key: cord-0937751-og34t4bt authors: Rochoy, M.; Fabacher, T.; Cosperec, I.; Wendling, J.-M. title: Experimental efficacy of the face shield and the mask against emitted and potentially received particles date: 2020-11-24 journal: nan DOI: 10.1101/2020.11.23.20237149 sha: 7ba93220979613410be85eb25cb963635ef4ee50 doc_id: 937751 cord_uid: og34t4bt The aim of this study was to evaluate the comparative performance of masks and face shields in different experimental configurations. An experimental setup with two mannequin heads positioned at 1.70m high and at 25 cm each other was used. A fogger generated a particle's airflow with a speed of 5m/sec from the emitter to the receiver head mannequin. Our aerosol generator produced 3 000 times more particles than a physiological cough situation. A particle counter allowed us to evaluate the number of particles received on a mannequin head located at a very short distance of 25 cm. The amount of all particles up to the selected particle sizes were counted with an optical particle counter on channels 0.3 m, 0.5 m, 1 m, 2.5 m, 5 m and 10 m. The reduction factors with a protection worn by the receiver alone, by the emitter alone and then the double protection of emitter and receiver were calculated. When the receiver alone wore a face shield, the amount of total particles was reduced (54.8%), while the reduction was less when the receiver alone wore a mask (21.8%) (p = 0.003). Wearing a protection by the emitter alone reduced much more the level of particles received by 96.8% for both mask and face shield. The double protection allowed for even better results, but close to the protection of the emitter alone: 98% reduction for the face shields and 97.3% for the masks (p=0.022). Even with small particle size emission ([≤]0.3m), results were of the same order. Considering our results, protection of the emitter alone or double protection is much more effective than protection of the receiver only. Face shield should be included as part of strategies to safely and significantly reduce transmission in the community setting, in addition to masks or for people with disabilities or medical intolerance to masks. At the end of 2019, a novel coronavirus named severe acute respiratory syndrome coronavirus 2 COVID (SARS-CoV-2) emerged [1] . The outbreak of the disease caused by this virus, Coronavirus Disease 2019 (COVID- 19) , was declared a pandemic by the World Health Organization (WHO) on 11/03/2020, and has caused nearly 1 million of fatalities as of 27/09/2020 [2] . The airborne transmission route for SARS-CoV-2 is virulent for the spread of COVID-19 [3] [4] [5] , as for SARS-CoV-1 [6] . At present, neither the aerosol viral load nor the minimum infectious dose of SARS-CoV-2 has been established [7] . A viable virus can be emitted by an infected person by breathing or talking, without coughing or sneezing [8] : the louder the voice, the greater the quantity of aerosols emitted; furthermore, a small fraction of individuals behave as "speech superemitters," consistently releasing an order of magnitude more particles than their peers [9] . A challenge in pandemic control is limiting the transmission of SARS-CoV-2 by asymptomatic or presymptomatic individuals [10] . A systematic review of the literature and meta-analysis revealed that face covering decreased the risk of airborne infections [11] . Surgical face masks significantly reduced detection of coronavirus RNA in aerosols [12] . Face covering by asymptomatic people (the primary case and family contacts before the primary case developed symptoms) is effective in reducing transmission [13] . In the United States, a recent analysis revealed that the difference with and without mandated face covering represented the principal determinant in shaping the trends of the pandemic [4] . However, in Western countries, there has been significant controversy over the face covering [14, 15] , including after the recommendation of the WHO on 5 June [16] . One of the keys to acceptance of the face covering could be the face shield, especially for younger children, during intensive aerobic physical activity, or for people who are anxious about wearing a mask [17] . Face shields could also be of economic and ecological interest. One reason for the controversy surrounding a recommendation to wear a face shield, during this period, is that little is known about the efficacy of different types of protective measures in the context of this pandemic [18] . CDC does not currently recommend the use of face shields as a substitute for masks because there is currently not enough evidence to support the effectiveness of face shields for source control [19] . It has already been shown that the use of face shields can significantly reduce healthcare workers' short-term exposure to larger infectious aerosol particles and can reduce contamination of their respiratory systems. They are less effective against smaller particles, which can remain airborne for long periods of time and can be easily circulated around a face shield for inhalation [20] . We hypothesized that face shields could reduce the amount of droplets emitted and received, particularly in the context of widespread use (worn by the emitter and worn by the receiver). We aimed (1) to quantify the number of particles of different sizes detected in front of a patient with different types of screens (face shield, mask or nothing); (2) compare these different types of screens. The evaluation was carried out in August 2020 on an experimental setup with two mannequin heads positioned at 1.70m high and at 25 cm each other (Figure 1) . The tests were carried out in an empty closed room of 18.40m² without drafts nor mechanical ventilation, and with a sectional door of size 8.4m² closed during the tests. One of the two heads (called an "emitter" or Em) has been hollowed out to reproduce a mouth. A pipe was introduced through the head to connect the fogger to the mouth. The pipe had an internal diameter of 18mm. The atomizer was generated by aerosolizing distilled water with a fogger TRIXIE Fogger XL: the flow rate was continuous at 5 ml/min coming out of the particle airflow generator with a speed of 5m/sec. The other head (called a "receiver" or Re) has also been hollowed out at the mouth. A short pipe of 5 cm long and 15 mm in diameter was connected to an optical particle counter. The optical particle counter (TROTEC PC220) is designed to measure the size and amount of particles in the air. It sucks in air for an adjustable amount of time and determines the size and amount of particles contained in it. The device is equipped with an integrated measuring cell with laser (class 3R laser, 780nm, 1,5-3mW). Particles of sizes less than 0.3 µm, 0.5 µm, 1 µm, 2.5 µm, 5 µm and 10 µm were treated equally during the process. The cumulative method of analysis was performed. The amount of all particles up to the selected particle sizes were counted (e.g. : "0.5 µm = 417" means that 417 particles had a size between 0.3 and 0.5 µm). The pumping time, air volume and the start delay are programmable. A HEPA filter was used on the counter to reset to zero before each measurement. The temperature, ambient humidity and the aerosol flow velocity were recorded using a hot-wire thermo-anemometer TROTEC TA300. This device comes equipped with a hot-wire sensor and microprocessor technology for signal amplification. This combination guarantees precise measuring results. Two PPE were tested and their performance were compared (Figure 2) : -EN14683 surgical medical device masks (over 95% of 3µm particle filtration), -Face shields (masks) (Spanish brand) respectively: height covering the eyes, mouth, nose -22.5 cm high with overhang under the chin of 7 cm, circumference of the visor 35 cm, front opening 4 cm high in line with the center of the forehead. The background noise (background level of particulate pollution) was first evaluated in the experimental room (10 measurements) (situation 0). The aerosol was then generated, without any PPE. The airflow speed at the emitting mouth was noted. The particle amount and the size distribution of the aerosol were measured at a distance of 25 cm (9,85 inch) (10 measurements) (situation 1). 6 configurations with PPE devices were then tested with each time a serie of 10 measurements: -surgical mask (situation 2), then face shield (situation 2A) on the receiver head only, -surgical mask (situation 3) then face shield (situation 3A) on the emitter head only, -surgical mask (situation 4) then face shield (situation 4A) on both emitter and receiver heads (double protection). The aerosol generator was started at maximum power. The counter was started at the same time but with a 5 second programmed delay from the generator. The particle counter calculated the total cumulative particles aspirated on a volume of 1.416 liters. Counting was performed on the 6 channels (0.3 µm -0.5 µm -1 µm -2.5 µm -5 µm -10 µm) during 30 seconds of sucking in air. After each measurement, the counter was reset to zero by the HEPA filter and the room ventilated by opening the sectional door for at least 5 minutes to remove airborne particles. Before testing, temperature and humidity were recorded. Our Emitter-Receiver basic configuration was not changed during the entire experimentation: only masks and face shields have been exchanged for the data acquisition. First, the amount of particles and the distribution of the particle size at a distance of 25 cm (9.85 inch) were studied without any PPE. For the experiments performed with the aerosol particle measurement instruments, the parameters studied and compared were: -the impact of the presence of PPE, -the difference between protection of the emitter, receiver or both protection, -the influence of the type of PPE, mask or face shield, -the influence of the aerosol particle sizes on the results (0.3 µm -0.5 µm -1 µm -2.5 µm -5 µm or 10 µm respectively). For descriptive analysis we used mean, median and standard deviation for continuous variables. Distribution of each particle type was described by their means frequencies and percentages. In order to analyze the impact of the type of protection (mask vs. face shield) and location of protection (emitter, receiver or both), a hierarchical linear regression was used. For categorical variables, face shield, emitter and aerosol particle < 0.3 µm were used as references. The logarithm of the number of each particle type detected in each experience was modeled. The covariates included in the model were the type of PPE and the location of protection. Two random effects, one on the experiment and the other on the type of particles, were added. Sensitivity analyses were performed with comparison between the different experiments and subgroup analysis for micro (<5µm) and macro particles (≥ 5µm). The alpha risk was set at 5% for all analyses, and R software version 3.6 was used. The temperature and hygrometry in the experimental room were regularly measured (n = 20). The mean values were respectively 27.73°C (sd = 0.50°C) and 68.3% (sd = 2%). Before aerosolization, the thermo-anemometer confirmed the absence of significant air current in the room around the test bench (values < 0.05m/sec in all directions). During the production of aerosol by the fogger, on the first step of the study, the speed of the airflow coming out the pipe without any protection was evaluated at the mouth of the mannequin head with the thermo-anemometer. The speed airflow mean value was 4.95 m/sec (sd = 0.17 m/sec) (n = 20). All the measurements were cumulative countings of the different particles at the same distance (25cm or 9.85 inch) during the same time (30 sec) after the same delay (5 sec), between emission and start . CC-BY-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; of the particle counter on the receiver side. The air volume sucked (1.416 l) was identical in all experiments. The background level of particulate pollution on the sucked in air volume standard in the experimental room showed a mean value of 13,837 particles (sd = 2,436). The mean concentration was 9,171 total particles/liter (Appendix Table 1 ). The aerosol produced has a distribution of the particles, received at 25 cm, decreasing according to size. The average amount of particles measured at 25 cm without protection was 1,000,763 particles (sd = 165,118) with 60% particles received less than 0.3µm (Figure 3) . The number of detected particles decreased with increasing particle sizes. This distribution was homogeneous in the different experiments: the linear regression model shows a constant decrease in the estimates compared to less than 0.3 µm particles (p<.001). All experiences statistically lead to detected more particles compared to background level of particulate pollution (p<0.05) (Appendix Table 2 ). All experiences statistically lead to less detected particles compared to experiment without any PPE (p<0.05) (Appendix Table 3 ). When the receiver head wore a mask, it significantly reduced the amount of total particles found by 21.8%; when the receiver head wore a face shield, the reduction was significantly higher: 54.8% (p = 0.003). When the emitter head wore a mask or a face shield, it significantly reduced the amount of total particles found by a greater level of 96.7%. When the receiver and the emitter wore a mask, it significantly reduced the amount of total particles found by a greater level of 97.3%; when they two wore a face shield, the reduction was significantly higher: 98.0% (p = 0.022) ( Table 1 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint Considering only small in size particles (≤0.3µm) when the receiver head wore a mask, the reduction was significantly more important with the face shield 48 % than with the mask 13.2% (p=0.004). For this range of particles, when the emitter head wore the protection, reduction with face shield was 96.9% vs 96% for the mask. With the double protection, we found 97.7% reduction in the number of particles for the face shield compared to 96.8% for the masks (p=0.024) (Figure 4 ) (Appendix Table 4 ). Statistically more particles are detected when the protection is worn by the receiver compared to the protection worn by the emitter alone (p < 0.001). Double protection further reduced the number of particles, but not significantly compared to the protection worn by the emitter alone (p = 0.084). ( Table 2) . Sensitivity analyses in micro and macro particles sub-groups (⪯2,5µm and >2,5µm) found the same results: less particles detected with PPE in both emitter and receiver compared to emitter, less particles detected with PPE on emitter than on receiver and less particles detected with face shield compared to mask (Appendix, Table 5 and 6). In our experimental situation, face shields and masks could reduce the amount of particles emitted and received. The number of particles detected decreased mainly when the emitter wore face protection, and more so when the receiver also wore face protection, especially for micro-particles. As expected, our results indicate more effective protection when the emitter wears face protection. As part of the fight against the spread of SARS-CoV-2, the masks have been generalised to prevent propagation from asymptomatic emitters, as effective "anti-spittoons screens" [21] [22] [23] . The current hypothesis is that the particles emitted during breathing and speech are mainly formed by a mechanism of "fluid film burst" inside the small airways of the lungs and/or by the vibration and adduction of vocal folds in the larynx [4, 24] . More loud is the voice, greater is the amount of aerosols emitted; furthermore, a small fraction of individuals behaves as "speech superemitters," consistently releasing more particles in size and quantity than their peers [9, 25] . Under our experimental conditions, the total number of particles received was significantly lower when wearing a face shield than when wearing a mask by the receiver even for micro-particles. One recent pre-print study, with measurements conducted with a 60 cm horizontal distance between the breathing and the coughing simulators, found that the face shield provided better protection than surgical mask and blocked more than 90% of the otherwise inhaled particles. For droplets larger than 3μm, the efficiency of face shields was found to be comparable to medical masks. The big advantage of the face shield lies on the enhanced protection of parts not covered by the mask. For finer particles (0.3 μm), face shield performed better, as in our study [26] . In another experimental pre-print study [27] , researchers used Background Oriented Schlieren (BOS) imaging to compare seven types of face protections including surgical mask, hand-made mask, heavy duty commercial face shield and a lightweight face shield based on a 3D printed headband. The results showed that during quiet, heavy breathing or cough, no front throughflow was discernible for the . CC-BY-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; lightweight printed 3D face shield. For surgical masks, air front jets travelled more than 23.7 cm. The maximum distances travelled by the air jets (crown, brow, side and back jets) were all higher for the surgical mask compared to the face shield. Completely consistent with our observations, face shields generate particularly strong down leakage jets towards the ground while no flow was discernible with the surgical mask in that direction: it seems to us more interesting to fold down the flow of infected particles towards the ground rather than to let them escape in the direction of the receiver. During aerosol-generating procedures (e.g., bronchoscopy, suctioning of the respiratory tract endotracheal intubation) on patients who are not suspected of being infected with an agent for which respiratory protection is otherwise recommended (e.g., M. tuberculosis, SARS or hemorrhagic fever viruses), CDC strongly recommends to wear one of the following: a face shield that fully covers the front and sides of the face, a mask with attached shield, or a mask and goggles [28] . Face shields are commonly used as an infection control alternative to goggles. As opposed to goggles, a face shield can also provide protection to other facial areas. To provide better face and eye protections from splashes and sprays, a face shield should have crown and chin protection and should wrap around the face to cover entirely both ears, which reduces the likelihood that a splash could go around the edge of the shield and reach the eyes. To our knowledge, no studies have compared face shields and masks in an identical configuration or dual protection (emitter and receiver) [29] . Face shields are often used according to the paradigm of personal protection. For example, in 2014, Lindsley and al. studied face shields and concluded that they can reduce the short-term exposure of health care workers to large infectious aerosol particles, but smaller particles can remain airborne longer and flow around the face shield and be more easily inhaled [20] . So, face shields do not seem to be a good alternative to protect oneself, for example for people working in restaurants, from unmasked people who could potentially transmit COVID-19 [20] . But here we propose the generalized use of face shields to protect the environment, since it has been shown (and found in our study) that experimental efficacy is greater [26] . Surgical masks are now the gold standard in the fight against COVID-19. There are limitations to the surgical mask, including potential permeability to particles less than 3 microns, important leaks even if the filtration performance is announced at more than 95%, often improper wear and variable acceptability. Despite these limitations, its effectiveness has been widely demonstrated, particularly in real life in the fight against the spread of COVID-19 [12, 14, 16, 19, 21, 22, 30] . Face shields provide a barrier to acutely-expelled aerosols of body fluids and are commonly used by the health workers as an alternative to goggles because they confer protection to a larger area of the face. Face shields offer a number of advantages [17, 18] . First, they are durable equipments because they are primarily made of plastic. While medical masks have limited durability, face shields can be reused indefinitely, and are easy to clean with soap or antiseptic [31] . The economic interest is obvious for the poorest populations throughout the world. Second, they reduce the potential for autoinoculation by preventing the wearer from touching their mouth and nose. A recent study found that mask does not fit tightly enough on the face to protect against droplet infection, and that only a particle-filtering half-mask that fits tightly offers protection against droplet infection [32] . Third, people wearing medical masks often have to remove them to communicate with others around them or have to adjust the mask; this is not necessary with face shields. The use of a face shield is also a reminder for people around to maintain social distancing, but allows lips movements for speech perception, notably for people with hearing loss [32] . Fourth, face shields appear to significantly reduce the amount of inhalation exposure to influenza virus, another droplet-spread respiratory virus, when only the receiver is protected [18, 20] . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; Finally, face shields reduce the potential for inoculation by eyes, whose interest in the fight against COVID-19 seems under-considered and could be effective in community settings [11, 33] . A recent study found in Hubei hospital found that only 5,8% of COVID-19 patients (16 of 276 patients) wore glasses compared to an estimated 31.5% of the general population, suggesting that eye protection, eyeglasses or face shields could be useful against COVID19 [33] . The ocular mucosa and the nasopharynx are connected by the nasolacrimal duct. When liquid splashes occur at eye, it can be partially absorbed by the cornea and conjunctiva but also get into the nasal cavity through the nasolacrimal duct, be transported toward the nasopharynx and trachea, and at this point swallowed into the gastrointestinal tract. So logically viruses contained in the liquid splashes to which the eye is exposed can be transported to the nasopharynx and respiratory tract mucosa [34, 35] . Ocular manifestations seem frequent, and high-frequency hand-eye contact was correlated with conjunctival congestion [36, 37] . Across all eye specimens, immunohistochemical analysis revealed expression of ACE2 and TMPRSS2 in the conjunctiva, limbus, and cornea, with especially prominent staining in the superficial conjunctival and corneal epithelial surface, suggesting that ocular surface cells including conjunctiva are susceptible to infection by SARS-CoV-2, and could therefore serve as a portal of entry as well as a reservoir for person-to-person transmission of this virus [34, 38, 39] . Very few studies have evaluated the effects or potential benefits of face shields on source control, ie, containing saliva emission during speaking, singing, sneezing or coughing, when worn by asymptomatic or symptomatic infected persons. However, with efficacy ranges of 68% to 96% for a single face shield on the receiver, it is likely that adding source control would only improve efficacy [17] . To evaluate the performance of the face shield as source control, Verma and al. used a cough simulator, a synthetic smoke (particles emitted are possibly more volatile because of the high temperature), a horizontal laser sheet in addition to a vertical laser sheet revealing how the droplets cross the horizontal plane. By placing a plastic face shield they were able to map out the paths of droplets and found that droplets also spread in the reverse direction. The authors did not quantify the number and the distribution of particles emitted, or the decreased concentration with distance. The face screen was positioned semiopen, in an improper way, facilitating the exit of a plume of exhaled air ; we used a right angle position close to the chin down, a chin and forehead overhang, and very important, a position of the visor parallel to the face [40] . Our study has some limitations. The study was only conducted on one surgical and one face shield model. The way to wear it, the shape or design of the face-shield can determine a more or less performance [27] . In addition, this is an experimental study, with fixed conditions such as only one short distance, comparing identical situations regarding masks and face shields, favouring droplets and aerosols transmissions (short duration of 30 seconds). The experimentation situation is an extreme configuration of exposure. Considering human particles (p) emission levels (breathing = 0.31 p/s, speaking = 2.77p/sec, coughing = 10.1 p/s [30] ), our aerosol generator produced 3000 times more particles than a real cough situation, with a very short distance in an enclosed, unventilated space. Real life leads to situations where the emitter is not always facing the receiver, but potentially at all 360° points around the receiver. We made the ballistic assumption that the most risky situation was face-to-face [9, 25] . Our experimentation is based on short and intense transmission. Transmission by aerosolization could also be achieved by long, low-intensity exposure, not tested here in our study. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org /10.1101 /10. /2020 Another situation at risk for SARS-CoV-2 transmission may be feco-oral route and a manuporting route: indeed, SARS-CoV-2 can survive up to 9 hours on human skin, much longer than influenza-virus and is found in wastewater and stools [41] [42] [43] . Surgical masks may be less protective for this pathway than face shields that prevent finger-to-eyes in addition to mouth access; however, current evidence does not suggest that this manuporting pathway is predominant [5] . More studies are needed in real-life situations to determine the best possible uses for face shields. Some microbiological studies on sick people equipped with face shields with environmental analysis on air bio collectors should be conducted. Further experimentations should improve performance through the shape and the face shield design. To protect others in the best possible way, the objective should be to send the exhaled air flow towards the wearer or towards the ground with the most efficient design of the visor. However, the results of our study suggest that they should be used now, in addition to masks as part of an expanded arsenal against SARS-CoV-2: Pennsylvania, Oregon, North Carolina and Singapore have already integrated the face shield as a possible face covering alternative, especially for people with disabilities or medical intolerance to mask. Finally, the efficiency of the face shields also encourages the development of their use where no facial protection is possible (bars, restaurants), through the integration of Plexiglas between users. Our results show that wearing a face shield or a mask by the emitter is much more efficient than only by the receiver. When the receiver alone wore a face shield, the amount of total particles was significantly reduced (54.8%), better than the receiver alone wore a mask (21.8%), even with small particle size emission (≤0.3µm). Wearing a protection by the emitter alone reduced much more the level of particles received by 96.7% for both mask and face shield. The double protection allowed for even better results: 98% reduction for face shields and 97.3% for masks. Face shields are durable, can be reused, and are easy to clean. They prevent the wearer from touching their mouth, nose or eyes, possible route transmissions of SARS-CoV-2. So, face shields should be used in addition to masks to better reduce transmission of SARS-CoV-2. . CC-BY-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.23.20237149 doi: medRxiv preprint The species Severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it SARS-CoV-2 Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1 Identifying airborne transmission as the dominant route for the spread of COVID-19 Detection of Airborne Severe Acute Respiratory Syndrome (SARS) Coronavirus and Environmental Contamination in SARS Outbreak Units Estimation of airborne viral emission: Quanta emission rate of SARS-CoV-2 for infection risk assessment Infectious virus in exhaled breath of symptomatic seasonal influenza cases from a college community Aerosol emission and superemission during human speech increase with voice loudness Natural History of Asymptomatic SARS-CoV-2 Infection Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis Respiratory virus shedding in exhaled breath and efficacy of face masks Reduction of secondary transmission of SARS-CoV-2 in households by face mask use, disinfection and social distancing: a cohort study in Beijing, China Face masks for the public during the covid-19 crisis Advice on the use of masks in the community, during home care and in healthcare settings in the context of the novel coronavirus Moving Personal Protective Equipment Into the Community: Face Shields and Containment of COVID-19 Face shields for infection control: A review Use of Masks to Help Slow the Spread of COVID-19 Efficacy of Face Shields Against Cough Aerosol Droplets from a Cough Simulator Widespread use of anti-spittoons screens (homemade masks): a means to fight the Covid-19 epidemic Face Masks Against COVID-19: An Evidence Review Surgical mask partition reduces the risk of non-contact transmission in a golden Syrian hamster model for Coronavirus Disease 2019 (COVID-19) Modality of human expired aerosol size distributions Visualizing Speech-Generated Oral Fluid Droplets with Laser Light Scattering Examining the protection efficacy of face shields against cough aerosol droplets using water sensitive papers Face Coverings, Aerosol Dispersion and Mitigation of Virus Transmission Risk Isolation Precautions -Guidelines Library Effectiveness of Face Masks in Preventing Airborne Transmission of SARS-CoV-2. mSphere Efficacy of masks and face coverings in controlling outward aerosol particle emission from expiratory activities Stability of SARS-CoV-2 in different environmental conditions Fundamental protective mechanisms of face masks against droplet infections Association of Daily Wear of Eyeglasses With Susceptibility to Coronavirus Disease Role of the Eye in Transmitting Human Coronavirus: What We Know and What We Do Not Know. Front Public Health Evaluation of SARS-CoV-2 in Tears of Patients with Moderate to Severe COVID-19 Ocular manifestations and clinical characteristics of 535 cases of COVID-19 in Wuhan, China: a cross-sectional study Characteristics of Ocular Findings of Patients With Coronavirus Disease 2019 (COVID-19) in Hubei Province Expression of SARS-CoV-2 receptor ACE2 and TMPRSS2 in human primary conjunctival and pterygium cell lines and in mouse cornea Detection of severe acute respiratory syndrome Coronavirus-2 in the tears of patients with Coronavirus disease 2019 Visualizing droplet dispersal for face shields and masks with exhalation valves Survival of SARS-CoV-2 and influenza virus on the human skin: Importance of hand hygiene in COVID-19 SARS-CoV-2 in wastewater: potential health risk, but also data source SARS-CoV-2 in wastewater: State of the knowledge and research needs