Aggravated exposure risks of children to multipath transmitted pathogens in indoor environments Authors: Zhijian Liu, Lina Hu, Chenxing Hu, Qian Liu, Zhenyu Liang, Rui Rong, Haiyang Liu Published: 2023 ABSTRACT SummaryConsidering the significance of multipath transmissions of respiratory pathogens in the post-epidemic era, surprisingly little is acknowledged regarding the susceptibility, short-term aerodynamics, and exposure risk of children in indoor environments. Here, experimental and computational investigations were conducted to evaluate the exposure risks associated with respiratory pathogens. The dominant effect of recirculation structure originating from indoor ventilation, including air supply modes and air change rate, on aerosol dispersion was quantitatively proved. A large proportion of deposited aerosol particles was captured by the human body, and deposited particles may further increase under high air change rate, which required a balanced ventilation strategy. Little discrepancies existed between adults and children in exposure risk by airborne transmission. The infection probability by contact transmission for children, however, may be surprisingly high due to high frequency in interactive activities, deposition on upper and lower limbs of accompanying parents, and the wall within arm span. Introduction In the past decade, the prevalence of respiratory diseases such as influenza, respiratory syncytial virus (RSV), acute severe respiratory syndrome (SARS), or novel coronavirus type 2 (SARS-CoV2) has brought about substantial human losses.1,2,3 Evidences show that infants and young children are susceptible to many respiratory pathogens due to their still-developing immune systems, narrow airway, and low attention to hygiene.4,5,6 Notably, a so-called tripledemic of RSV, seasonal influenza, and SARS-CoV2 surged in the population at once in the US over the winter of 2022, resulting in a remarkable increase of hospitalization numbers for children (roughly 4,000 cases per week) according to the Centers for Disease Control and Prevention.7,8 In the era of pre- and post-pandemic, statistical models estimating the seasonal variation of the clinical and syndromic surveillance for hospitalized adults and children due to respiratory diseases are not rare to find in previous reports.9,10 A growing body of evidence has shown that the young children (below 5 years old) are more vulnerable to RSV and influenzas virus, but catch COVID-19 less often and develop milder symptoms than adults.5,11 Although immune preparedness and natural antibodies of children are addressed by researches to unravel this perplexing mystery,4,12 more questions remain to be answered, and focus of the research community has gradually shifted to the young children. A key question under debate is how the susceptibility to respiratory virus differs by age in the transmission chain, especially in densely populated, confined settings with high probability of intensive contact. While considerable attention has been devoted to the individual effects of pathogen activities, mask usage, and environmental/building factors on respiratory infections among adults,13,14,15,16,17 the relative contribution of each transmission mode to children infection under indoor environment remains a mystery. Accurate estimating the transmission risk for children in close contact is not a straightforward task since a child’s susceptibility to multipath transmitted bioaerosols may vary substantially along with the activities of infected and susceptible hosts, such as coughing, sneezing, and touching the nose or eyes.18 Uncertainty surrounding the age-related differences in transmission risk of respiratory pathogens in ventilated indoor environment hinders the interpretability of bioaerosol models quantifying the threats to children’s health. This knowledge gap should be bridged promptly in preparation for the spreading of new variant of SARS-CoV2 or other respiratory diseases. Precisely how the viral infections transfer between infected and susceptible hosts has long been a matter of debate. Direct contact with contaminated surfaces or body parts and airborne transmission of exhaled droplets (>5 μm in diameter) or droplet nuclei (<5 μm in diameter), however, are supported with abundant evidences according to World Health Organization guidelines.19,20 Airborne transmission of virus-laden aerosol is strongly backed up with several famous super-spread events in early 2020, including a poorly ventilated restaurant outbreak in Guangzhou, China,21 an outbreak among members of a Washington choir,22 and a rapid outbreak in a nursing facility with high mortality in Boston.23 As to contact transmission, it was also proved with animal tests by putting the animals in the cage that was previously used to house the infected animals.24,25 Children are supposed to have higher respiratory rate but lower inspiratory capacity compared with adults, which means naturally lower inhalation of droplets laden with the infectious virus once released from occupants.26 In an indoor microclimate such as classrooms or households, the aerodynamic and microbiological activities of airborne particles carrying respiratory pathogens may vary both temporally and spatially, typically influenced by the environmental/building factors such as temperature, relative humidity, mechanical ventilation, and filtration.14,15 In addition, the deposited containments on solid or body surfaces also contribute to the infection spread as indirect route and are highly associated with human activities.27,28 For children in household or confined settings, high-touch surfaces29 including the surfaces of tables, chairs, door handles, toys, walls below arm’s length and accompanying parents’ body are especially critical. Cumulative evidences reveal that the deposition pattern of virus-laden aerosols change significantly with heating, cooling, and air conditioning (HVAC) or recirculating portable air filtration (PAF),30,31,32 since the deposition flux of particle depends on Brownian and eddy diffusion as well as gravitational settling.33 Modeling the infection risk of respiratory disease from hospitalization statistics,34 clinical trials,35 or computational fluid dynamics15,36 is arguably the most critical step to derive epidemiological insights into infection spread. Among the attempts to quantitatively assess infection risk of respiratory disease through airborne transmission, the Wells-Riley-based and dose-response-based models are the primary choices in indoor scenarios. Following the classic concept originated from Wells-Riley model,37 the quanta values represented the minimum number of these three respiratory pathogens required for infection, which were determined based on the estimation analysis performed previously. The detailed ranges of quanta values for each pathogen were shown in Supplementary Table, and median values were selected in the present analysis approach, which was 127 for SARS-CoV-2 Alpha variant,38 1535 for SARS-CoV-2 Omicron variant,38 and 258 for influenza virus.39 Recent reports show that spatial variations of infectious aerosols driven by ventilation and fluid dynamics are the main concern in the improved Wells-Riley models.40Conversely, there exist two greatest obstacles to estimate the risk of infection induced by aerosol deposition, which are the significant influence of aerodynamic size of deposited particles, and the randomness in the activities of susceptible individuals, particularly prominent for children.40,41 Up to date, surface sampling tests and computational-fluid-dynamics-based techniques remain to be the dominating approaches in the quantification of deposition behavior of respiratory pathogens.42 Clearly, a full-scale, well-characterized, quantitative modeling of the aerodynamics and infection risk by airborne and close-contact virus-laden aerosols in indoor environment is highly desired, and holds promise toward a better understanding of the transmission chain for children’s respiratory diseases. The innovation of this study is two-folded: (1) systematically differentiate the infection probabilities between adults and children considering various ventilation conditions and respiratory pathogens; and (2) quantitatively evaluate the significance of two transmission modes based on activities involving inhalation and contact on the susceptivity in a confined setting. To this end, computational fluid dynamics analysis and aerosol tracking tests are conducted in a real-world infusion center for children in China. In this facility, pathogen-containing aerosols are released into the indoor environment through the respiratory behaviors of infected occupants. A structured scheme is devised to combine aerosol dispersion and deposition for estimating the exposure risk of both adults and children to the SARS-CoV-2 Alpha variant, SARS-CoV-2 Omicron variant, and influenza virus (Figure 1). Concentration of aerosols particles around the nose of the person (within the radius of 0.2 m for adults and 0.162 m for children) is incorporated into an improved Wells-Riley model to quantify the airborne transmission contributions driven by ventilation. The contribution of deposited contaminant is estimated by statistical analysis of the particles captured on walls below arm’s length of adults and children, seat surfaces, floor and individual bodies.Figure 1Overview of multipath-transmitted aerosol behaviors and exposure risk assessment(A) Typical transmission route of virus-laden aerosol or droplet in a children’s infusion center. Airborne transmission driven by human activities and indoor ventilation is of crucial importance. Contact transmission of aerosol realized by wall capture, body contact, seat or superficial deposition should not be underestimated, especially for children.(B) A common scenario in a children’s infusion occupied by children receiving transfusion therapy, their accompanying adults, and a potential respiratory disease infector.(C) Experimental tests for the indoor ventilation and dispersion of aerosol particles exhaled from a manikin in a ward. A six-stage Anderson sampler is used for aerosol sampling.(D) Computational fluid dynamics simulation of airflow, aerosol dispersion and validation against experimental data. Different breathing rates and velocities for children and adults are considered.(E and F) Airflow pattern, aerosol concentration, and deposited particles on walls/human body in the infusion room shown in contours and plots.(G) Exposure risk assessment of respiratory pathogens based on the aerosol concentration in a spherical area surrounding a child’s or an adult’s nose. Overview of multipath-transmitted aerosol behaviors and exposure risk assessment (A) Typical transmission route of virus-laden aerosol or droplet in a children’s infusion center. Airborne transmission driven by human activities and indoor ventilation is of crucial importance. Contact transmission of aerosol realized by wall capture, body contact, seat or superficial deposition should not be underestimated, especially for children. (B) A common scenario in a children’s infusion occupied by children receiving transfusion therapy, their accompanying adults, and a potential respiratory disease infector. (C) Experimental tests for the indoor ventilation and dispersion of aerosol particles exhaled from a manikin in a ward. A six-stage Anderson sampler is used for aerosol sampling. (D) Computational fluid dynamics simulation of airflow, aerosol dispersion and validation against experimental data. Different breathing rates and velocities for children and adults are considered. (E and F) Airflow pattern, aerosol concentration, and deposited particles on walls/human body in the infusion room shown in contours and plots. (G) Exposure risk assessment of respiratory pathogens based on the aerosol concentration in a spherical area surrounding a child’s or an adult’s nose. Results Ventilation-dependent airflow pattern To demonstrate how airflow pattern and ventilation mode alter the short-range transmission of aerosols for children, aerodynamic analysis of the vortical structures around the susceptible individuals in the infusion center under two types of ventilation mode was performed in typical iso-surface and streamline. Evidence of strong correlation between the airflow pattern and ventilation mode can be found in Figure 2, indicating two dominating vortexes in scenario VV1 and one dominating vortex in scenario VV2.43 High magnitude of air velocity (>0.7 m/s) and vorticity (1 or −1) is observed near the room corner or the individuals seated near walls, indicating the presence of turbulent structures and recirculation phenomena. Airflow was not necessarily clustered near the individual sitting close to the air supply vent. Comparing airflow around adults and children, vertical development of airflow driven by ventilation and thermal plume originating from human body led to slight discrepancy of air velocity at the respiratory heights for adults and children. The form of airflow determined that when an infected person was in the center of the room, the exhaled aerosol particles would be transported to the whole room under the action of airflow, and it was not easy to be fully replaced in a confined setting with high density of susceptible individuals. Another important factor evaluating different ventilation options was the air changes per hour (ACH) at which the air got replenished with fresh air (Figures 2B and 2D).44 As one might expect, area-averaged air age at the breathing plane for children, which was employed to qualitatively assess ventilation effectiveness, deceased prominently (66.43% for VV1 and 66.52% for VV2) when ACH increased from 6 h−1 to 18 h−1.Figure 2Airflow patterns under two ventilation modes(A) Iso-surface of λ2 criterion with λ equals 1 and -1 under scenario VV1-12 at stable state.(B) Area-averaged air age at the breathing height for children under scenario VV1 with different ACH values.(C) Iso-surface of λ2 criterion with λ equals 1 and -1 under scenario VV2-12 at stable state.(D) Area-averaged air age at the breathing height for children under scenario VV2 with different ACH values.(E and F) Contours and vectors of velocity at cross-section planes through adults and children, and horizontal planes at the breathing height of adults and children. Airflow patterns under two ventilation modes (A) Iso-surface of λ2 criterion with λ equals 1 and -1 under scenario VV1-12 at stable state. (B) Area-averaged air age at the breathing height for children under scenario VV1 with different ACH values. (C) Iso-surface of λ2 criterion with λ equals 1 and -1 under scenario VV2-12 at stable state. (D) Area-averaged air age at the breathing height for children under scenario VV2 with different ACH values. (E and F) Contours and vectors of velocity at cross-section planes through adults and children, and horizontal planes at the breathing height of adults and children. Airborne transmitted aerosol and risk assessment To investigate whether aerosol particles follow the airflow pattern and indoor ventilation, the migration strategy of aerosols was assessed by analyzing the concentration contour at selected planes and the behavior of suspended particles under scenario VV1, considering a release time frame ranging from 30s to 240s (Figure 3). Once exhaled from the infected individual C3P, the aerosol concentration increased significantly at the nearest plane 3 (c > 1 × 10−14 kg/m3). As the aerosol further evolved, the particles were first transported toward the ceiling direction (t = 60s), then spread to the north wall (t = 120s), which indicated the association between the recirculating vortex and airborne transmission of aerosol. Quantitative comparisons of the surface integral of aerosol concentration (Figure 3B) further confirmed this association as the aerosol particles spread from plane 3 located at the center of the infusion room to other planes. At 240s, surface integral of aerosol concentration at plane 3 reduced to 9.4 × 10−13 kg/m, and similar concentration values at all planes indicated the fully mixed state of aerosol in the whole room. From the viewpoint of particle number, suspension rate of indoor aerosol particles gradually increased and then stabilized, and the change of suspension rate per second is less than 0.2% (Figures 3C and 3D).Figure 3Evolution of aerosol concentration and suspended particle number(A) Concentration of aerosol particles at five selected planes after released for 30s, 60s, 120s, and 240s.(B) Surface integral of aerosol concentration at five planes at four typical moments.(C) Particles in the infusion room at four typical moments colored by birth time.(D) Variation of suspended aerosol number with time. Changes of suspended aerosol remain at low level after 240s, indicating a dynamic stable state of the migrating aerosol. Evolution of aerosol concentration and suspended particle number (A) Concentration of aerosol particles at five selected planes after released for 30s, 60s, 120s, and 240s. (B) Surface integral of aerosol concentration at five planes at four typical moments. (C) Particles in the infusion room at four typical moments colored by birth time. (D) Variation of suspended aerosol number with time. Changes of suspended aerosol remain at low level after 240s, indicating a dynamic stable state of the migrating aerosol. To capture the airborne-transmitted aerosols posing a potential threat to the susceptible individuals through respiratory activities, the volume-averaged aerosol concentration within the breathing zone was illustrated (Figures 4A and 4B), wherein the breathing zone was delineated by a sphere encompassing the nose, featuring distinct radii for adults (0.2 m) and children (0.162 m). What was unexpected was that the aerosol concentration in the breathing zone of child C3 was not necessarily protruding among all the susceptible individuals under scenario VV2, especially compared with the peak value of concentration around nurse’s nose at 30s, which further verified the significance of ventilation mode on airborne transmission. The enlarged plots excluding host C3 and C3P indicated that children and adults shared similar peak value of concentration in general except from the nurse in the corridor under scenario VV1-12 (Figure 4A). Once ventilation mode changed into VV2-12, the statistical data indicated that children, C5 and B4 for instance, were more susceptible than adults (Figure 4B).Figure 4Aerosol concentration within breathing zone and infection probability for susceptible individuals(A) Evolution of volume-averaged aerosol concentration within breathing zone for each individual under scenario VV1-12, and the enlarged plot excluding the host C3 and C3P. The child C3 has the highest aerosol concentration among the susceptible individuals except from the infected host C3P.(B) Evolution of volume-averaged aerosol concentration within breathing zone for each individual under scenario VV2-12, and the enlarged plot excluding the host C3 and C3P.(C and D) Infection probability of SARS-CoV-2 Alpha variant, SARS-CoV-2 Omicron variant, and influenza virus for susceptible individuals calculated based on improved Wells-Riley model under scenario VV1-12 and VV2-12. The horizontal and vertical coordinates respectively represent the sitting position in the real scene. To facilitate identification, side bar with corresponding sitting position is added. The red pentagram symbol represents the position of the infected host C3P. Aerosol concentration within breathing zone and infection probability for susceptible individuals (A) Evolution of volume-averaged aerosol concentration within breathing zone for each individual under scenario VV1-12, and the enlarged plot excluding the host C3 and C3P. The child C3 has the highest aerosol concentration among the susceptible individuals except from the infected host C3P. (B) Evolution of volume-averaged aerosol concentration within breathing zone for each individual under scenario VV2-12, and the enlarged plot excluding the host C3 and C3P. (C and D) Infection probability of SARS-CoV-2 Alpha variant, SARS-CoV-2 Omicron variant, and influenza virus for susceptible individuals calculated based on improved Wells-Riley model under scenario VV1-12 and VV2-12. The horizontal and vertical coordinates respectively represent the sitting position in the real scene. To facilitate identification, side bar with corresponding sitting position is added. The red pentagram symbol represents the position of the infected host C3P. Additionally, directly identifying the exposure risk of each individual to respiratory pathogens relied on the aerosol concentration in the breathing zone, and also associated with lung ventilation, quantum generation rate produced by infected hosts, and exposure time. How three typical respiratory pathogens SARS-CoV-2 Alpha variant, SARS-CoV-2 Omicron variant, and influenza virus,45,46 which initiated global or seasonal epidemics with widely presumed ability of airborne transmission, affected the infection probability for both adults and children was statistically assessed (Figures 4C and 4D). Unfortunately, no available data on the discrepancy of quanta values between adults and children, even with high possibility to differ, were reported previously to our knowledge. Pairwise comparisons demonstrated that similar infection pattern of spatial distribution can be observed for all three pathogens if individual immunity state was not considered. Notably, infection probability of SARS-CoV-2 Omicron variant can reach a much higher level (88.83% for VV1-12 and 76.06% for VV2-12) than those of the other pathogens with the same amount of exhaled aerosol. Previous reports and real-world cases can strongly back up this result given the widely established infectivity of Omicron virus.47 Additionally, the infection probability of SARS-CoV-2 Alpha variant was almost half of that of influenza virus with the peak value of 16.58% and 30.81% under scenario VV1-12, respectively. Increased exposure risk in the surrounding area of the infected host C3P was observed in all test cases, and two different ventilation modes leads to higher risk of the child C3 (VV1) and the nurse (VV2), respectively. Although multiple factors were considered in assessing the exposure, the volume-averaged concentration was found to be dominant given their high consistence in spatiality. Taking the SARS-CoV-2 Alpha variant for instance, the peak value of estimated risk can be found at the back row of the infection source for scenario VV1 (14.64%), and the corner near north wall for scenario VV2 (11.16%). Also, of note, the turbulent recirculation region localized at the corner near north wall observed in Figure 2 may be highly associated with the increased risk in corresponding position. Throughout the comparing study between adjacent children and adults, it was found that unlike the public’s general perception, children did not have a lower risk due to their smaller tidal volume or lower breathing height, and the infection probability of children was comparable to that of adults in the route of airborne transmission. Aerosol deposition and contact transmission To further explore how the contact transmission in the form of surface containment affected the infection of respiratory pathogens, the presence of aerosol particles deposited on human body, wall, seat, ceiling, and floor was statistically investigated when steady state of ventilation was achieved at 240s as shown in Figure 5. Among all the surfaces concerned, the largest amount of the aerosol particle deposition per unit area was observed on human body, especially under scenario VV2 with high ACH. It was indicated that exposure risk may indirectly rise with increased frequency of personnel interaction, which was more applicable to susceptible children as the bodies of themselves and the accompanying parent were most accessible objects. Contradictorily, high air change rate gave rise to deposition on human body, although it was significantly and negatively associated with airborne transmission of aerosol particles. This underscored the need to balance the ventilation condition on transmission dynamics,41,48 to interpret the epidemiological relevance of pathogen under multiple aerodynamic factors.Figure 5Aerosol particle deposition in the whole infusion room(A) Amount of aerosol particle deposition per unit area of human body, wall, seat, ceiling and floor under six ventilation modes.(B and C) Proportion of deposited aerosol particles on human body including adults and children, walls including the area under the arm span for adults and children, seat, ceiling and floor under scenario VV1-12 and VV2-12.(D and E) Distribution of deposited aerosol particles on four vertical walls, floor, ceiling and selected susceptible individuals under scenario VV1-12 and VV2-12. The pair of susceptible individuals with the largest number of deposited aerosol particles apart from the infected host, referring to A1 and A1P for scenario VV1-12, and C5 and C5P for scenario VV2-12, is selected. To facilitate comparison, the arm span for adults and children are marked with orange and green dash lines on the deposition diagrams of vertical wall. Aerosol particle deposition in the whole infusion room (A) Amount of aerosol particle deposition per unit area of human body, wall, seat, ceiling and floor under six ventilation modes. (B and C) Proportion of deposited aerosol particles on human body including adults and children, walls including the area under the arm span for adults and children, seat, ceiling and floor under scenario VV1-12 and VV2-12. (D and E) Distribution of deposited aerosol particles on four vertical walls, floor, ceiling and selected susceptible individuals under scenario VV1-12 and VV2-12. The pair of susceptible individuals with the largest number of deposited aerosol particles apart from the infected host, referring to A1 and A1P for scenario VV1-12, and C5 and C5P for scenario VV2-12, is selected. To facilitate comparison, the arm span for adults and children are marked with orange and green dash lines on the deposition diagrams of vertical wall. The next objective was to perform a comprehensive statistical analysis of the total amount of deposited particles on each surface while distinguishing the individual proportions of adults and children (Figures 5B and 5C). The vertical walls, ceiling and human body captured most of the deposited aerosol particles due to their large covering area. As to the human body, most of the deposited particles were on adults under both ventilation modes, for instance, 24.8% for adults and 3.1% for children under scenario VV1-12. The ceiling and the vertical walls beyond the range of arm span for adults or children, however, were not of significance in the viewpoint of contact transmission. Taking the distribution of the deposited particles on each surface into account (Figures 5D and 5E), only a small part of the deposited particles (5.7% for adults and 0.9% for children) on vertical wall was accessible for adults or children under scenario VV1-12. For VV1-12, however, nearly half of the wall deposition (14.0% for adults and 8.8% for children) posed a threaten in term of contact transmission, which was suspected to be largely attributed to the turbulent recirculation near the corner according to the distribution of particles on north and east walls. In addition to the above-mentioned holistic analysis that covers the entire infusion room, the detailed deposition dynamics of aerosol particles on the body or clothes of each individual were considered, including the divided parts for adults and children, as depicted in Figure 6. Increased air change rate per hour was found to homogenize the distribution of particle deposition on human body (Figures 6A‒6C), and lead to manifolding personnel deposition in total amount combining the observation mentioned in Figure 5A. The deposition per unit area on the most susceptible child C3 in the viewpoint of airborne transmission (VV1) dropped from 464 pcs/m2 to 84 pcs/m2, when ACH changed from 6 h−1 to 18 h−1. Contrarily, deposition per unit area on the individuals sitting near the south wall and the nurse increased prominently attributing to the more well-distributed ventilation. One notable exception existed under scenario VV2, which was the unusual increase of deposition on host pair C5 and C5P. It was implied that the recirculation region near the north wall in term of airflow pattern was not removed or transported under higher ACH. Besides, the deposition number per unit area between adjacent adults and children did not differ much, and even in multiple cases, children suffered more from deposited particles than adults. Despite the dynamic behavior of medical staff was not considered, the deposition of aerosol particles was remarkable under most ventilation conditions, suggesting special care and protection should be taken.Figure 6Deposition of aerosol particle on human body(A‒C) Deposition number of aerosol particle per unit area of human body under scenario VV1 with three ACH. The individuals with high amount of deposited aerosol are highlighted.(D‒F) Deposition number of aerosol particle on unit area of human body under scenario VV2 with three ACH.(G) Four main components of human body including head, thorax, upper limbs, and lower limbs are divided for adults and children.(H and I) Proportion of deposited aerosol on each component of the adults and children with the highest deposition amount except from the infected host are shown, which are A1P and A1 under scenario VV1-12, and C5P and C5 under scenario VV2-12. Deposition of aerosol particle on human body (A‒C) Deposition number of aerosol particle per unit area of human body under scenario VV1 with three ACH. The individuals with high amount of deposited aerosol are highlighted. (D‒F) Deposition number of aerosol particle on unit area of human body under scenario VV2 with three ACH. (G) Four main components of human body including head, thorax, upper limbs, and lower limbs are divided for adults and children. (H and I) Proportion of deposited aerosol on each component of the adults and children with the highest deposition amount except from the infected host are shown, which are A1P and A1 under scenario VV1-12, and C5P and C5 under scenario VV2-12. Statistically speaking, physical contact was usually performed between one’s upper limbs to the other’s upper limbs and thorax, while for children in transfusion treatment, lower limbs of accompanying adults should also be considered given their high possibility of random touch driven by curiosity, eating, or drinking now and then. To harmonize the interpretation of contact transmission of aerosol particle between adults and children, a comparison was made regarding the deposition proportion of each component between adults and children (Figures 6H and 6I). On the whole body, the lower limbs (36.23% for VV1-12 and 66.72% for VV2-12) and head had the largest and smallest deposited aerosol number for adults in both scenarios, respectively. Considering the height difference between adults and children, the deposited aerosol on adults’ lower limbs showed strong relevance on the contact transmission. Although the proportion of upper limbs were relatively low (26.09% for VV1-12 and 13.06% for VV2-12) for adults, the highest probability of physical contact with children by hands or arms suggested that they should not be underestimated. Discussion The central aspect of this work is to comprehensively interpret how the airborne and contact transmission of virus-laden aerosol conduct in ventilated indoor environment full of susceptible adults and children. The relative significance of multiple influencing factors, such as ventilation mode, air changes per hour, sitting position, exposure time, and tidal volume of susceptible hosts, on the transmission dynamics is demonstrated through a statistical approach. To present a straight-talking feature comparison of the exposure risk, the airborne transmission of SARS-CoV-2 Alpha variant, SARS-CoV-2 Omicron variant, and influenza virus was estimated with an improved Wells-Riley model. Given the fact that rare well-recognized methods existed to directly estimate the infection probability of physical contact, the statistical significance contact transmission was interpreted with detailed analysis of individualized deposition behaviors, especially in the presence of young children. Although several prior studies have examined the dynamics of indoor transmission of exhaled aerosol,49,50,51 the present work extends the fundamental science of aerosol transmission in three important ways. First, the current and previous studies clearly indicate that the ventilation conditions including air supply modes and ACH drive the aerodynamics of aerosol dispersion, and play one of the leading roles in airborne transmission.30,52 In our study, it was found that the increased aerosol concentration in the breathing zone of susceptible hosts strongly associated with the turbulent vortex and recirculation structure, which were frequently observed around the room corner or leeward walls. Through improving ACH from 6 h−1 to 18 h−1, the area-averaged air age in the infusion room was reduced with 66 percent, indicating that the suspended particles can be removed rapidly, and the short-term exposure of airborne aerosol can be significantly suppressed. The infection probability of SARS-CoV-2 Alpha variant, SARS-CoV-2 Omicron variant, and influenza virus were compared and proved to be consistent with the previous report and real-world cases. Unlike the public’s general perception, children did not differ much from their accompanying adults in exposure risk by airborne transmission, although they were assumed to have smaller tidal volume and lower breathing height. Second, it should be highlighted that a large proportion of aerosol particles were captured by human body, and high ACH was not always beneficial to lowering exposure level by contact transmission regarding to the increased aerosol deposition. Actually, increasing ACH led to a more even distribution and larger number in total amount for deposition statistics on human body in the whole room, which required a balanced ventilation strategy considering both airborne and contact transmission. Deposition of aerosol particles on the vertical wall that was accessible for adults or children relied much on the airflow dynamics, which was similar to the airborne transmission. Under the ventilation mode VV2-12, for instance, nearly half of the wall deposition was within the arm span of adults and children, which further added in the exposure risk. The medical staff in the infusion room may suffer a lot from the deposited particles in either ventilation conditions due to larger exposed body, let along their walking behaviors may induce resuspension of deposited aerosol.53 Third, in an attempt to implement respiratory disease interventions for young children, special attention should be paid on the contact transmission due to their high probability of interaction with people, walls, or items.54 Rare available approaches, however, can be referenced to statistically weigh the infection probability of contact transmission similar to the exposure risks originated from airborne transmission assessed by Wells-Riley model, which was partly attributed to the unpredictable, individualized, and unstable nature of children’s behavior tendency. Also, of note, the height difference and interaction idioms between adults and children determined that the deposited aerosol on adults’ upper and lower limbs, which occupied more than half of the deposition number of human body, showed strong relevance on the contact transmission. It was suggested that ensuring hand hygiene with disinfecting wipes or other means and avoiding high-risk activities may reduce the exposure risk of children by contact transmission while transfusion treatment was conducted. Based on the aforementioned research, the following suggestions are put forward to minimize children’s risk of exposure to infections: First, they should avoid sitting near walls or corners and prioritize well-ventilated seating positions. Second, protective measures during medical visits should be heightened, and unnecessary physical contact minimized. Equally important is post-visit disinfection and clothing change to prevent potential contamination. Taken together, these results provide multiple insights into the migration strategy of respiratory pathogens within a confined facility and the related exposure risks in both airborne and contact transmission in response to the global or seasonal epidemics of respiratory diseases. It is demonstrated that the bioaerosol transport patterns are significantly subjected to indoor ventilation modes, air change rate, and age-related characteristics of susceptible individuals. The methodology proposed in the paper, which involves a systematic analysis of spatial airflow patterns and bioaerosol migration mechanisms to assess the risks of both airborne and contact transmission, can be applied to similar indoor environments, such as hospital waiting rooms, school buses, and classrooms. Proper implementation of this approach could enable the visualization of respiratory infection mechanisms in a variety of scenarios, enhancing people’s understanding of the dynamic spread of respiratory diseases. This, in turn, could provide valuable guidance for non-pharmaceutical interventions aimed at safeguarding the health of children during epidemics. These insights hold the potential to provide guides on disease intervention, not only within the context of pediatric medical settings but also in diverse indoor environments like schools, transportation vehicles, and public spaces. The analytical approach employed in this study, combined with the pragmatic recommendations formulated, exemplifies a pathway toward safeguarding vulnerable populations, enhancing public health resilience, and fostering a more informed response to infectious disease challenges. In a world continually confronted by evolving pathogens and epidemiological threats, this research can serve as an indispensable tool in our arsenal to better understand, mitigate, and ultimately prevent the transmission of respiratory diseases. The methodology proposed in the paper, which involves a systematic analysis of spatial airflow patterns and bioaerosol migration mechanisms to assess the risks of both airborne and contact transmission, can be applied to similar indoor environments, such as hospital waiting rooms, school buses, and classrooms. Proper implementation of this approach could enable the visualization of respiratory infection mechanisms in a variety of scenarios, enhancing people’s understanding of the dynamic spread of respiratory diseases. This, in turn, could provide valuable guidance for non-pharmaceutical interventions aimed at safeguarding the health of children during epidemics. Limitations of the study There are, to be sure, uncertainties and several open questions remaining in our analysis approach: The aerosol particles generated by the respiratory behaviors such as talking, sneezing, or coughing are not considered as the most representative scenario of continuous breathing is concerned. This limits our ability to determine which type of releasing conditions pose the most threaten to other hosts. How to quantitatively determine the risk originated from deposited aerosol in contact transmission for children is not fully understood given that related high-risk activities are often unpredictable, subjective, and individualized, although its connection with children’s exposure to respiratory pathogens is recognized. Furthermore, secondary suspension of deposited aerosol particles randomly induced by mechanical ventilation or human activities can cause increased risk of airborne transmission of respiratory pathogens, which are not accounted for in the present work. Nevertheless, despite these caveats, strong empirical support for both airborne and contact transmission of respiratory pathogens are present with quantitative evidence of the migration strategy of virus-laden aerosol. Among measures to directly assess the aerodynamic significance such as mechanical ventilation, airflow pattern, aerosol particle diffusion and deposition, the associated influence on children’s exposure risk and health are addressed compared with those of well-documented adults. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERBacterial and virus strainsEscherichia coliAcademy of Military SciencesCMCC (B) 44102; RRID: WB-STRAIN:WBStrain00040993Software and algorithmsAnsys 2022 R1ANSYShttps://www.ansys.com/Tecplot 360EX 2022 R1Tecplothttps://www.tecplot.com/OtherTSI 9535 anemometerTSIhttps://tsi.com/cn/Six-stage AndersonJunrayhttps://www.instrument.com.cn/netshow/SH102701/Bioaerosol generatorJunrayhttps://www.instrument.com.cn/netshow/SH102701/ REAGENT or RESOURCESOURCEIDENTIFIERBacterial and virus strainsEscherichia coliAcademy of Military SciencesCMCC (B) 44102; RRID: WB-STRAIN:WBStrain00040993Software and algorithmsAnsys 2022 R1ANSYShttps://www.ansys.com/Tecplot 360EX 2022 R1Tecplothttps://www.tecplot.com/OtherTSI 9535 anemometerTSIhttps://tsi.com/cn/Six-stage AndersonJunrayhttps://www.instrument.com.cn/netshow/SH102701/Bioaerosol generatorJunrayhttps://www.instrument.com.cn/netshow/SH102701/ Resource availability Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Chenxing Hu (chenxinghu@bit.edu.cn). Materials availability All unique/stable reagents generated in this study are available from the lead contact with a completed Materials Transfer Agreement. Data and code availability •All data reported in this paper will be shared by the lead contact upon request.•All original code has been deposited at Dataverse and is publicly available as of the date of publication. DOIs: https://doi.org/10.7910/DVN/ID4DGD.•Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. •All data reported in this paper will be shared by the lead contact upon request. All data reported in this paper will be shared by the lead contact upon request. •All original code has been deposited at Dataverse and is publicly available as of the date of publication. DOIs: https://doi.org/10.7910/DVN/ID4DGD. All original code has been deposited at Dataverse and is publicly available as of the date of publication. DOIs: https://doi.org/10.7910/DVN/ID4DGD. •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. Experimental model and study participant details Escherichia coli Escherichia coli with the ATCC 13706 reference number is used in the model bacteria experiment. Method details Study design and scenario description In response to the seasonal outbreaks of respiratory diseases, this study developed a computational modeling framework of dispersion and deposition quantification for the exhaled aerosols in a children’s infusion room (7.6 m × 7.4 m×4.2 m) divided into 5 rows and 6 columns by the seats (Figure S1A). Considering the epidemic prevention and control, scattered arrangement of 16 children and their accompanying adults under the most intensive state of personnel, are seated next to each other. Due to the large number of personnel, individuals are marked by the horizontal and vertical numbers, children said A1, A3, A5 …, and adults are A1P, A3P, A5P … Two representative types of ventilation in medical institutions according to Capital Medical University Affiliated Beijing Luhe Hospital are selected in the infusion room,55,56 which is the mixed ventilation with the square diffusers of 0.6 m × 0.3 m and horizontal direction of 30° (marked by VV1, Figure S1A), and the displacement ventilation with the square diffusers of 0.3 m × 0.3 m and normal wind direction (marked by VV2, Figure S1B). Variations of air changes per hour (ACH) including 6 h−1,12 h−1 and 18 h−1 are taken into account for both ventilation types.44,57 Aerosol particles were released at a mass flow rate of 9.048 × 10−15 kg/s in the nasal cavity of an infected person (C3P) only during exhalation since only normal breathing action was considered. Considering the complexity of respiratory activities and the significant proportion of 0.4 μm aerosol particles in the total particle size distribution, following Gupta et al.'s recommendation, there are a total of 525 aerosol particles with a diameter of 0.4 μm emitted from the nostrils with each breath58 (Figure S1D). Continuous breathing process follows Gupta59 as V1=−3.73×sin(π×0.498×t), where V1 is the adult respiratory rate and t is the breathing time. For children with remarkable discrepancy from adults in respiratory rate and expiratory volume, the respiratory model for children was adjusted into V2=−2.65×sin(π×0.667×t) according to the tidal volume of school-age children (4000 mL/min) and adults (7500 mL/min).60,61 Simulations of aerosol dispersion and deposition To describe the airflow pattern and aerosol behaviors in the infusion room, the computational modeling of the unsteady Navier-Stokes equations with RNG k-ε turbulence model62 was performed. SIMPLE algorithm was used for velocity-pressure coupling,63 and Boussinesq model was employed to approximate the buoyancy effect.64 Aerosol particles were represented as discrete phase particles whose spatial and temporal evolution was governed by a Lagrangian model.65 Given that the density of aerosol particle is quite small compared with that of air, the pressure gradient, Basset force, and virtual mass force are ignored.66 Thus, the investigation encompassed the impact by thermophoretic force, Saffman’s lift force, and Brownian force. Figure S2 describes the procedure for grid independence verifications for two scenarios of the infusion room. In the numerical simulation, air supply vents were set as the inlet of the computational domain with the turbulence intensity of 5%. For the scenario VV1, the air supply vent is in the form of square diffuser, which was divided into 16 uniform-sized units to simulate the air supply angle of 30°.67,68,69 For the scenario VV2, a typical grid diffuser was employed to supply uniform and normal wind speed. The temperature of air supply was 26°C, the human body surface 31°C,70 the aerosol particle and exhalation 35°C.71 The aerosol particles released from the infector C3P were allowed to escape through the air supple vents and captured by human body, walls or seats. Based on the approximate perfect mixing theory proposed by Gupta,72 it is assumed that the droplets are fully mixed in the cabin after a period of time. Change of aerosol particle suspension rate per second in the room is less than 1% when it reaches 240 s. Sample collection and validations To assess the effectiveness of the computational model, an experimental assessment of indoor aerosol dispersion was conducted at North China Electric Power University.73 An isolation ward with a thermal manikin as the source of the exhaled aerosol, as described in Figure S3, is ventilated with two air supply vents (0.24 m × 0.135 m) and one exhaust vent (0.27 m × 0.27 m). The airflow velocity measurement is conducted using the TSI 9535 anemometer, with a measurement range of 0–30 m/s and an accuracy of ±3%. A thermal dummy model is placed on the central bed of the measurement room (Figure S4C). The nose of the dummy is used as the divergence source of the divergent aerosol particles in the experiment. The artificial saliva74 (Figure S4A) is sprayed into the room at a speed of 0.5m/s using a collision bioaerosol generator (ZR-C04) to generate aerosol particles with an initial particle size of 1μm (Figure S4B), simulating 13 min breathing behavior (Figures S4A and S4B). When preparing artificial saliva, glycerol and PBS are added to it, taking into account its activity. Studies have shown that Escherichia coli has been widely used in bioaerosol diffusion experiments. Escherichia coli was round and white in ordinary agar medium, with obvious identification. In addition, the biosafety level of E. coli was 1, which is commonly used in bioaerosol diffusion experiments in medical environments. In the experiment, therefore, Escherichia coli was added to artificial saliva to clearly observe the divergence of aerosol particles. In order to reduce the influence of background concentration on the experimental results, the ward must be fully ventilated before the start of each experiment. Four six-stage Anderson samplers were prepared in advance. Six-stage Anderson sampler was placed in many blank agar plates (Figure S4D). After the experiment, the thermal manikin released the aerosol particles for 10 min, and the sampler was opened for sampling for 3 min. After sampling, the collected agar plates were placed in a 37°C incubator for 24 h (Figure S4E), and the CFU (Colony-Forming Units) of Escherichia coli was recorded.75,76 In order to reduce the experimental error, the above operation is repeated three times. An analysis of experimental uncertainty was undertaken, encompassing considerations for the precision of the anemometer, leading to an uncertainty of 0.03 for the wind speed measurement experiment. Furthermore, a comprehensive assessment was conducted, taking into account factors such as the accuracy of the aerosol generator, the uncertainty associated with the six-stage Anderson sampler, deviations in incubation time, and operational errors within the CFU counting process. As a result of this comprehensive evaluation, the uncertainty for the aerosol concentration collection experiment was determined to be 0.118. Quantification and statistical analysis Graphs and statistical analysis In the experiment, data analysis was conducted using Origin 2023, and the statistical results can be found in Figure S4. The analysis involved the assessment of aerosol concentrations at specified wind speeds. The reported values represent the dimensionless arithmetic mean of three independent replicate experiments. The size of the error bars was obtained by calculating the standard deviation of dimensionless values from the three replicate experiments, reflecting the range of variability in the experimental measurements. The results of numerical simulations are processed into images using Tecplot 360 EX 2021 R1. Exposure risk assessment and quantification To quantitatively assess the exposure risk to the respiratory pathogens including SARS-Cov-2,45 and influenza,46 an improved Wells-Riley model evaluating the infection probability by airborne transmission was employed. The Wells-Riley equation is incorporated into computational fluid dynamics, with ignoring the effect of microbial activity. The aerosol concentration in the respiratory zone around the nose of a susceptible person, which is defined as a spherical area with the radius of 0.2 m for adults77 and 0.162 m for children based on the discrepancy in expiratory volume, is taken as the concentration of inhaled aerosol determining the exposure risk.