Objective: To investigate the association between vegetable and fruit consumption and incidence of lung cancer.
Methods: Self-administered questionnaires were used to assess diet in two large population-based cohorts with 42,224 and 51,114 subjects in 1990 and 1993, respectively. After ten and seven years of follow-up, we ascertained 428 newly diagnosed case of lung cancer. Relative risk (RR) estimates were calculated using the Cox proportional hazards model with pooling of estimates from the two cohorts.
Results: Total vegetable and fruit intake was not associated with lowered risk of lung cancer, with RR approximating unity. The null relation between vegetable and fruit consumption and lung cancer incidence was consistent across strata of smoking status (never or ever smokers). When dividing lung cancers into adenocarcinoma and non-adenocarcinoma, risk for middle and high intakes of vegetables only, fruit only, and vegetables and fruit combined were all below one for non-adenocarcinoma and above one for adenocarcinoma, although no statistically significant differences were noted. Similar patterns of results were found when the two cohorts were analyzed separately.
Conclusions: Contrary to popular belief, our results suggest that vegetables and fruit do not appear to confer protection from lung cancer.