[Source: The Lancet, full page: (LINK). Abstract, edited.]

The Lancet, Early Online Publication, 18 March 2014


Copyright ? 2014 Elsevier Ltd All rights reserved.

Tuberculosis prevalence in China, 1990?2010; a longitudinal analysis of national survey data

Original Text

Lixia Wang MS a ?, Hui Zhang PhD a ?, Yunzhou Ruan PhD a ?, Daniel P Chin MD b ?, Yinyin Xia PhD a, Shiming Cheng BS a, Mingting Chen MS a, Yanlin Zhao PhD a, Shiwen Jiang BS a, Xin Du BS a, Guangxue He PhD a, Jun Li MS a, Shengfen Wang PhD a, Wei Chen PhD a, Caihong Xu MS a, Fei Huang MS a, Xiaoqiu Liu MS a, Dr Yu Wang MD a



China scaled up a tuberculosis control programme (based on the directly observed treatment, short-course [DOTS] strategy) to cover half the population during the 1990s, and to the entire population after 2000. We assessed the effect of the programme.


In this longitudinal analysis, we compared data from three national tuberculosis prevalence surveys done in 1990, 2000, and 2010. The 2010 survey screened 252 940 eligible individuals aged 15 years and older at 176 investigation points, chosen by stratified random sampling from all 31 mainland provinces. All individuals had chest radiographs taken. Those with abnormal radiographs, persistent cough, or both, were classified as having suspected tuberculosis. Tuberculosis was diagnosed by chest radiograph, sputum-smear microscopy, and culture. Trained staff interviewed each patient with tuberculosis. The 1990 and 2000 surveys were reanalysed and compared with the 2010 survey.


From 1990 to 2010, the prevalence of smear-positive tuberculosis decreased from 170 cases (95% CI 166?174) to 59 cases (49?72) per 100 000 population. During the 1990s, smear-positive prevalence fell only in the provinces with the DOTS programme; after 2000, prevalence decreased in all provinces. The percentage reduction in smear-positive prevalence was greater for the decade after 2000 than the decade before (57% vs 19%; p<0?0001). 70% of the total reduction in smear-positive prevalence (78 of 111 cases per 100 000 population) occurred after 2000. Of these cases, 68 (87%) were in known cases?ie, cases diagnosed with tuberculosis before the survey. Of the known cases, the proportion treated by the public health system (using the DOTS strategy) increased from 59 (15%) of 370 cases in 2000 to 79 (66%) of 123 cases in 2010, contributing to reduced proportions of treatment default (from 163 [43%] of 370 cases to 35 [22%] of 123 cases) and retreatment cases (from 312 [84%] of 374 cases to 48 [31%] of 137 cases; both p<0?0001).


In 20 years, China more than halved its tuberculosis prevalence. Marked improvement in tuberculosis treatment, driven by a major shift in treatment from hospitals to the public health centres (that implemented the DOTS strategy) was largely responsible for this epidemiological effect.


Chinese Ministry of Health.

a Chinese Center for Disease Control and Prevention, Beijing, China; b China Office, The Bill & Melinda Gates Foundation, Beijing, China

Correspondence to: Dr Yu Wang, Chinese Center for Disease Control and Prevention, Beijing, 102206, China

? Authors contributed equally