Does insufficient adjustment for smoking explain the preventive effects of fruit and vegetables on lung cancer?

Recent reports have raised the question, whether the previously observed protective effects of high intake of fruit and vegetables on the risk of lung cancer were due to insufficient adjustment for smoking leading to residual confounding. Association of intake of fruit and vegetables on lung cancer risk was examined, using the Danish prospective cohort study, "Diet, Cancer and Health". Participants completed a food-frequency and lifestyle questionnaire, and age-standardized incidence rates and rate ratios were estimated for quartiles of dietary exposure. In 1993-2001, 247 out of the 54158 participants were diagnosed with lung cancer. The incidence rate of lung cancer was highest in the lowest quartile of intake of plant food (fruit, vegetables, legumes and potatoes) and the age-standardized rate ratio of lung cancer decreased significantly with increasing intake of plant food to 0.35 (95% CI, 0.27-0.45) but after control for smoking it was attenuated to 0.65 (95% CI, 0.45-0.93). The incidence rate differences of current smokers with high (> or = 400 g per day) and low (< 400 g per day) daily intake of plant food were independent of smoking intensity; assuming a true biological protective effect, 80-90 lung cancer cases per 100000 current smokers could be prevented in our cohort if all smokers had a high intake of plant food. The observed inverse association between high intakes of plant food seems chiefly to be a real protective effect, and not solely due to residual confounding.