Prevalence of tuberculosis infection among South African adolescents

Abstract

Background: We aimed to measure changes in adolescent prevalence of latent tuberculosis (TB) infection (LTBI) over a 10-year period (2005-2015), in which TB disease incidence fell 34% from 1,038 – 682 per 100,000 per annum, in order to evaluate impact of effective TB control efforts on transmission to children. Methods & Materials: We compared baseline data from a TB cohort study (2005-2007) and a TB vaccine trial (2014-2015), which enrolled adolescents from the same 8 high schools near Cape Town, South Africa. LTBI was defined by positive QuantiFERON®-TB Gold In-Tube (QFT) test. Data were analysed by two-sample test of equality of proportions for survey data and logistic regression for survey data. Results: We analysed data from 6,848 adolescents with median age of 15.5 years (range 12.3 - 19.1). 4,880 (71%) adolescents were tested between 2005-2007 and 1,968 (29%) between 2014-2015. Prevalence of LTBI was 43.8% (95% Confidence Interval (CI) 28.4-59.1) vs 48.5% (CI: 41.1-55.8); and annual risk of Mycobacterium tuberculosis infection (ARTI) was 3.6% (CI: 2.1-5.5) vs 4.2% (CI: 3.4-5.2), in the period 2005-2007 vs 2014-2015, respectively. Marked heterogeneity in LTBI prevalence was observed at school level, ranging from 12-65%. LTBI prevalence increased on average by 13% between age 12-18 years in lower socio-economic quintile schools, in which average LTBI was 54% in 2005-2007 and 53% in 2014-2015, consistent with ongoing, but stable risk of TB transmission. By contrast, LTBI prevalence did not increase with age in highest socio-economic quintile schools, in which average LTBI prevalence increased from 20-38% over the same period, consistent with increased risk of transmission prior to school entry. Conclusion: Adolescent LTBI prevalence remained high over a decade, suggesting TB transmission to children was not yet impacted by effective TB control efforts that had reduced regional TB disease notification rate by more than one-third. Trends in school LTBI prevalence should be interpreted in the context of socio-economic and other TB risk factors that affect risk of transmission prior to school entry.