Environmental Health, Occupational Health, Environmental Justice, and Climate Change
Estimating exposure to infectious influenza aerosols in roommates of influenza cases during the 2012-2013 season on the University of Maryland campus
(School of Public Health (UMD) Maryland Institute for Applied Environmental Health Doctoral Student)
Bueno de Mesquita, Jacob (UMD SPH Maryland Institute for Applied Environmental Health), Heidarinejad, Mohammad (Illinois Institute of Technology), Addo, Kofi (UMD A. James Clark School of Engineering), Dalgo, Daniel (UMD A. James Clark School of Engineering), Mattise, Nicholas (UMD A. James Clark School of Engineering), Srebric, Jelena (UMD A. James Clark School of Engineering)

Background: The purpose of this study is to estimate exposure to infectious, airborne influenza virus in a dormitory room given an infected roommate. To understand a typical quantity of infectious virus in the exhaled breath of campus cases, we take advantage of observed viral shedding from 142 influenza cases at the University of Maryland, described in Yan et al., 2018. We couple this data with newly collected, dormitory room CO2, a marker of exhaled breath exposure, to estimate the dose of inhaled virus following the Rudnick-Milton equation (2003).


Objectives: Given the challenges with effective vaccine match, strategies designed to reduce influenza virus exposure are paramount to protecting population health. Effective exposure reduction draws upon knowledge of transmission via direct contact, large droplet spray, and fine particle aerosols. Contact and large droplet transmission can be mitigated by reducing close contact with infected individuals, washing hands, and sterilizing surfaces, as recommended by CDC. Protecting against airborne exposure poses a greater challenge, however, a household transmission study estimated that about half of influenza is spread by airborne, fine particle aerosols (<5um in aerodynamic diameter), and infection initiated this way may lead to enhanced respiratory symptoms (Cowling et al., 2013). Given the potential for fine particle aerosols to carry infectious virus, and to mix suspended in air, we employed continuous CO2 monitoring of dormitory rooms to estimate the rebreathed air fraction as a proxy for exposure to airborne virus. Transmission risk is inferred given the assumption that a single virus is infectious and follows a Poisson distribution.

Approach: Influenza virus shed into fine particle aerosols was quantified, using a G-II bioaerosol sampler from community acquired influenza cases season (Yan et al., 2018). 30% of cases had culturable virus by focus assay and the geometric mean of positive cases was 37 with geometric standard deviation of 4.4. Per Rudnick and Milton, a rebreathed air fraction is calculated as the difference between indoor and outdoor CO2 concentrations (PPM) divided by the volume fraction of CO2 added to exhaled breath during breathing, assumed to be 0.038. Then average virus concentration in a room was calculated for a triple with 1 case and 2 susceptibles, for an exposure period and later the average number of viruses breathed by a susceptible person, using equation 1, µ ̅=ptN ̅ (1) where p is the pulmonary ventilation rate (l/hr) assumed to be 480, t is the exposure time, assumed to be 8 hours, and N ̅(per l) is the average virus concentration in the room for total exposure time t. There is a poisson distributed probability that a susceptible person remained uninfected, and thus the probability of infection.

Results: The findings show that the rebreathed air fraction for each of the two representative dorm rooms is dramatically different, and the building with improved ventilation (dorm 2) somewhat reduces transmission risk, given the low dose for influenza. From an epidemic prevention standpoint, the critical rebreathed air threshold is 0.0035, which is an order of magnitude smaller than what was observed in dormitory 2. The average CO2 level at night (8am-8pm) for dormitories 1 and 2 were 1.57E+04 and 1.70E+03, respectively. Rebreathed air fractions, based on these CO2 levels were 0.40 and 0.033. Geometric mean for number of inhaled infectious viruses over an eight hour period were 79 and 7 respectively. Assuming infection can be caused by a single infectious particle, as suggested by Alford et al. (1966), the risk of airborne transmission for the 2 susceptibles in both dorms would be 100%. If the number of viruses inhaled by susceptibles were at 1 standard deviations below the geometric means, then that would result in infection risk of 100% and 50% for dormitories 1 and 2, respectively. If the number of viruses inhaled by susceptibles were at 1 standard deviations below the geometric means, then that would result in infection risk of 88% and 16% for dormitories 1 and 2, respectively. These infection risk outcomes warrant validation through additional analyses to account for various exposure conditions and infectious dose estimations.

Importance to public health: Results from this analysis show that there is significant risk of airborne transmission between roommates, given a single shedder. Overall this work suggests that there is room to further reduce aerosolized virus exposure through improved ventilation and/or filtration, and/or inactivation methods such as upper-room germicidal UV.