RESUMO
OBJECTIVE: To investigate the relationship between physical activity and lung cancer among smokers and whether this relationship differed according to physical activity intensity, smoking status, and gender. DESIGN: Meta-analysis. METHODS: A computerized bibliographical search was conducted in five databases. Study inclusion criteria were: (i) the study population was not diagnosed with lung cancer at baseline; (ii) the study provided information concerning the effect size of physical activity on the risk of developing lung cancer in smokers; and (iii) the study distinguished different physical activity intensity levels. Two authors independently extracted data and assessed the methodological quality. Pooled rate ratios (RR) were calculated for all data, and for subgroups of physical activity intensity, smoking status, and gender. RESULTS: Pooled RRs of 7 cohort studies showed that physical activity was associated with a reduced risk of lung cancer in smokers (RR=0.82, 95% CI=0.77; 0.87). We did not find clear dose-response relationship regarding exercise or smoking intensity, i.e. high levels of physical activity did not show a higher risk reduction than moderate physical activity levels, and the association between physical activity and risk reduction did not differ between heavy and light smokers. The reduced risk associated with physical activity was greater in women than in men (p=0.03), but this finding was based on only one study that reported data on women. CONCLUSIONS: Results of this meta-analysis indicate that leisure time physical activity is associated with reduced risk of developing lung cancer among smokers. Future studies should provide insight into a potential dose-response relationship, and should use reliable and valid physical activity measurements.