Disease Persistence and Population Dynamics: Modeling Measles under Mass Vaccination

Abstract Following smallpox, several other infectious diseases including measles, polio and rubella have been targeted for eradication. However, elimination of these latter infections has proven challenging. While mass vaccination has halted endemic measles transmission in most of the Americas, with similar high vaccination rates, measles continues to cause large epidemics in some parts of the world. Further, these epidemics can spread to other regions due to high global connectivity and reduced local vaccine coverage (e.g. the recent increases in measles cases in the US). These observations indicate that current understanding of disease persistence in complex population systems remains incomplete and must be improved to effect eradication of infections such as measles. To improve this understanding of disease persistence, the proposed work will develop model-Bayesian inference systems using mathematical modeling and statistical methods to identify the spatial, temporal, and demographical factors contributing to the persistent transmission of measles in the mass vaccination era. A range of hypothesized transmission mechanisms will be tested, including changes in i) vaccination rate, ii) demographics (e.g. birthrates and age structures), iii) contact pattern, iv) spatial connectivity and migration, and/or v) strength of maternal immunity. Further, the project will test potential intervention measures based on the identified transmission mechanisms as well as generate predictions of future measles epidemic dynamics to inform measles elimination efforts. By leveraging detailed measles surveillance data, infectious disease modeling, and Bayesian inference methods, the proposed work will yield new understanding of measles transmission dynamics in modern populations and provide model-guided intervention strategies. Project findings may also inform control strategies for other infections targeted for eradication (e.g. rubella). In addition, the model-inference systems developed here can be adapted to study a broad range of (re)emerging infectious diseases.