Predominant airborne transmission and insignificant fomite transmission of SARS-CoV-2 in a two-bus COVID-19 outbreak originating from the same pre-symptomatic index case

Abstract

The number of people infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to increase worldwide, but despite extensive research, there remains significant uncertainty about the predominant routes of SARS-CoV-2 transmission. We conducted a mechanistic modeling and calculated the exposure dose and infection risk of each passenger in a two-bus COVID-19 outbreak in Hunan province, China. This outbreak originated from a single pre-symptomatic index case. Some human behavioral data related to exposure including boarding and alighting time of some passengers and seating position and mask wearing of all passengers were obtained from the available closed-circuit television images/clips and/or questionnaire survey. Least-squares fitting was performed to explore the effect of effective viral load on transmission risk, and the most likely quanta generation rate was also estimated. This study reveals the leading role of airborne SARS-CoV-2 transmission and negligible role of fomite transmission in a poorly ventilated indoor environment, highlighting the need for more targeted interventions in such environments. The quanta generation rate of the index case differed by a factor of 1.8 on the two buses and transmission occurred in the afternoon of the same day, indicating a time-varying effective viral load within a short period of five hours.

Keywords: Airborne transmission; COVID-19; Fomite transmission; Quanta generation rate.

Graphical abstract

Fig. 1

Seating arrangements of the two buses involved in the Hunan two-bus COVID-19 outbreak. (a) B1, (b) B2. The seats on which the index case and infected cases sat are colored dark blue and light cyan, respectively; empty seats are colored gray. All of the seats are labeled by a unique number and letter combination, except for the seats of the driver and conductor, which are labeled as D and C, respectively. The seat labels of passengers who wore a face mask are colored dark blue and bolded.

Fig. 2

The trip timelines of the two buses involved in the Hunan two-bus COVID-19 outbreak. The bus measured ventilation rate is affected by its speed, and each speed range is referred to as a phase. There are 8 phases (1–8) with two sub-phases in Phase 1 (i.e. 1a and 1b) and Phase 8 (i.e. 8a and 8b) each. The bus speed in each phase of the two buses is summarized in Table S1. (a) B1. Two 15-min time slots (11:55–12:10 and 15:30–15:45) were allocated before and after the journey, during which passengers boarded or alighted the bus. (b) B2. Two 5-min time slots (15:38–15:43 and 16:43–16:48) were allocated before and after the journey, during which passengers boarded or alighted the bus. Note that two 5-min time slots and three 2-min time slots were considered for passengers to board or alight B1 or B2, respectively, during the journeys of these buses (as the actual boarding and alighting durations during the journeys were unknown). Each phase is represented by a number or a number and letter combination. Each number marks a specific driving speed, bus location and ventilation rate. The letter “a” and “b” means that the index case was not on the bus and on the bus, respectively. T_id,on and T_id,off are the time when the index case boarded and alighted the bus, respectively. For details, please refer to Tables S1 and S2.

Fig. 3

Passenger and driver positions on B1. (a) Passenger positions and behaviors at 15:17:28. (b) Passenger positions and behaviors at 15:28:05. (c) Driver’s behavior at 15:17:28.

Fig. 4

Passenger arrangement on B2. (a) Passenger positions and behaviors at 15:40:36. (b) Passenger positions and behaviors at 15:52:25. (c) Passenger positions and behaviors at 16:00:36. (d) Passenger positions and behaviors at 16:43:33.

Fig. 5

Spatial distribution of the predicted average infection risk [in decimal, not percent] (for 1000 simulations) via fomite or airborne transmission at the end of the computational period. (a) Fomite transmission risk in B1, (b) airborne transmission risk in B1, (c) fomite transmission risk in B2, and (d) airborne transmission risk in B2. The dose–response parameters in the respiratory tract (η_r) = 3.2/RNA copy and that in the mucous membranes (η_m) = 3.2 × 10⁻³/RNA copy, and the viral load (L_o) = 1 × 10⁹ RNA copies/mL. Different levels of infection risk are represented by the intensity of red shading. Seats of the index case and the infected passengers are marked with thick dark blue and light cyan squares, respectively. There are two numbers in each seat square: the number at the top is the seat label, while that at the bottom is the simulated infection risk.