Quantifying averted burden from infectious disease control policies

Abstract

This dissertation explores methods for calculating averted burden from infectious disease control policies and applies those methods in the contexts of pediatric HIV in sub-Saharan Africa and malaria in Uganda. In my first aim, I describe the Shapley value estimate as the preferred approach to decomposition and then introduce an innovation to the application of Shapley value estimation in the context of interventions implemented at different times with overlapping effects. I call this innovation sequential Shapley value estimation and detail the algorithm for its application. In my second aim, I apply sequential Shapley value estimation to calibrated estimates of pediatric HIV burden in sub-Saharan Africa. I present intervention coverage level for three biomedical interventions and highlight variable impact made through these interventions in preventing new HIV infections and mortality among children under-15. I also estimate avertable burden in 2023 and discuss the implications for future policy. In my third aim, I propose a framework for evaluating outbreak detection approaches using averted burden and apply it to malaria outbreak detection in Uganda.