Using Mendelian randomization (MR) approach, our objective was to determine whether there was a causal association between dietary factors and gastroesophageal reflux disease (GERD), Barrett's esophagus (BE), or esophageal cancer (EC).Genome-wide association study (GWAS) data for eighteen types of dietary intake were obtained from the UK Biobank. GWAS data for GERD, BE, and EC were sourced from the FinnGen consortium. We performed univariable and multivariable MR analysis to assess the cause effect between dietary factors and esophageal diseases. MR results were expressed as odds ratios (OR) with 95% confidence intervals (CI).Raw vegetable intake was associated with a lower risk of GERD (OR = 0.478; P = 0.011). On the contrary, cooked vegetable intake increased the risk of GERD (OR = 1.911; P = 0.024). Bread intake was associated with increased odds of BE (OR = 6.754; P = 0.007), while processed meat intake was associated with reduced risk of BE (OR = 0.210; P = 0.035). We also observed evidence that increased consumption of dried fruit (OR = 0.087; P = 0.022) and salt added to food (OR = 0.346; P = 0.045) could prevent EC. The results of multivariable MR showed that the protective effect of consumption of salt added to food on EC was no longer significant after adjusting for the consumption of dried fruit.Vegetable consumption was associated with GERD, whereas consumption of bread and processed meat was associated with BE. Dried fruit intake was associated with a lower risk of EC, and the protective effect of consumption of salt added food on EC may also be mediated by consumption of dried fruit. Future research should be performed to investigate the mechanisms behind these cause-and-effect relationships to reduce the burden of disease caused by dietary habits.Copyright: © 2023 Zou et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.