A map for all seasons: Tracking transmission dynamics and mortality of childhood infections through the year

Abstract

Among infectious diseases, the dynamics of directly-transmitted, immunizing childhood infections are arguably the best understood. Yet, there are key gaps in our understanding of the complex interactions between disease dynamics, demography, socioeconomic factors, and the environment. This dissertation addresses some of these gaps. In the first chapter, I study six common childhood infections in Mexico, and show that the aggregation of children in schools is a key determinant of the population-level dynamics of these infections. In addition, for some pathogens, I find evidence for the role of socioeconomic and environmental drivers.

My findings from Mexico add to a body of research that has shown that childhood infection dynamics are influenced by age-dependency in contact rates. This first analysis was grounded in the oft-used Time-series Susceptible-Infected-Recovered (TSIR) model, which assumes mass-action or homogeneous mixing, yet is able to accurately predict dynamics of childhood infections through forward simulation in many contexts. This is remarkable, considering that childhood infection dynamics are inherently age-structured. In the second chapter, I show, theoretically and using simulations, that the success of the TSIR lies in the fact that age structure can be collapsed to a simple mass-action mixing model, by allowing the transmission rates to vary over time.

Finally, in the last chapter, I turn to the mortality burden of measles, a major cause of childhood morbidity and mortality in many parts of the world. I use historical mortality data from Matlab, Bangladesh, and a mechanistic model of measles transmission, to quantify changes in the case-fatality rate - an important component of disease burden - in a population that experienced war and famine. My results suggest that the period of turmoil during and after the war, and the sustained effects of the famine, is likely to have contributed to the high fatality burden of the 1976 measles outbreak in Matlab.

Overall, the findings of this dissertation strengthen our ability to model and accurately predict the dynamics of directly-transmitted, immunizing, childhood infections, and to design effective control strategies. This work also opens up a number of intriguing directions for further work, including in disentangling climatic drivers of transmission.