Determining the importance of different locations to Mycobacterium tuberculosis transmission in high tuberculosis burden settings

Over nine and a half million people worldwide developed tuberculosis (TB) in 2014, and 1.5 million died from the disease. Current control strategies are insufficient, with global TB incidence falling by only 1-2% per year. Despite its importance as a global public health problem, very little is known about where Mycobacterium tuberculosis transmission (M.tb., the bacterium that causes the majority of TB disease) occurs, particularly in high burden countries such as South Africa. This is primarily because it is difficult to know whether someone has been recently infected, and it can be decades between infection and disease. Studies of social contacts have been useful in giving some indication of which locations may be important to M.tb. transmission. However, these studies are insufficient to estimate where the majority of transmission occurs, for three main reasons:

1) There is a lot of variation in the infectiousness of people with TB disease, and in how susceptible people are to infection and disease (e.g. people living with HIV are at increased risk). For this reason, chance meetings between the most infectiousness and susceptible people are likely be very important to overall transmission in a population. These meetings are most likely to occur in the places where numbers of new people met are highest, but these locations may not be the places where people spend most time, or where people meet the most other people per visit.
2) Transmission is likely to occur more often in crowded, badly ventilated buildings with little natural light.
3) Transmission is likely to be more common in places that are both visited by people with TB, and also visited by the people most susceptible to TB (e.g. clinics).

In this study, we will develop a better understanding of where M.tb. transmission occurs in high TB burden countries, and use this information to inform strategies to reduce TB death and suffering. We will use Cape Town, South Africa, as a case study. We have four main tasks:

1) Develop a model of people's contacts and infection transmission to investigate how variation in susceptibility and infectiousness affects where transmission occurs, for M.tb. and other infections.
2) Collect new data using questionnaires that will allow us to estimate the proportion of contacts that are between people who have not met before, by location. We will also collect data on the levels of ventilation and natural light in different types of location.
3) Using the new data we have collected, as well as existing data, build a model of M.tb. transmission by location. Using this model, we will estimate the proportion of transmission that occurs in different locations in Cape Town.
4) Adapt the model to represent contact patterns, susceptibility levels (e.g. HIV prevalence), and levels of M.tb. transmission in other high burden sub-Saharan African settings. We can then estimate in what types of location transmission may be common in other countries. We will also investigate what the most effective locations are to target using infection control interventions (e.g. improving ventilation or installing ultraviolet (U.V.) lights).

This study will help to fill a major gap in our understanding of TB: where does transmission occur? This has very real, practical implications. Trials of U.V. lights suggest that they may reduce transmission by 70-80%, and improving ventilation can also be very effective in preventing infections. These infection control measures are often implemented in TB clinics, but we currently do not know where else they should be targeted. Our results will inform the targeting of infection control measures, and if successful, should lead to reductions in M.tb. transmission, and in TB death and suffering. Our new understanding should also contribute to meeting the 2025 WHO End TB Strategy targets in high TB burden settings, the settings most in need of novel, innovative, and cost effective prevention strategies.

Currently, little is known about where M.tb. transmission occurs in high incidence settings, limiting the potential effectiveness of infection control interventions. The aim of this study is to estimate the proportion of transmission that occurs in different locations in Cape Town and other high incidence, sub-Saharan African settings, and to investigate the implications for TB control. There are four objectives:

1) Use network modelling to determine the effect of variation in infectiousness and susceptibility on the contribution of different location types to transmission.
2) Conduct primary data collection and analysis to inform model parameterisation. This objective has three parts: A. Estimate numbers of unique and repeat contacts by location (primary data collection using a questionnaire). B. Determine the suitability of different types of location for M.tb. transmission (data collection in buildings, including measuring CO2 and UV levels). C. Determine how contacts vary between people with different levels of behavioural and biological susceptibility to M.tb. infection and progression, by location (analysis of existing data).
3) Develop an individual-based model of contact behaviour and M.tb. transmission, and estimate the proportion of transmission that occurs in different locations in Cape Town.
4) Generalise the model to other settings, and investigate the impact of location-based infection control measures on M.tb. incidence and mortality.

This study will directly address two of the WHO's research priorities: determining effects of infection control in 1) healthcare facilities and 2) other settings. Our findings will be of interest to international policy makers, and our results will be useful in informing and improving the effectiveness of infection control interventions. This understanding should contribute to meeting the 2025 WHO End TB Strategy Targets, in the high TB burden settings most in need of novel and innovative prevention strategies.

There is an increasing interest in using location-based infection control interventions to reduce Mycobacterium tuberculosis (M.tb.) transmission, as a means of reducing morbidity and mortality due to tuberculosis disease (TB). The World Health Organization (WHO) has prioritised quantifying the effects of infection control strategies in 1) health care facilities and 2) other settings (congregate settings and households) as two of its key research priorities. Recent trials of ultraviolet germicidal irradiation devices have shown 70-80% reductions in M.tb. transmission to guinea pigs from TB patients in clinics. Improving ventilation can also be highly effective in reducing transmission, and can cost little in warm climates. Current understanding of where M.tb. transmissions occurs in high tuberculosis burden settings is poor however, impeding the effective implementation of location-based infection control measures. For this reason, one of the Stop TB Partnership's high priority questions is "What are the relative contributions of the various foci of TB transmission (e.g. household, community, nosocomial transmission) at the population level?"

This study will improve understanding of transmission by location in Cape Town, South Africa, and other high burden settings. This will enable the potential impact of location-based infection control measures in clinics and other locations to be determined, and will allow the most effective locations to target with infection control measures to be identified. Our findings will therefore be of interest to the National TB Control programmes in South Africa and other high burden countries, as well as the WHO and other global policy makers. The manufacturers of UVGI systems may also benefit, through an increased demand for their products.

In addition to improving understanding of where M.tb. transmission occurs, this study will also lead to an improved understanding of where transmission occurs for other infections that are transmitted directly between humans. In particular, through objective 1, we will investigate the role that variation in infectiousness and susceptibility plays on transmission by location. This may lead to the improvement of transmission models of infections such as influenza, and improved recommendations of the infection control measures (e.g. school closures) that should be employed during epidemics, leading to reduced morbidity and mortality.

More effective TB control measures will also have a wider societal impact. Informed targeting of location-based interventions has the potential to greatly increase both the impact and cost-effectiveness of control measures, ensuring that optimum use is made of the limited resources available for TB care and control. It will also reduce onward transmission of M.tb., reducing the incidence and prevalence of TB disease, contributing towards the WHO goal of TB elimination, and reducing the large economic cost to high incidence countries of reduced productivity due to TB disease.