Increasing contributions of airborne route in SARS-CoV-2 omicron variant transmission compared with the ancestral strain.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron variant has become the dominant lineage worldwide. Experimental studies have shown that SARS-CoV-2 Omicron variant is more stable on various environmental surfaces than the ancestral strains of SARS-CoV-2. However, the influences on the role of the contact route in SARS-CoV-2 transmission are still unknown. In this study, we built a Markov chain model to simulate the transmission of the Omicron and ancestral strains of SARS-CoV-2 within a household over a 1-day period from multiple pathways; that is, airborne, droplet, and contact routes. We assumed that there were two adults and one child in the household, and that one of the adults was infected with SARS-CoV-2. We assumed two scenarios. (1) Asymptomatic/presymptomatic infection, and (2) symptomatic infection. During asymptomatic/presymptomatic infection, the contact route contributing the most (37%-45%), followed by the airborne (34%-38%) and droplet routes (21%-28%). During symptomatic infection, the droplet route was the dominant pathway (48%-71%), followed by the contact route (25%-42%), with the airborne route playing a negligible role (<10%). Compared to the ancestral strain, although the contribution of the contact route increased in Omicron variant transmission, the increase was slight, from 25%-41% to 30%-45%. With the growing concern about the increase in the proportion of asymptomatic/presymptomatic infection in Omicron strain transmissions, the airborne route, rather than the fomite route, should be of focus. Our findings suggest the importance of ventilation in the SARS-CoV-2 Omicron variant prevention in building environment.

Introduction

Since the first emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in December 2019, the virus has spread across the globe, causing over 469 million cases and 6.1 million deaths as of March 21, 2022 [1]. SARS-CoV-2 is a respiratory infection believed to be transmitted mainly through direct contact, droplets, fomites, and airborne routes [2]. Exploring the relative importance of different transmission routes is crucial for developing targeted infection-control strategies. However, the extent to which different transmission routes contribute to SARS-CoV-2 transmission between humans remains unclear.

Modelling studies have evaluated the relative importance of evaluating transmission through different routes for SARS-CoV-2 wild strains in healthcare settings [3,4] or households [5]. One study in healthcare setting suggested that droplet and airborne transmission routes may predominate over the contact route [3]. While Mizukoshi and his colleagues suggested fomite route was the major contributor at high virus concentration in the saliva in healthcare setting [4]. Fomites as a mode of SARS-CoV-2 transmission are thought to play an important role at the beginning of the pandemic, with laboratory studies revealing that SARS-CoV-2 can persist on plastic, stainless steel, and other surfaces for hours to days [[6], [7], [8]]. By May 2020, the World Health Organization (WHO) and other agencies were recommending careful and thorough washing of hands and disinfection of frequently touched surfaces. The importance of fomite in SARS-CoV-2 transmission was first confirmed in July 2020 [9,10], with the literature in different settings strengthening this argument. The Centers for Disease Control and Prevention (CDC) had stated that “people can be infected with SARS-CoV-2 through contact with surfaces. However, based on available epidemiological data and studies of environmental transmission factors, surface transmission is not the main route by which SARS-CoV-2 spreads, and the risk is considered low [11].”

The newly emerged Omicron SARS-CoV-2 variant was first identified on November 19, 2021, in South Africa [12], and soon became the dominant strain worldwide, suggesting its high transmissibility in humans. A recent structural study indicated that its spike protein is more stable than that of the ancestral strain [13], and an experimental study showed that the SARS-CoV-2 Omicron variant was more stable on various environmental surfaces compared with ancestral strain [14]. In addition, Omicron variant has a much higher rate of asymptomatic infection than ancestral strain [15]. And infection with Omicron was less severe than with ancestral strain [16]. Omicron variant has difficulty to mitigate its transmission for its high transmissibility under the strict public health and social measures in other countries [17]. Thus, the transmission routes of Omicron variant may be different to these of ancestral strain. The transmission of SARS-CoV-2 Omicron via packaging has been reported over the past few months. However, how the changes in surface stability would influence the role of fomite route in SARS-CoV-2 Omicron variant transmission remains unknown.

The relative importance of different transmission routes varies in different scenarios [18]. In this study, we considered a household environment because households are high-risk settings for the transmission of COVID-19 [19] and are an important factor in wider community spread [20,21]. It has been reported that most clustered COVID-19 infections in the first wave in China were within families [[22], [23]], suggesting high rates of intra-family transmission and urging prioritization of studies on risk factors for household transmission.

Methods

Environmental setting

The mean household size is 2.95 members in China [24], as compared to 2.52 in the USA [25]. Thus, the household size was assumed to be three, that is, two parents and one child around 10 years of age, since the contacts between little baby and parents are unique and different to these between parents and children with age>6. So here we considered one child around 10 years of age to represent a more general child-to-parent contact patterns. Since adults are more susceptible to SARS-CoV-2 and there is no gender difference in susceptibility to COVID-19 [2], one of the parents was assumed to be infected with SARS-CoV-2 and the other two individuals were assumed to be susceptible to SARS-CoV-2. In this study, we modeled the risk of SARS-CoV-2 infection in two susceptible individuals during a 1-day exposure to the household environment.

Transmission route definition

As a respiratory infection, SARS-CoV-2 can be transmitted by contact, droplets, fomites, airborne routes, and possible fecal-oral, bloodborne, and intrauterine routes [2]. In this study, we considered four major transmission routes: contact, droplet, fomite, and airborne routes, and classified them into the following three categories:

The airborne route refers to the direct inhalation of an infectious agent through small droplet nuclei, where the residue of large droplets containing microorganisms that have evaporated to an aerodynamic diameter of less than 10 μm (termed respirable) [26]. These respirable droplets can then be deposited in the respiratory tract.

The droplet route refers to the inhalation of the virus carried in respirable airborne particles with a diameter between 10 and 100 μm (termed inspirable) [26], and the droplet spray of large droplets (>100 μm in diameter) onto facial target membranes.

The contact route includes both direct and indirect contact routes. Direct contact route refers to infection transmission through person-to-person contact such as shaking hands. The indirect contact route is also called the fomite route, which refers to infection transmission by touching objects or surfaces that have previously been contaminated by hands or by the direct deposition of infectious pathogens from infected individuals. Direct and indirect contact routes are difficult to distinguish because hands can be contaminated via both routes, and the virus concentration on the hands of susceptible individuals sequentially influences the exposure level. Thus, in this study, the two routes have merged to one route. And we evaluated the relative importance of direct and indirect contact routes via the relative infection risk when the susceptible individuals did not touch fomites or hands of the infector.

The exposure pathway model

A Markov chain was used to model the movement of the virus between select compartments in the household environment (Fig. 1 ), which described virus moved between the states according to rate constants in the direction of the arrows. Based on this model, the viral load after n time steps can be calculated. In addition, we also described this model in detail in the Method. We used this model to study multi-route infection transmission in airplanes, households, and hospitals [5,27,28]. Eight compartments and three transmission routes were considered in the study. The virus is emitted into the room air (compartment 1) in the droplets exhaled by the infector. In this study, the environmental surfaces were divided into two types: porous surfaces, which is typically rough, such as cloth, the bed cover surfaces; and non-porous surfaces, such as the glass table surfaces. Some of the virus in the room air could be deposited on porous (compartment 2) and non-porous surfaces (compartment 3), infector hands (compartment 4), exhausted by ventilation (compartment 5), and inhaled by susceptible individuals (compartment 6). The virus can be transferred between porous/non-porous surfaces and the infector's or susceptible individuals' hands (compartment 7) during hand-to-surface contact. The virus on the hands of susceptible individuals could be transmitted to the mucus membranes and lead to infection transmission (compartment 8).

Fig. 1

The exposure pathway model in a household environment.

Airborne route model

For airborne route, we assumed that the airborne droplets were uniformly distributed in room air. On one hand, the infector could exhale virus-contained airborne droplets into room air, on the other hand, the virus-contained airborne droplets could be removed via ventilation, natural inactivation and deposition on horizontal environmental surface. We elected not to consider virus inhalation by individuals in the household and deposition on the vertical environmental surface during the 1-day simulation period because of their negligible contribution. Then the concentration of droplet nuclei in room air () can be calculated by equation (1).where is the concentration of virus in the droplets with initial radius ; is the ventilation rate of the room, is the inactivation rate of SARS-CoV-2 in air, is the room volume, and quantifies the deposition rate of droplets with particle size on the horizontal environmental surface. is the generation rate of droplets of radius exhaled by the infector. The total number of droplets generated by the infector per hour corresponds to ; droplet size distribution is , and . At a steady state, (), there is

Then the airborne exposure dose () of a susceptible person during exposure time can be calculated by equation (3):where is the largest radius of respirable droplets, is the pulmonary ventilation rate. is the initial radius of the droplet immediately after exhalation. We assume that the final radius after complete evaporation is [27].

Droplet route model

There are two main droplet transmission routes: (1) inhalation of the virus carried in inspirable droplets with diameters between 5 and 50 [26,27]. These inspirable droplets are mainly deposited in the upper respiratory tract; and (2) droplet spray of virus-containing droplets onto the facial membranes of susceptible individuals during talking and coughing. It was assumed that the trajectory of the droplets exhaled by the infector formed a jet, and the cone angle formed by the jet is , the mucosal area of a person is , at a horizontal distance of S (S ≤ 2 m) from the infector, the virus concentration in the droplet is:

We assume that the exposure time of susceptible individuals in the droplet transmission route is . The exposure dose caused by inhalation is:where is the maximum radius of the inspirable droplets. Small droplets with particle sizes less than 10 μm evaporated completely before being inhaled by the susceptible person, however, large droplets may not evaporate completely when inhaled by a susceptible person; thus, is determined by the evaporation time and inhalation time of susceptible persons. Assuming that is the traveling time for the exhaled droplets from the source to reach a susceptible person a distance S away, is the evaporation time for the droplets with radius .

Xie et al. [29] studied the evaporation time of droplets with different diameters using a fitting function,  =  was used in this study. , where is the droplet speed after exhalation. The exposure dose by spraying onto the mucous membrane isWhere is the maximum radius of the exhaled droplets. is the area of mucous membranes. The average interpersonal distance between two individuals (S) during talking was set as 0.81 m [30].

Contact route model

By assuming the uniformly distribution of SARS-CoV-2 on environmental surfaces and hands after each touch, the time varying SARS-CoV-2 concentrations on each environmental surface and hand can be calculated via ordinary differential equations, which had been used in previous literatures [[31], [32], [33]]. In this study, the environmental surfaces were divided into two types, i.e., porous and non-porous surfaces. Virus on hands could be removed by hand hygiene and natural inactivation. And we assumed there was no surface cleaning during the 1-day simulation period.

Assuming that 1% of the virus exhaled by the infector deposits by droplet spread on the infector's hands, the rate of virus deposition on the infector's hand is

Viruses on the hands also lose activation and are removed by hand hygiene. The natural inactivation rate of the virus on hands is represented by . Hand hygiene frequency was assumed to be and hand hygiene efficiency was assumed to be . Hand hygiene () is considered to remove 90% of the virus [34]. To represent the discrete hand hygiene process in the continuous governing ordinary differential equations, we made the following translation. After hand hygiene, only a fraction (1–) of the viruses remained on the hands. Hand hygiene occurs times per hour, and the time-averaged rate of pathogen removal due to hand hygiene is denoted by . On average, if there is 1– = , then  = −log (1–90%) [33].

At the steady state, the amount of virus on the patient's hands is

Assuming the infector and the susceptible adult touch the environmental surface with rate , and the susceptible child touch the environmental surface with rate , the frequency of hand-to-hand contact between the infector and the susceptible adult is , and that between the infector (and the susceptible adult) and the susceptible child is . Then, the amount of virus on the th susceptible individual's hand, ( = 1 for the adults,  = 2 for the child), and the kth environmental surface, ( = 1 represents the porous surface,  = 2 represents the non-porous surface), can be calculated by equations ((10), (11), (12)): where and represent the transmission efficiency of the virus from the hand to the kth environmental surface, from the kth surface to hand, and from hand to hand, respectively., , , and represent the area of the th susceptible individual's hand, area of the th environmental surface, contact area during hand-to-environmental surface contact, and contact area during hand-to-hand contact, respectively. is the inactivation rate of the virus on the kth environmental surface. We assumed that hand washing and surface cleaning are performed at a certain frequency at certain time points.

Then, the total exposure dose of th susceptible individual via the contact route at the exposure time can be calculated by equation (13):where is the contact rate on the mucous membrane of th susceptible individual, is the transmission efficiency of the virus from the hand to mucous membranes, and is the contact area between the hands and mucous membranes of the th susceptible individual.

Detailed model parameterization were summarized in Supplementary Information Part 1 Model Parameterization.

Infection risk assessment

The negative exponential dose-response model [27] was used to estimate the infection risk, which implies that a single particle can start an infection and that all single particles are independent of each other. The infection risk of individual during 1-day exposure can be calculated according to the following equation:where , , and are the dose-response rates in the lower respiratory tract, upper respiratory tract, and mucous membranes, respectively. To the best of our knowledge, a dose-response relationship for both SARS-CoV-2 ancestral and Omicron variant as a cause of SARS-CoV-2 has not been reported. Therefore, in this study, we assumed that the dose-response rates for the ancestral and Omicron variants were the same. Animal experiments have suggested that airborne transmission is more efficient than fomite transmission [35]. Recent findings on SARS-CoV-2 also revealed that angiotensin-converting enzyme 2, which is used to infect humans, is expressed in the nose at higher levels than in the lower respiratory tract with a ratio of difference approximately 4:1 [36]. Therefore, this is a case of . In this study, we assume that [4], Although the 100:1 ratio for infectivity in the lower respiratory tract and upper respiratory tract/mucous membranes may be an overestimate, we set the ratio mainly based on the data on influenza virus in humans; for influenza, there is [26,37]. For adults, , which was the dose-response rate for SARS-CoV-2, and was also the best available model in the present study [4]. For children, the dose response rates were assumed to be half of those for adults, since studies have reported that susceptibility to SARS-CoV-2 increases with age [38,39]. The detailed model parameterization is summarized in the Supplementary Information Part 1 Model Parameterization.

Virus shedding by the infector

For both Omicron and ancestral variant, SARS-CoV-2 infected individuals shed viruses into the environment via respiratory activities such as breathing, talking, coughing, and sneezing. The proportion of asymptomatic/presymptomatic infections is high in SARS-CoV-2 transmission [40]. We defined ‘asymptomatic’ as an individual with laboratory-confirmed SARS-CoV-2 infection but without symptoms throughout their entire course of infection, or after 14 days of follow up; and ‘pre-symptomatic’ as an individual who reports no symptoms at the time of the initial positive test result, but who subsequently develops symptoms attributable to COVID-19, i.e., during the incubation period [41]. In this study, we considered the following two scenarios: 1) asymptomatic/presymptomatic infection, where infected individuals shed viruses into the environment via breathing and talking. 2) Symptomatic infection, where infected individuals shed viruses into the environment via breathing, talking, and coughing. Detailed size distribution of droplets from breathing, talking and coughing were in Supplementary Information Part 1.

During asymptomatic/presymptomatic infection, we assumed that 49.5% and 49.5% of the droplets exhaled by the patient were partitioned into porous and non-porous surfaces, respectively, and the remaining 1% would be deposited on the infector's hands. During symptomatic infection, we assumed that 89% and 10% of the droplets exhaled by the patient were partitioned into porous and non-porous surfaces, respectively [4], as the infected individuals would lie in bed most of the time. The remaining 1% was deposited in the hands of the infector. The viral load in nasopharyngeal swabs peaked 1 d before symptom onset, with values of 8–9 log 10 RNA copies per mL, and then decreased exponentially [42,43]. Thus, in this study, in asymptomatic/presymptomatic infections, the viral concentration in the droplets was set from 105 to 109 RNA copies per mL, and in symptomatic infections, the viral concentration was set from 104 to 108 RNA copies per mL.

Sensitivity analysis

Some parameters had uncertainties that could not be controlled by the scenario settings. Thus, we performed the sensitivity analysis of these uncertain parameters, including hand-to-surface contact rate, hand hygiene frequency, dose response rates and proportion of virus deposition on different surfaces, which were summarized in Supplementary Information Part 2 Sensitivity Analysis.

Results

Estimation of the infection risk

The risks of each pathway and the overall risk depended on the virus concentration in saliva (Fig. 2 ). Under the same conditions, the overall infection risk was albeit slightly higher for the Omicron strain than for the ancestral strains (Fig. 2). This was mainly because of the assumption that Omicron variant had higher dose response rates than the ancestral strains due to its higher infectivity. With higher dose-response rates, the infection risk of the Omicron variant would be higher (Supplementary Information Part 3 Fig. S5). When the salivary virus concentration was 104–108 RNA copies/mL, the overall infection risk for the susceptible child was approximately half of that of the susceptible adult, as the dose response rate of the child was set as half of that of the adult, but when the salivary virus concentration was 109 RNA copies/mL, the overall infection risks were close to 1 for both the susceptible child and the adult. During asymptomatic/presymptomatic infection, for the susceptible child, the overall infection risk was 2.210−4 for salivary virus concentration of 105 RNA copies/mL, 2.210−2 for salivary virus concentration of 107 RNA copies/mL, and 0.89 for salivary virus concentration of 109 RNA copies/mL. For the susceptible adult, the overall infection risk was 4.410−4 for salivary virus concentration of 105 RNA copies/mL, 4.310−2 for salivary virus concentration of 107 RNA copies/mL, and 0.99 for salivary virus concentration of 109 RNA copies/mL. During symptomatic infection, for the susceptible child, the overall infection risk was 2.310−4 for salivary virus concentration of 104 RNA copies/mL, 2.310−2 for salivary virus concentration of 106 RNA copies/mL, and 0.90 for salivary virus concentration of 108 RNA copies/mL. For the susceptible adult, the overall infection risk was 4.710−4 for salivary virus concentration of 104 RNA copies/mL, 4.610−2 for salivary virus concentration of 106 RNA copies/mL, and 0.99 for salivary virus concentration of 108 RNA copies/mL. When the infector was symptomatic, the infector was expected to produce a much larger number of droplets than during asymptomatic/presymptomatic infection, as expected by the addition of cough to the modes of infector droplet generation. However, the peaked virus concentration in the saliva was supposed to be lower. Thus, the peaked overall infection risk of two susceptible individuals with asymptomatic/presymptomatic infection was close to that of symptomatic infection, which suggests that asymptomatic/presymptomatic infections could contribute to approximately 50% of SARS-CoV-2 transmission.

Fig. 2

Absolute infection risk from each pathway and overall risk. (A, B, C, D) during asymptomatic/presymptomatic infection, with salivary virus concentration at 105–109 RNA copies/mL respectively; (E, F, G, H) during symptomatic infection, with salivary virus concentration at 104–108 RNA copies/mL respectively; (A, C, E, G) for ancestral strain; (B, D, F, H) for Omicron variant; (A, B, E, F) for the susceptible adult; (C, D, G, H) for the susceptible child.

The contribution of each pathway according to different virus concentrations in the saliva was shown in Fig. 3 . For both strains, the dominant routes during asymptomatic/presymptomatic infection were different from those during symptomatic infection. When the infector was asymptomatic/presymptomatic, all three routes contributed a role in SARS-CoV-2 transmission (Fig. 3A and B). Contact route contributed the most (37%–45%), followed by the airborne route (34%–38%) and droplet route (21%–28%). When the infector was symptomatic, the droplet route was the dominant pathway (48%–71%), followed by the contact route (25%–42%). The airborne route played a negligible role (<10%). Compared to the ancestral strain, although the contributions of the contact route increased in Omicron variant transmission, the increase was slight (Fig. 3). Further, the sensitivity analysis showed that the slight increase in the contribution of contact route in SARS-CoV-2 Omicron variant transmission compared with the ancestral strain was robust to the key parameters in the model (Supplementary Information Part 2 Sensitivity Analysis).

Fig. 3

Contribution of airborne, droplet, and contact transmission routes to the (A, C) susceptible child and (B, D) the susceptible adult, (A, B) during symptomatic infection and (C, D) during asymptomatic/presymptomatic infection, for different salivary virus concentrations. Left: ancestral strain; Right: Omicron strain.

We also explored the relative contributions of direct and indirect contact routes to the contact transmission of SARS-CoV-2 (Fig. 4 ). The indirect contact route dominated the contact transmission of SARS-CoV-2 in households. This may be mainly because the frequency of hand-to-environmental surface contact (3 times per hour) was much higher than the frequency of hand-to-hand contact (0.1–0.5 times per hour), and the hand hygiene frequency was also high, at 0.5/h; however, environmental surfaces were not considered to be clean, so the virus concentration on environmental surfaces was higher than that on the hands (Supplementary Information Part 3, Fig. S5). Because of the relatively higher hand-to-hand contact between the susceptible child and the infector, the direct contact route contributed more in contact transmission of SARS-CoV-2 for the susceptible child than that for the susceptible adult (Fig. 4).

Fig. 4

Contribution of direct and indirect (fomite) contact route in contact transmission of SARS-CoV-2. (A, C) Susceptible child and (B, D) the susceptible adult, (A, B) during symptomatic infection and (C, D) during asymptomatic/presymptomatic infection, for different salivary virus concentrations. Left: ancestral strain; Right: Omicron strain.

Discussion

The emergence of the new coronavirus variant strain, Omicron, has led to a rapid increase in the proportion of asymptomatic infections, making it more difficult to prevent infection and control the pandemic. To control the transmission of SARS-CoV-2, the relative importance of the different pathways should be quantified, to determine the effective non-pharmaceutical interventions. In this study, based on our current knowledge about SARS-CoV-2 and its related information, we calculated SARS-CoV-2 infection risk through multiple pathways of exposure to SARS-CoV-2 in the context of a prototypical household and estimated the relative contribution of each pathway to transmission. We found that the relative importance of different routes varied between asymptomatic/presymptomatic and symptomatic infections. During asymptomatic/presymptomatic infection, all three routes contributed a role. During symptomatic infection, the droplet route was the dominant pathway (48%–71%) and the airborne route played a negligible role (contribution<10%). The main reason for this is that, during asymptomatic/presymptomatic infection, SARS-CoV-2 is mainly emitted via breathing and talking, and most droplets emitted from breathing are respirable droplets with diameters less than 10 μm. When the infector developed symptoms, most viruses were emitted from coughing, since the volume of saliva from coughing was much larger than that from breathing. Droplets from coughing had a relatively larger size, so most exhaled droplets were deposited on environmental surfaces and inhaled by susceptible individuals via the droplet route. With the emerging concern about the increase in the proportion of asymptomatic/presymptomatic infections due to the Omicron strain [44], evaluating the relative importance of different transmission routes in asymptomatic/presymptomatic infection is critical for SARS-CoV-2 prevention. In addition, in both asymptomatic/presymptomatic and symptomatic infections, although the SARS-CoV-2 Omicron variant was more stable on various environmental surfaces than the ancestral strain, the contribution of the contact route to SARS-CoV-2 transmission increased slightly for the Omicron variant.

Previous studies have explored the relative contributions of different routes to SARS-CoV-2 ancestral variant transmission via both animal models [45] and mathematical models [3,4]. In the animal experiment study, the absolute risks of airborne, droplet, and overall risk were 0, 3/10, and 7/13, respectively; thus, considering the inclusion and exclusion of contact transmission, the estimated contributions of airborne, droplet, and contact routes could be 0%, 57%, and 43%, respectively. This tendency is consistent with the results obtained when the infector was symptomatic. The modelling studies conducted by Jones [3] and Mizukoshi et al. [4] both considered virus emission via coughing; the study by Jones [3] considered virus emission solely via the coughing as a possible action for spread. Jones found that the droplet and airborne routes predominated, contributing 35% and 57%, respectively, in hospitals. Mizukoshi et al. [4] found that the airborne route was much less important than the contact route and droplet route, only contributing 4%–10% of SARS-CoV-2 transmission in hospitals. This is consistent with the results obtained when the infector was symptomatic. In addition, Mizukoshi et al. [4] found that the role of the droplet route decreased with an increase in salivary virus concentration, contributing 65%–70% when the virus concentration was 101–104/mL, and 20%–46% when the virus concentration was 105–108/mL. When the virus concentration was low, the estimated contribution of the contact route was consistent with the results of this study (67%–71%). However, when the virus concentration was high, the estimated 20%–46% contributions were much lower than those in this study (48%–71%). The main reason for this could be that the estimated overall risk in the study by Mizukoshi et al. [4] was much higher than our estimation under the same salivary virus concentration. When the infection risk by contact route was high, with further rapid increase in salivary virus concentration, the infection risk by contact route would increase slowly, while infection risk via other routes would still increase quickly; thus, the contribution of the contact route was relatively decreased. The relatively low overall infection risk in this study was mainly due to the relatively small volume of droplets considered to be exhaled by talking (0.16 VS 3.2 mL saliva per 100 s speaking) and coughing (4.4 VS 8.1 mL per cough).

SARS-CoV-2 transmission through fomites has been demonstrated in animal model studies [35,45,46]. SARS-CoV-2 has been detected on various surfaces [[47], [48]]. In long-term care facilities with COVID-19 patients, SARS-CoV-2 concentrations on environmental surfaces ranged from 1.3 to 3661.2 genomes/cm2, with median values at 76.6 genomes/cm2 [49]. The estimated virus concentration on the environmental surfaces in this study was similar to the results of field measurements. The transmission of microns of SARS-CoV-2 Omicron via packaging has attracted widespread attention. Because the virus inactivation of the Omicron variant is lower than that of the ancestral strain [14], concerns are rising about a significant increase in contact transmission risk. Since both Omicron and ancestral variants survive well on environmental surfaces, hand hygiene plays a key role in restricting contact transmission of SARS-CoV-2. In our study, we found that the role of the contact route increased slightly in Omicron variant transmission, by no more than 5%. Therefore, contact transmission is should not be the cause of excessive concern. In contrast, we found that, with the increase in asymptomatic/presymptomatic infections in Omicron strain transmission [44] compared with the ancestral strain, the airborne route should gain more attention in SARS-CoV-2 transmission and prevention. When the infector was symptomatic, the airborne route played a negligible role (<10%), but this pathway becomes important (34%–38%) when the infector is asymptomatic/presymptomatic.

This study had several limitations. First, the results depend on the model assumptions. For example, the relative contribution of the droplet route is dependent on the emission of the virus in respirable droplets and the exposure time of susceptible individuals. Hand hygiene frequency and efficiency affected the results related to contact route. However, the real-life effect and variables relating to these parameters remain unclear. To improve the accuracy of the model, the data pertaining to SARS-CoV-2 should be updated in future studies. Second, it should be noted that the scenario we assume at home is relatively simple; we only consider close contact distance (0.81 m between two individuals, and contacts of infected and susceptible individuals were only considered for 30 min of conversation). In reality, there are more complex activities, such as dining together, that may increase the risk of the droplet and contact routes. In other public places such as shopping malls and restaurants, the dominant route may be different; thus, we would study further scenarios of the other settings. Lastly, the behavioral settings of susceptible adults and susceptible children are the same; however, children may touch more surfaces and wash their hands less often, which may lead to a higher risk of contact transmission. In a word, the overall risk of SARS-CoV-2 infection predicted by the mathematical model should be interpreted with caution, though the result comparisons are informative.

Conclusions

In SARS-CoV-2 Omicron strain transmission, concern about the increase in the proportion of asymptomatic/presymptomatic infections and the airborne route, rather than the fomite route, should receive much attention; interventions such as increased ventilation should be recommended for Omicron strain transmission prevention.

Funding

This study was supported by grants from the (Grant No. 82003509 to H.L.), the (Grant No. LQ20H260009 to H.L.), the (Grant No. 2020A1515110455 to S.X.), Fundamental Research Funds for the (to H.L.), and the Shenzhen Science and Technology Program under Grant (Grant No. RCBS20210706092345029 to S.X.).

CRediT authorship contribution statement

Shuyi Ji: Writing – original draft, Software, Formal analysis. Shenglan Xiao: Writing – review & editing, Software, Formal analysis, Data curation. Huaibin Wang: Writing – review & editing, Data curation. Hao Lei: Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.