A meta-analysis on the association between smoking and tuberculosis

Objective To assess quantitatively the association between smoking and the risk of disease and death from pulmonary tuberculosis(TB) by a meta-analysis. Methods Both Chinese and English literatures published from Jan. 1999 to Jun. 2010 about the case-control study of the association between smoking and pulmonary TB were retrieved by searching the electronic resource databases. A meta-analysis was then performed on the comparison and synthesis of findings from included studies. Pooled odds ratio (OR) and 95% confidence interval (CI) were calculated using a software RevMan 4.2. Finally,possible publication bias was tested and a sensitivity analysis was conducted. Results The meta-analysis for total 22 high-quality case-control studies found a pooled OR of 2.56 (95% CI 1. 80 ～3.64) with a random-effects model. Sensitivity analysis showed that pooled OR was 2.56 (95% 1.76 ～3.71 ) by removing two papers of nested case-control study. The OR were 2.13 (95% CI1.76～2.58) and 4.60 (95% CI 1.81～11.69) for disease and death from TB,respectively. After confounding factors,including age and gender,had been controlled the OR was 2.53 (95% CI 1.27 ～5.03 ) whereas 1. 99(95% CI 1.54 ～ 2.56 ) without control of them. Conclusion Smoking can increase the risk of incidence and death of TB. TB control programs and treatment of patients might benefit from interventions aimed at reducing tobacco exposure.