ANALYSIS, QUANTIFICATION AND SIMULATION OF THE RISK FROM AIRBORNE INFLUENZA

Abstract

Despite the development of effective vaccines, influenza still remains as a global concern. For appropriate public health intervention, it is crucial to accurately determine the routes of transmission. Influenza is believed to have three primary modes of transmission: big droplet, direct contact and aerosol particles. Considerable evidence points to both aerosol and droplet transmission routes as being significant. Because of the limitation of sampling and analysis, the quantitative dynamics of the aerosol mode of transmission are not completely understood. In this dissertation I have characterized the physical and biological collection efficiency of a novel exhaled breath aerosol collector named Gesundheit II (G-II). The device was proven to successfully collect and preserve infectivity with different types of influenza virus. I have also been involved in epidemiological data analysis, experimental quantification and numerical modeling. On experimental quantification, I have been part of a multi-member team that has conducted a study of characterization of respiratory droplets from influenza infected individuals at the University of Maryland campus during the flu seasons of 2012-2013. The exhaled breath was collected with the G-II for accurate quantification of the influenza virus. 218 pairs of fine (< 5 µm) and coarse (≥ 5µm) exhaled breath samples were obtained from 142 subjects and analyzed. The relationship between culturability, coughing frequency, and symptoms were investigated. The high rate of RNA detection and the frequent recovery of influenza virus by culture from fine aerosol samples suggest a contribution of fine particle aerosols in the transmission of influenza. Given these new findings, to understand the risk of influenza infection from these finer droplets, we have modified an existing mathematical risk analysis model and studied the effect of these droplets on subjects in presence or absence of a respiratory protective device (RPD). Two of the major enhancements in our model are (1) the ability to account for subject-to-subject variability over a wide range of age groups and (2) the heterogeneous population was introduced into the model with some infectees or susceptibles not wearing RPDs.