Abstract: Objective  To investigate prevalences of pulmonary tuberculosis among different regions in China.  Methods  Multi-stage stratified clustered sampling method was adopted, 176 clusters were selected. All 252 940 residents over 15 years old received investigation and took chest X-ray films. Sputum smear and culture were conducted for 9825 patients with abnormal X-ray film or tuberculosis related symptoms. Strains isolated via culture were taken an identification test to identify M. tuberculosis complex and non-tuberculosis mycobacteria for 380 patients. A weighted adjustment method accounting for complex sampling was used to calculate prevalence and 95% confidence intervals.Second-order Rao-Scott adjusted chi-square test was used to compare prevalences among different groups.  Reporting incidences were abstracted from Chinese Infectious Disease Detection Report Information System to compare with prevalences in different regions.  Results  The weighted prevalence of active, and bacteriological positive pulmonary tuberculosis was 832/100 000(497/51 333), 241/100 000(145/51 333) respectively in rural areas in western provinces, significantly higher than those in urban areas of western provinces［286/100 000(70/18 563), 70/100 000(16/18 563)］，χ²=148.16（P<0.01）and 13.82（P<0.01）respectively. While for the prevalence of smear positive pulmonary tuberculosis in western provinces, there were no difference found between rural and urban areas（χ²=3.16，P=0.08）. In eastern rural areas, the weighted prevalence of active, smear positive and bacteriological positive pulmonary tuberculosis（364/100 000（190/41 763），62/100 000（31/41 763），91/100 000（47/41 763）） were all significantly higher than those in urban areas（228/100 000（150/60 355），28/100 000（18/60 355），44/100 000（28/60 355））, χ²=17.61（P<0.01）,5.19(P=0.03) and 6.90(P=0.01) separately. In middle provinces, no difference was found between rural and urban areas（χ²=0.35，P=0.56；χ²=1.18，P=0.28；χ²=0.01，P=0.91）respectively for comparison of weighted prevalence of active, smear positive and bacteriological positive pulmonary tuberculosis. Ratio of prevalence and reporting incidence of smear positive pulmonary tuberculosis in western provinces was 284：1, which was higher than that in eastern provinces(1.47：1) and in middle provinces(1.50：1).  Survey sites with the highest number of active pulmonary tuberculosis patients detection were all located in remote areas.  Conclusion  The prevalences of tuberculosis are still high in western, rural or remote areas, which should be regarded as key areas for enhanced tuberculosis control.

Key words: Tuberculosis, pulmonary/epidemiology, Prevalence, Small-area analysis