key: cord-290277-ndfoppoq
authors: Bahl, Prateek; Doolan, Con; de Silva, Charitha; Chughtai, Abrar Ahmad; Bourouiba, Lydia; MacIntyre, C Raina
title: Airborne or droplet precautions for health workers treating COVID-19?
date: 2020-04-16
journal: J Infect Dis
DOI: 10.1093/infdis/jiaa189
sha: 
doc_id: 290277
cord_uid: ndfoppoq

Cases of COVID-19 have been reported in over 200 countries. Thousands of health workers have been infected and outbreaks have occurred in hospitals, aged care facilities and prisons. World Health Organization (WHO) has issued guidelines for contact and droplet precautions for Healthcare Workers (HCWs) caring for suspected COVID-19 patients, whilst the US Centre for Disease Control (CDC) has recommended airborne precautions. The 1 – 2 m (≈3 – 6 ft) rule of spatial separation is central to droplet precautions and assumes large droplets do not travel further than 2 m (≈6 ft). We aimed to review the evidence for horizontal distance travelled by droplets and the guidelines issued by the World Health Organization (WHO), US Center for Diseases Control (CDC) and European Centre for Disease Prevention and Control (ECDC) on respiratory protection for COVID-19. We found that the evidence base for current guidelines is sparse, and the available data do not support the 1 – 2 m (≈3 – 6 ft) rule of spatial separation. Of ten studies on horizontal droplet distance, eight showed droplets travel more than 2 m (≈6 ft), in some cases more than 8 meters (≈26 ft). Several studies of SARS-CoV-2 support aerosol transmission and one study documented virus at a distance of 4 meters (≈13 ft) from the patient. Moreover, evidence suggests infections cannot neatly be separated into the dichotomy of droplet versus airborne transmission routes. Available studies also show that SARS-CoV-2 can be detected in the air, 3 hours after aeroslisation. The weight of combined evidence supports airborne precautions for the occupational health and safety of health workers treating patients with COVID-19.

M a n u s c r i p t Coronaviruses are respiratory pathogens, and the SARS-CoV-2 has been identified in both upper and lower respiratory tract samples from patients [3] . Fever, dry cough, malaise, lethargy, shortness of breath, myalgia are the commonest symptoms [2] . Less common symptoms are headache, productive cough and diarrhoea. Mild cases may present with a common cold like syndrome, whilst severe cases may develop severe acute respiratory distress syndrome and pneumonia. According to the WHO 21% of cases in China have a severe illness [2] . Early estimates of the reproduction number, R0, give values around 2.2 with a mean incubation period of 5.2 days [4] , and a range up to 24 days. A review found the average R0 value for COVID-19 to be up to 3.28 and median value to be around 2.79 [5] . A more recent study estimated the Maximum-Likelihood (ML) value of R0 to be 2.28 for the Diamond Princess cruise ship [6] . All these estimates are similar to R0 estimates for SARS [7] .

In the past epidemics of SARS and MERS Coronavirus, health care workers (HCWs) have paid a heavy toll. During SARS, HCWs comprised 21% of all cases and in some countries, such as Hong Kong, Singapore and Canada, more than half the cases were HCWs, with deaths reported among them [8] . HCW deaths have already been reported with A c c e p t e d M a n u s c r i p t 5 The WHO has issued guidelines for protection of HCWs which recommend contact and droplet precautions for HCWs caring for suspected COVID-19 patients [9]. Specifically, a medical mask is recommended for routine care, while a respirator (airborne precautions) is recommended if HCWs are conducting an aerosol-generating procedure such as endotracheal intubation, bronchoscopy or airway suctioning, along with droplet precautions [9] . Droplet precautions includes the recommendation to maintain spatial separation of 1 m (≈3 ft) with an infected patient, in the belief that large droplets can only spread horizontally to a maximum of 1 m (≈3 ft) [10] . The initial guidelines released by US Centers for Disease Control recommended a more precautionary approach, which includes the use of a mask by the patient (source control [11] ), and airborne precautions for HCWs [12] .

We aimed to review the evidence supporting the rule of 1 m (≈3 ft) spatial separation for droplet precautions in the context of guidelines issued by the World Health Organization (WHO), US Center for Diseases Control (CDC) and European Centre for Disease Prevention and Control (ECDC) for HCWs on respiratory protection for COVID-19.

A systematic review was conducted for evidence of horizontal distance travelled by respiratory droplets, using the PRISMA criteria [13] 

We found 393 papers in the initial search. After reviewing the titles and abstracts 28 papers were selected for full text review. Finally, 10 papers were included in the review (Figure 1 ).

Eight of the ten studies discussed a horizontal trajectory greater than 2 m (≈6 ft) for a range of droplet sizes of less than 60 µm [14] [15] [16] [17] [18] [19] [20] [21] . Seven out of ten studies are based on modelling and A c c e p t e d M a n u s c r i p t 7 among them the extent of horizontal spread of droplets vary between 2 -8 m (≈6 -26 ft) [14] [15] [16] [17] [18] [19] [20] , highlighting differing findings between them, which can be partially attributed to the methodologies employed. Specifically, four of these studies rely on computational fluid dynamics (CFD) approaches that do not account accurately for the multiphase particle-flow interaction physics [14, 15, 18, 20] and three of them model cough as a turbulent jet (continuous ejection with conservation of momentum flux) instead of a turbulent puff (short sudden ejection with conservation of momentum) [15, 18, 20] . The fourth study used Lagrangian modelling for the droplet dispersion and it was acknowledged that this approach assigns a larger momentum to air hence, making it difficult to translate the results into relevant settings for hospital infection control [14] .

Two studies used analogous water tank experiments to validate the mathematical modelling developed and reported distances up to 1.4 m (≈4.5 ft) and 2.5 m (≈8.2 ft) [17, 22] . One of these two studies modelled coughs as turbulent jets (continuous emission) [22] despite contrary evidence showing that the physics of violent exhalations is captured by puffs, sudden high momentum emission of moist and hot air [17] .

Five studies performed experiments on human subjects [14, 17, 19, 21, 23] , four of them generated undisturbed/natural sneezes and coughs, without injestion of fluid or powders by the human subjects [17, 19, 21, 23] . Out of five, two studies used the human subject measurements to develop and validate the mathematical modelling of the droplet dispersal and showed the importance of the exhaled gas cloud of hot and moist air in trapping and extending the range of all droplets [17, 19] . One involved injection of powder in the mouth of the human subject potentially shifting the natural droplet sizes ejected [14] . The other two used still photographs [23] and particle counters [21] and the distance reported among these two vary from 1 -3 m (≈3 -10 ft). Table 1 The interim guidelines for COVID-19 appear to assume only droplet and contact spread and the general risk limit defined for healthcare workers is 1 m (≈3 ft) from the patient [10, 31] .

The transmission of COVID-19 is not well characterised, but is likely to be similar to SARS, which was spread by contact, droplet and airborne routes [32] . Given the presence of SARS-CoV-2 viral loads in both the lower and upper respiratory tract [3] , as well as persistence of the virus in the air 3 hours after aeroslisation [33] , airborne transmission is possible. A recent study showed that seasonal coronaviruses were more commonly emitted in aerosols than in droplets, A c c e p t e d M a n u s c r i p t 9 even through normal tidal breathing [34] . It is timely to review the evidence informing the 1 -2 m (≈3 -6 ft) rule of infection control, which drives guidelines for droplet precautions. Most studies of horizontal transmission of droplets show distances of greater than 2 m (≈6 ft). The maximum distance recorded in the few available studies is 8 m (≈26 feet) [19, 35] . We note, although the studies employed very different methodologies and should be interpreted cautiously, they still confirm that the spatial separation limit of 1 m (≈3 ft) prescribed for droplet precautions, and associated recommendations for staff at ports of entry [10] , are not based on current scientific evidence.

The horizontal distance of droplet spread depends on various factors such as viscoelasticity of the expiration fluid, type of ventilation, velocity of expiration, rate of evaporation and the dynamics of turbulent cloud generated during exhalations, sneezing, or coughing [15, [17] [18] [19] .

The 1 -2 m (≈3 -6 ft) limit is based on very limited epidemiologic and simulated studies of some selected infections [36] . Some studies cite Jennison (1942) [23] as the evidence in support of the 1 -2 m (≈3 -6 ft) risk limit. This study used high speed exposure to capture still photographs of the atomising secretions generated by human sneezing, coughing and talking, imaged very close to the mouth. It was concluded that the distance to which the majority of droplets were expelled is 2 -3 ft (≈1 m) but, no details were provided about how they reached this conclusion. The study acknowledges that the motion picture film used for the experiments was not sensitive enough to capture all the droplets. The lighting technique used inherently selects for the largest sizes of droplets and fluid ligaments, not capturing the rest of the emissions and gas cloud carrying them. The author used still photographs, in which many droplets move out of focus and become unrecordable very quickly, especially using photographic technology from the 1940s. More recent studies have shown the extent of droplet spread to be greater than 2 m (≈6 ft) [16] [17] [18] [19] [20] [21] 35] , and that infection risk exists well beyond the recommended range of spatial separation.

A c c e p t e d M a n u s c r i p t 10 Further, there is no agreement on the definition of "droplet" route of transmission. There is some agreement that particles with diameters less than 5 µm are airborne particles but, there is significant variation in the literature when it comes to the classification of the lower size limit of droplets. Wells (1934) [37] considered 100 µm as the cut-off limit for the droplet route. But, later studies considered a cutoff particle diameter of more than 10 µm to more than 100 µm [14, 15, 20] . The World Health Organization (WHO) employs a cut off limit of 5 µm to differentiate between aerosols (≤5 µm) and droplet (>5 µm) [38] transmission routes. However, even particles with a diameter of more than 10 µm can remain airborne long enough to not fall under the framework of classification of "droplet" route [39] . In addition, the size of a droplet is dynamic and changes within seconds during the transit from the respiratory tract to the environment due to evaporation [39] . A large droplet expelled during coughing or sneezing can become an airborne particle in less than a second [39] and that timescale changes depending on the cloud dynamics of exhalation [17, 19] . Hence, it is not possible to characterize droplet and airborne spread as separate, mutually exclusive modes of transmission and further studies of the risks accounting for combined ambient conditions and patient exhaled cloud are needed.

Indeed, another important consideration is the effect of temperature, relative humidity, ventilation etc. on the extent of droplet spread which has been examined by only a few studies.

To summarise, they have shown that relative humidity plays an important role in the evaporation of the droplets and the distance a droplet can travel. They report that as the relative humidity increases the extent of droplet spread decreases [18, 20] , yet the horizontal range of the cloud propelling the drops was found to increase with increase in relative humidity, due to the role of buoyancy of the exhaled cloud [17] . For droplets less than 20 µm in diameter, local airflow field due to body heat is an important factor in determining the extent of spread since it can lift the droplets upwards into the breathing zone [40] . Studies have also shown that depending on the flow direction and airflow pattern, increasing ventilation rate can effectively A c c e p t e d M a n u s c r i p t 11 reduce the risk of long range airborne transmission [41] . Most patients spend the majority of time in normal breathing and can saturate the room air with airborne particles expelled during breathing. Moreover, despite negative pressure isolation conditions, airflow due to door motion can cause breakdown in isolation conditions and as a result pathogen can escape the room and there is probability of infection spread outside the room [42] . In general recent studies show distances reached by potentially pathogen-laden droplets of a continuum of sizes to be far greater than 2 m (≈6 ft) [16] [17] [18] [19] [20] , therefore the probability of infection well beyond the defined risk limit can be significant. For example, SARS was classified as predominantly transmitted through contact and droplet modes, but, aerosolised transmission well beyond 2 m (≈6 ft) was reported in the Amoy Gardens outbreak [32] .

The ability of countries to respond effectively depends on the safety and confidence of the health workforce, especially in low income countries with low ratios of HCWs per head of population and protective measures are crucial to ensure a functional health workforce. We have previously shown that masks do not have clinical efficacy against respiratory infections [43, 44] , and that intermittent use of respirators (which depends on HCWs to assess their own risk and use the device when they judge they are at risk) is as equally ineffective as mask use [44] . A recent trial confirmed there is no difference between targeted respirator use and surgical mask use, but did not have a control arm and so may have shown equal efficacy or inefficacy [45] . Proven efficacy of a respirator is seen when the device is worn continually during the shift [43] . The SARS-CoV-2 has been found in both upper and lower respiratory tract specimens, often early in the upper and later in the lower respiratory tract [3] , which means it can potentially be dispersed in fine, airborne particles. Influenza studies show that in a busy emergency department or hospital ward, airborne particles with viable virus can persist for hours in the air [46] . A study of SARS-CoV-2 in a hospital in Wuhan found virus at least 4 m (≈13 ft) within a hospital ward, and virus was identified in air samples and on multiple air outlet A c c e p t e d M a n u s c r i p t 12 vents [47] . Other studies have also found SARS-CoV-2 on air vents in a patient room [48] .

Another study found virus in air samples three hours after aersolisation [33] . We have also shown that airborne precautions are more efficacious in protecting HCWs even against infections assumed to be spread by the droplet route [49] . This further supports the conclusion that infections cannot be neatly separated into droplet versus airborne transmission routes, and that it is likely both airborne and large droplets, carried by the respiratory cloud, are emitted close to the patient and further away. In light of the lack of definitive transmission data for SARS-CoV-2 , as well as persistence of the virus in the air 3 hours after aeroslisation [33] , the precautionary principle in the initial CDC guidance was justified. This includes use of a mask by the patient, for which the limited evidence is supportive [11] . Guidelines should be precautionary in ensuring protection of the occupational health and safety of health workers treating COVID-19 [50] . Although the majority of the studies reviewed point towards horizontal spread of more than 2 m (≈6 ft), these results cannot be translated directly to hospital settings, as the studies used varying range of assumptions. The recent data on SARS-CoV-2 in a hospital ward shows a distance travelled by the virus of at least 4 m (≈13 ft), double the assumed safe distance [47] . This review reveals the limited scientific data to inform spatial separation guidelines, and a growing body of evidence that droplet precautions are not appropriate for SARS-CoV-2. Hence, future works on carefully documenting and studying the mechanisms shaping transmission distances are warranted, particularly with experiments over a large number of subjects and a variety of conditions, to update current spatial separation guidelines and the current paradigm of droplet and airborne respiratory transmission routes. M a n u s c r i p t 21 

Novel Coronavirus (2019-nCoV) situation reports

Virological assessment of hospitalized patients with COVID-2019

Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia

The reproductive number of COVID-19 is higher compared to SARS coronavirus

Estimation of the reproductive number of novel coronavirus (COVID-19) and the probable outbreak size on the Diamond Princess cruise ship: A data-driven analysis

Transmission dynamics and control of severe acute respiratory syndrome

Infection prevention and control during health care when novel coronavirus (nCoV) infection is suspected

World Health Organization. Management of ill travellers at points of entryinternational airports, seaports and ground crossings -in the context of COVID-19 outbreak: interim guidance

World Health Organization

Cluster randomised controlled trial to examine medical mask use as source control for people with respiratory illness

Interim Healthcare Infection Prevention and Control Recommendations for Patients Under Investigation for 2019 Novel Coronavirus

Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement

Study on transport characteristics of saliva droplets calm indoor environment

How far droplets can move in indoor environments -revisiting the Wells evaporation-falling curve

Theoretical analysis of the motion and evaporation of exhaled respiratory droplets of mixed composition

Violent expiratory events: On coughing and sneezing

Enhanced spread of expiratory droplets by turbulence in a cough jet

Evaporation and dispersion of respiratory droplets from coughing. Indoor Air

Quantity, size distribution, and characteristics of coughgenerated aerosol produced by patients with an upper respiratory tract infection

Human cough as a two-stage jet and its role in particle transport

Atomizing of mouth and nose secretions into the air as revealed by high-speed photography. Aerobiology. 17th ed. American Assn. for the Advancement of Science

Interim Domestic Guidance on the Use of Respirators to Prevent Transmission of SARS

Infection prevention and control during health care for probable or confirmed cases of Middle East respiratory syndrome coronavirus (MERS-CoV) infection: interim guidance: updated

Interim Infection Prevention and Control Recommendations for Hospitalized Patients with Middle East Respiratory Syndrome Coronavirus (MERS-CoV)

Rapid risk assessment: Severe respiratory disease associated with Middle East respiratory syndrome coronavirus (MERS-CoV)

CDC updates guidance on PPE for health care personnel; COVID-19 declared a pandemic

Interim Infection Prevention and Control Recommendations for Patients with Suspected or Confirmed Coronavirus Disease 2019 (COVID-19) in Healthcare Settings

Infection prevention and control and preparedness for COVID-19 in healthcare settings -second update

Evidence of Airborne Transmission of the Severe Acute Respiratory Syndrome Virus

Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1

Respiratory virus shedding in exhaled breath and efficacy of face masks

Turbulent Gas Clouds and Respiratory Pathogen Emissions

Guideline for Isolation Precautions: Preventing Transmission of Infectious Agents in Health Care Settings

On air-borne infection. II. Droplets and droplet nuclei

World Health Organization. Infection prevention and control of epidemic-and pandemic-prone acute respiratory diseases in health care

Aerosol technology: properties, behavior, and measurement of airborne particles

Thermal effect of human body on cough droplets evaporation and dispersion in an enclosed space

Ventilation control for airborne transmission of human exhaled bioaerosols in buildings

Door-opening motion can potentially lead to a transient breakdown in negative-pressure isolation conditions: The importance of vorticity and buoyancy airflows

A cluster randomized clinical trial comparing fit-tested and non-fit-tested N95 respirators to medical masks to prevent respiratory virus infection in health care workers

A randomized clinical trial of three options for N95 respirators and medical masks in health workers

N95 Respirators vs Medical Masks for Preventing Influenza Among Health Care Personnel

Measurement of Airborne Influenza Virus in a Hospital Emergency Department

Aerosol and Surface Distribution of Severe Acute Respiratory Syndrome Coronavirus 2 in Hospital Wards

Surface Environmental, and Personal Protective Equipment Contamination by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) From a Symptomatic Patient

The efficacy of medical masks and respirators against respiratory infection in healthcare workers. Influenza Other Respi Viruses

Uncertainty, risk analysis and change for Ebola personal protective equipment guidelines

M a n u s c r i p t 23 Wei and Li