A meta-analysis was conducted to summarize evidence from prospective cohort studies about the association of fruit and vegetable consumption with the risk of lung cancer.

Pertinent studies were identified by a search of Embase and PubMed databases to October 2014. A random-effects model was used to combine study-specific relative risks and 95% confidence interval [RR (95% CI)]. Dose-response relationship was assessed by restricted cubic spline.

The RR (95% CI) of lung cancer for highest versus lowest category of fruit and vegetable (FV) consumption was 0.87 (0.79-0.95) (8 studies including 12,942 cases among 1,571,494 subjects), and the effect was 0.84 (0.79-0.90) for fruit (16 studies including 15,421 cases among 1,791,469 subjects) and 0.90 (0.84-0.96) for vegetable (19 studies including 16,422 cases among 1,877,375 subjects). The above-mentioned associations did not differed significantly in subgroup analysis by country, age, number of covariates adjusted, quality score, sex, smoking status and histological subtypes; however, studies with follow-up duration of ≥10 years and with FV assessed by interview showed a stronger association than those of <10 years and by self-administrated food frequency questionnaires, respectively. The risk of lung cancer decreased by 3% (P=0.07), 5% (P<0.01) and 3% (P=0.09) for every 1serving/day increment in FV, fruit and vegetable consumption, respectively. There was a threshold around 2servings/day of fruit and 2servings/day of vegetable, respectively, after which the risk of lung cancer did not reduce further.

Fruit and vegetable consumption are inversely associated with risk of lung cancer.