Identification of populations at high risk of tuberculosis infection in rural China: a population-based, multicentre, prospective study

Abstract

BackgroundAcquisition and persistence of tuberculosis infection are important epidemiological measurements for evaluation of tuberculosis transmission in a community. However, prospective population data on the incidence of tuberculosis infection have been sparely reported in the global literature. MethodsWe conducted a population-based, multicentre, prospective study to investigate the annual risk of tuberculosis infection and its persistence in rural populations in China from July 2013 to June 2015, using yearly tests with tuberculin skin test (TST) and an interferon-γ release assay (IGRA). Findings15 907 eligible participants from four rural sites (Xiangtan, Hunan province; Zhongmu, Henan Province; Longxi, Gansu Province; Danyang, Jiangsu Province) who were identified as TST-negative (<10 mm) or IGRA-negative in a baseline survey were included in the first year's follow-up examination. 13 580 (85·4%) with completed questionnaires and tests results were included in this study. Annual seroconversion rates among the study sites ranged from 2·1% to 4·9% (average 3·1%) when assessed by IGRA and between 6·0% and 31·1% (14·5%) when assessed by TST. Infection persistence was successfully followed-up in 330 participants who underwent IGRA conversion during the second year of follow-up, 49·7% (164) were found to be consistently IGRA-positive. Both conversion and persistence of IGRA-positivity were increased significantly in association with increasing age (p<0·001 for trend). History of close contact with patients with active tuberculosis and smoking were positively associated with risk of IGRA conversion (adjusted OR 1·54, 95% CI 1·04–2·27, and 1·53, 1·13–2·05, respectively). InterpretationConsistent with published cross-sectional data, the prospective results provide stronger evidence that close contacts, older people, and smokers are high-risk populations for tuberculosis infection in rural China. These groups should be prioritised for tuberculosis control in China and areas with a similar epidemiological profile. The unexpectedly high annual infection rates based on TST could not be explained by local epidemiology and were probably overestimated because of the boosting effect of serial TST testing, BCG vaccination, or environmental mycobacteria. FundingNational Science and Technology Major Project of China (2013ZX10003004-002, 2014ZX10003001-001), and Program for Changjiang Scholars and Innovative Research Team in University of China (IRT13007).