OBJECTIVETo investigate the association between obesity and the risk of lung cancer and evaluate a dose-response relationship between body mass index (BMI) and incidence risk of lung cancer in the Chinese population.

METHODSA systematic literature search for BMI and incidence risk of lung cancer in the Chinese population, as well as through the reference lists of retrieved articles. The literature databases including Chinese National Knowledge Infrastructure, Wanfang, PubMed, Embase and Google Scholar. Time range was from the founding of each database to September 2014 and a total of 93 research papers were collected. Meta-analysis was conducted to calculate pooled odds ratio and corresponding 95% CI. Generalized least-squares regression methods were used to make a dose-response meta-analysis between BMI and incidence risk of lung cancer.

RESULTSSeven studies were included in the meta-analysis, with a number of 2 351 lung cancer cases. Results showed that obesity was inversely associated with lung cancer incidence (OR = 0.68, 95% CI: 0.59-0.79) (heterogeneity test: I² = 0, P = 0.594). The association did not change with stratification by study design, sex, smoking status, BMI measurement method and study population. A linear dose-response association between BMI and risk of lung cancer was visually significant, and lung cancer risk would be reduced 21% for per 5 kg/m² BMI increase (OR = 0.79, 95% CI: 0.71-0.89) (heterogeneity test: q = 22.43, P = 0.002).

CONCLUSIONThe results of this meta-analysis indicated that higher BMI was a protective factor against lung cancer, but smoking may play a stronger role as a confounding factor for the most important role with lung cancer incidence.