Abstract: Objective To assess the correlation between body mass index and lung cancer risk by means of Meta-analysis. Methods Relevant studies were identified by a through literature search. Results A total of 18 studies were used for this analysis.9 case-control studies with 4 246 cancer cases and 34 771 controls,and 9 prospective studies with 2 434 947 person years and 8 899 cancer cases were included.For case-control studies,a combined OR of 1.4295％CI 1.06,1.91,0.8095％CI 0.64,1.01 and 0.9495％CI 0.67,1.31,respectively,for lower,middle and higher BMI.For cohort studies,a combined RR of 1.5995％CI 1.20,2.12,1.02 95％CI 0.96,1.08 and 0.7895％CI 0.63,0.95,respectively,for lower,middle and higher BMI.In the studies about overweight and obesity,BMI was inverse associated with lung cancer risk.For those non-smoking population,there was no statistical relationship between BMI and lung cancer risk. Conclusion Our findings suggest inverse association between BMI and lung cancer risk,but it can't exclude the mixed impact of smoking.