Epidemiology and control of tuberculosis in the antiretroviral therapy era: towards a mathematical model for Cape Town, South Africa


This project aimed to use a variety of epidemiologic methods to describe the local TB epidemic and its drivers, and to determine the potential impact of different public health interventions in Cape Town, South Africa. State-of-the-art statistical techniques were used to perform an in-depth analysis of all TB cases in Cape Town in the last decade. These data were used to develop and validate a mathematical model of the local TB epidemic. Projections of this model were used to estimate the future impact of new and redesigned public health strategies for TB control.


As a Marie Curie International Outgoing Fellowship, it would also provide an opportunity for the postdoctoral researcher to expand her epidemiological and statistical skills, to broaden her expertise in TB and HIV co-infection research, to gain experience in multidisciplinary collaborations including with regional health authorities and to further develop her transferable skills.


The resurgence of tuberculosis (TB) has been underway for almost 20 years, and despite increasingly implemented comprehensive control strategies, the epidemic in sub-Saharan Africa is continuing to expand. There is an urgent need to elucidate the reasons underlying the failing of TB control and to develop innovative public health approaches.


This project was completed in March 2017. It found that the trajectory of TB incidence in Cape Town over the last 100 years, compared to London and New York, was very different. The introduction of chemotherapy did not reduce the TB epidemic in Cape Town. In addition, even before the advent of the HIV epidemic, TB rates were already at the level of the beginning of the century.The surge following the HIV epidemic has led to the life-time incidence of TB among the population in Cape Town to further increase over the 25% before HIV.


The most important finding was that the burden of recurrent TB in Cape Town is very high with over 30% of TB patients developing one or more additional episodes over a period of 12 years. The risk of another episode of TB increased greatly with every subsequent episode. Looking further in depth at the last 13 years, the analyses show a reduction in TB rates over the last five years which may be due to antiretroviral therapy roll-out, however the ecological study design did not allow for attribution of causality. An alternative or complementary explanation may be a decline in empirical treatment rates, which we identified in an evaluation of the impact of the roll-out of a new rapid molecular diagnostic test. We developed an age-stratified mathematical model of TB in Cape Town, which confirmed that the TB burden is determined by the underlying age structure, but that a protective effect of latent infection by previous TB strains as well as rates of progression to active TB disease after a prior episode of TB also play an important role.


The postdoctoral researcher spent two years at the Desmond Tutu HIV Centre at the University of Cape Town in South Africa and one year at AIGHD.


All data required were successfully collected and almost all analyses were completed. The mathematical model of TB in Cape Town was developed and validated. These achievements led to a current total of eight presentations of the results at international conferences, two presentations at local or regional meetings and five publications in peer-reviewed journals. An additional eight publications emanated from a supervisory or collaboratory role of the postdoctoral researcher. Three manuscripts are still in the process of write-up or peer review.

AIGHD Research Lead

Prof. Frank G.J. Cobelens


Desmond Tutu HIV Centre at the University of Cape Town
Centre for Infectious Diseases Epidemiology Research (CIDER) at the Faculty of Public Health and Family Medicine, University of Cape Town
Institute of Social and Preventative Medicine (ISPM) at the University of Bern


European Commission


South Africa