Author

T. Yates

Abstract

Tuberculosis is a major cause of premature mortality. Communities in Southern Africa are disproportionately affected. A growing body of evidence suggests that recent transmission within households can explain only a limited proportion of tuberculosis disease. However, our understanding of where transmission between households occurs is limited. I undertook a systematic review and meta-analysis of molecular epidemiology studies that described rates of strain discordance in co-prevalent cases of tuberculosis resident in the same household. I also conducted a tuberculin school survey in 6-8 year old children in a rural community in Northern KwaZulu-Natal. These children were all registered in a household surveillance programme operated by the Africa Centre for Population Health. I found that, across a range of both high and low burden countries, co-prevalent cases of tuberculosis in the same household often have different strains of Mycobacterium tuberculosis. These molecular epidemiological data suggest, at least in some settings, that recent transmission within households may explain a modest proportion of tuberculosis disease. I estimated the annual risk of tuberculous infection to be approximately two percent in the community around the Africa Centre. I found weak evidence that exposure to HIV positive adults in the household was associated with Mycobacterium tuberculosis infection in children. I found no strong evidence associating use of specific indoor public spaces with Mycobacterium tuberculosis infection. Transmission between households is likely an important determinant of tuberculosis disease. Further research locating Mycobacterium tuberculosis transmission might enable TB control interventions to be better targeted.