubMed, Embase, and Cochrane databases for systematic reviews and meta-analyses of prospective cohort studies investigating the association between dietary factors and GC risk. For each association, we recalculated the adjusted summary estimates with their 95% confidence interval (CI) and 95% prediction interval (PI) using a random-effects model. We used the I2 statistic and Egger's test to assess heterogeneity and small-study effects, respectively. We also assessed the methodological quality of each study and the quality of evidence. RESULTS: Finally, we identified 16 meta-analyses that described 57 associations in this umbrella review. Of the 57 associations, eight were statistically significant using random-effects, thirteen demonstrated substantial heterogeneity between studies (I2 > 50%), and three found small-study effects. The methodological quality of meta-analyses was classified as critically low for two (13%), low for thirteen (81%), and only one (6%) was rated as high confidence. Quality of evidence was rated high for a positive association for GC incidence with a higher intake of total alcohol (RR = 1.19, 95% CI 1.06-1.34) and moderate-quality evidence to support that increased processed meat consumption can increase GC incidence. Three associations (total fruit, vitamin E, and carotenoids) were determined to be supported by low-quality evidence, and two (pickled vegetables/foods and citrus fruit) were supported by very low-quality. CONCLUSIONS: Our findings support the dietary recommendations for preventative GC, emphasizing lower intake of alcohol and foods preserved by salting. New evidence suggests a possible role for total fruit, citrus fruit, carotenoids, and vitamin E. More research is needed on diets with lower quality evidence. REGISTRATION NUMBER: CRD42021255115.