There are few epidemiological data on the dietary risk factors of Barrett's oesophagus, a precursor of oesophageal adenocarcinoma. The present study investigated the association between vegetable, fruit and nitrate intake and Barrett's oesophagus risk in a large prospective cohort. The Netherlands Cohort Study recruited 120852 individuals aged 55-69 years in 1986. Vegetable and fruit intake was assessed using a 150-item FFQ, and nitrate intake from dietary sources and drinking water was determined. After 16 center dot 3 years of follow-up, 433 cases (241 men and 192 women) of Barrett's oesophagus with specialised intestinal metaplasia and 3717 subcohort members were analysed in a case-cohort design using Cox proportional hazards models while adjusting for potential confounders. Men exhibited a lower risk of Barrett's oesophagus in the highest v. the lowest quintile of total (multivariable-adjusted hazard ratio (HR): 0 center dot 66, 95% CI 0 center dot 43, 1 center dot 01), raw (HR 0 center dot 63, 95% CI 0 center dot 40, 0 center dot 99), raw leafy (HR 0 center dot 55, 95% CI 0 center dot 36, 0 center dot 86) and Brassica (HR 0 center dot 64, 95% CI 0 center dot 41, 1 center dot 00) vegetable intake. No association was found for other vegetable groups and fruits. No significant associations were found between vegetable and fruit intake and Barrett's oesophagus risk among women. Total nitrate intake was inversely associated with Barrett's disease risk in men (HR 0 center dot 50, 95% CI 0 center dot 25, 0 center dot 99) and positively associated with it in women (HR 3 center dot 77, 95% CI 1 center dot 68, 8 center dot 45) (P for interaction=0 center dot 04). These results suggest that vegetable intake may contribute to the prevention of Barrett's oesophagus. The possible differential effect in men and women should be evaluated further.