BACKGROUND: Progress in tuberculosis control in China has been the slowest in western areas, which have the highest prevalence. We assessed the prevalence of pulmonary tuberculosis in the Xinjiang province, China, 10 years after introduction of a control programme based on directly observed treatment, short course. METHODS: In this population-based, cross-sectional survey, we used a multistage stratified random cluster sample design to estimate the prevalence of smear-positive and bacteriologically confirmed (either smear positive or culture positive, or both) pulmonary tuberculosis among adults (aged ≥15 years) in Xinjiang who had been resident in their household for the last 6 months. The screening strategy and diagnosis followed WHO guidelines. We estimated prevalence by combining inverse probability weighting and multiple imputation of missing data. We compared our prevalence survey estimates with the ones from the 2010 China national pulmonary tuberculosis survey and the ones from a provincial pulmonary survey done in Xinjiang in 2000. The new smear-positive pulmonary tuberculosis notification rate in 2011 in Xinjiang was obtained to allow the calculation of patient diagnosis rate (PDR). FINDINGS: Between Sept 1, 2010, and July 31, 2011, 31 081 individuals were eligible, of whom 29 835 (96·0%) participated in the survey. We identified 50 (0·2%) smear-positive and 101 (0·3%) bacteriologically confirmed pulmonary tuberculosis cases. The weighted prevalence of smear-positive pulmonary tuberculosis was 170 (95% CI 103-233) per 100 000 people and of bacteriologically confirmed pulmonary tuberculosis was 430 (249-611) per 100 000 people. Compared with 2000 Xinjiang survey estimates, the prevalence of smear-positive pulmonary tuberculosis has decreased by 26·4% (from 231 [95% CI 148-314] per 100 000 people), whereas the prevalence of bacteriologically confirmed pulmonary tuberculosis has increased by 17·8% (from 365 [237-493] per 100 000 people). In each age group and sex, the pulmonary tuberculosis prevalence was higher in the 2010-11 Xinjiang survey than in the 2010 national survey. The PDR in 2011 was 0·34 (95% CI 0·25-0·44). INTERPRETATION: Despite progress in other parts of China, the prevalence of pulmonary tuberculosis in Xinjiang remains high. The very low PDR suggests poor access to diagnosis and care. Further studies are needed to understand the barriers to diagnosis and care of this population, and efforts are urgently needed to enhance tuberculosis screening in this area. FUNDING: Xinjiang Uyghur Autonomous Region Health Bureau.