Dead air - airborne covid 19 and poorly ventilated buildings
In the early stages of the Covid-19 crisis, there was little official recognition that airborne transmission was a risk. Has that view changed, and what role will building ventilation play when winter approaches?
This article was originally published in issue 34 of Passive House Plus magazine. Want immediate access to all back issues and exclusive extra content? Click here to subscribe for as little as €10, or click here to receive the next issue free of charge
The World Health Organisation (WHO) has long maintained that Covid-19 is not airborne, but its position becomes less clear when subjected to scrutiny. On 28 March, the agency made strong assertions in posts on Facebook and Twitter: “FACT: #COVID19 is NOT airborne,” the WHO account tweeted. A WHO article published the next day was less unequivocal, stating that the virus was “primarily transmitted between people through respiratory droplets and contact routes”, while conceding that airborne spread “may be possible in specific circumstances and settings in which procedures or support treatments that generate aerosols are performed,” such as intubating a Covid patient. The article acknowledged that research existed indicating airborne spread, but cautioned that it had not yet been subject to peer review, and proposed no additional precautions about airborne spread.
Passive House Plus contacted the WHO to ask, in light of the contradiction between its unequivocal Facebook and Twitter posts ruling out airborne Covid-19 on 28 March and its more nuanced position on 29 March, if it would unpublish the social media posts in question. The agency reiterated its position: “Covid-19 is primarily transmitted between people through respiratory droplets (for instance produced when a sick person coughs) and close contact with sick people or contaminated surfaces. These droplets are too heavy to hang in the air. They quickly fall on floors or surfaces. This is why WHO recommends that everyone continue to follow basic protective measures against Covid-19.”
“In health settings during certain procedures, it is possible for the virus to be airborne under some conditions. This is why WHO recommends precautions for health workers.”
The WHO response referenced a WHO guidance document on preventing Covid transmission in healthcare settings. In a section on contact and droplet precautions, the guidance states that “patients should be placed in adequately ventilated single rooms. For general ward rooms with natural ventilation, adequate ventilation is considered to be 60 l/s [litres per second] per patient”. For airborne precaution rooms, the target is set at 160 l/s per patient.
There is an apparent contradiction in the ventilation target for wards. If the WHO’s position is that Covid quickly falls on the floor or surfaces, and that one metre of distancing between beds is sufficient mitigation, then why specify a ventilation rate – and a high rate of 60/l/s/patient at that?
Linsey Marr, the Charles P. Lunsford professor of civil and environmental engineering at Virginia Tech agrees. “Recommending improved ventilation in buildings implicitly acknowledges that airborne transmission is important. Otherwise, social distancing alone would stop transmission,” says Marr. In responding to Passive House Plus, the WHO revealed plans to imminently publish revised advice on ventilation as part of its Covid-19 infection prevention and control (IPC) guidance in healthcare settings, while also issuing a statement to the magazine on airborne Covid-19. The statement, which is published in full on the Passive House Plus website, outlines the WHO position that there is no evidence of airborne transmission of Covid-19 outside of aerosol generating procedures, and that evidence of the virus in air samples did not demonstrate that the virus could be transmitted in this way. “So far there is no evidence of “transmission” of the virus as an airborne pathogen (such as TB),” the agency said. “This is different from finding the virus in air samples or showing that aerosol particles can be generated from bigger droplets when people cough, sneeze or talk loudly.”
According to Prof Jose L Jimenez, a fellow at the Cooperative Institute for Research in Environmental Sciences, this statement reflects a very narrow, binary definition of “airborne”. “It is like measles or TB, or it is not airborne at all. Makes no sense,” he says. “Why are there only those two possibilities, what is that assumption based on?”
Charles Haas, LD Betz professor of environmental engineering at Drexel University is equally vexed. “The dichotomy between ‘aerosol’ and ‘droplet’ is ancient and outmoded,” he says. “The problem with WHO is that initially they conveyed too much of a sense of certitude and they have had difficulty backing off.”
The WHO statement argues that the virus reproduction number in various countries, “does not indicate a typically airborne pattern of transmission. The occurrence of airborne transmission would have resulted in many more cases and even more rapid spread.”
Prof Jimenez calls this assertion “totally bogus”, adding that the reproduction rate “is not an indicator of aerosol transmission,” adding that Anthrax or hantavirus have a reproduction rate of zero, but are only transmitted via aerosols. Jimenez adds that unlike more efficient airborne diseases like measles and TB, Covid-19 is “opportunistic airborne”, and needs crowding, low ventilation rates and duration in order to spread.
Shelly L Miller, professor of mechanical engineering at the University of Colorado Boulder argues that the WHO are “demanding much more rigorous proof of airborne transmission compared to surface/touch transmission. Where has it been definitively proven that contact with contaminated surfaces got people sick? Yes, we find the virus on surfaces but also in the air.”
Replying to Passive House Plus in May, Prof Marr made a similar point: “I don’t think we can rule out airborne transmission of Covid-19, especially in close contact situations and in spaces with poor ventilation for the number of people present,” said Marr. “In fact, there is mounting evidence that airborne transmission is occurring. There isn’t actually any direct proof of transmission by large droplets either.”
The sense that the WHO and the medical community in general are not engaging sufficiently with aerosol scientists is palpable. The WHO told Passive House Plus it “convenes a group of more than 30 international experts in the fields of infectious diseases, epidemiology, public health and infection prevention and control” to regularly review the evidence. “The set of people at the table […] is narrow,” says Prof Haas. “Where are experts in engineering, ventilation, risk assessment? They have always had too narrow of a net that they cast for expertise.”
If the emerging consensus among aerosol scientists is right, then airborne transmission of Covid-19 is a significant factor – in particular in crowded, poorly ventilated buildings. The issue, then, of how we ventilate our buildings comes into sharp relief.
The aforementioned 60/l/s and 160/l/s per patient targets for hospital wards and airborne precaution rooms – expressed as minimum hourly averages – comes from a 2009 WHO document, titled ‘Natural ventilation for infection control in health-care settings’. It notes that this guidance, “only applies to new health-care facilities and major renovations,” an apparently implicit recognition that existing hospitals will have sub-optimal ventilation rates in airborne precaution rooms, and consequently higher risks of infection spread.
The guidance also states: “The design must take into account fluctuations in ventilation rate”, and adds that “when natural ventilation alone cannot satisfy the recommended ventilation requirements, alternative ventilation systems, such as hybrid (mixed-mode) natural ventilation should be considered, and then if that is not enough, mechanical ventilation should be used.”
Consulting ventilation engineer Mich Swainson points out that the ventilation rate required to maintain good indoor air quality in non-domestic buildings is typically taken as a minimum of 10 l/s, as per UK building regulations, CIBSE Guide A, and the Department of Health document, ‘HTM 03-01 A’. However, HTM 03-01 notes that a general ward or single room should have an air change rate of six air changes per hour (ACH).
“Relating 6 ACH to a dwelling, purge ventilation in a room is achieved by opening the windows, this aims to achieve a minimum of 4 ACH,” says Swainson. “Purge ventilation is deemed to be intermittent. The WHO and HTM 03-01 respectively indicate that a ventilation rate of 60 l/s/patient and 6 ACH is required, 24/7. Unless a natural ventilation system had been specifically designed to achieve such rates, in a single room this would require a significant window area to be open 24/7, and in a general ward could only be achieved with cross ventilation through large openings. In cool, windy or wet weather, the use of windows would be impractical to achieve such high flows, without significant impact on the thermal comfort of the occupants.”
As Dr Chris Iddon, chair of the CIBSE natural ventilation group, explains, the ability to deliver high levels of natural ventilation through window openings in hospitals may be considerably reduced from the design intent by window restrictors installed to stop patients escaping. “Window replacement is also an issue,” he adds, “as the tendency is to focus on Part L rather than whether the windows can help achieve the flow rates under Part F.”
Iddon warns that achieving higher ventilation rates may prove trickier in the coming months, given the tendency of occupants to close windows and vents during colder weather. “We’ll be coming to wintertime. We need to think about what amount of air we can reasonably deliver whilst not unduly increasing any transmission risk.”
“If you’re in your household bubble I think you have to think less about ventilation than in a public building,” says Iddon, reasoning that surface spread and large droplets will already make virus transmission highly likely in households. Iddon is more concerned about public buildings. “A lot of those buildings are probably well ventilated, but a lot aren’t. I’m not surprised we’ve seen large viral transmission events in halls, given how high up and inaccessible the windows tend to be.”
Iddon said CIBSE’s position has been to err on the side of caution, and try to maximise ventilation as much as reasonably possible, but to revisit this as we move in to autumn. “In a lot of circumstances where the ventilation is poorer than it should be, there seems to be an increased risk of infection,” he says.
But can natural ventilation be relied upon to satisfy a specific ventilation requirement? Useful insight on this point can be found in a 2015 review of 168 academic papers published since the year 2000 on ventilation and health, and which subjected 48 of these to more detailed analysis. It found that higher ventilation rates were generally associated with reduced adverse health outcomes, and that acute health symptoms tended to be lower in mechanically ventilated buildings. The picture on mechanical ventilation was mixed with regard to asthma and allergy symptoms, but the findings on natural ventilation bear repeating in full:
“Ventilation rates in naturally ventilated buildings can only be characterized with a high level of uncertainty because they depend on outdoor conditions and activities and on the behaviour of building occupants. Instantaneous (spot) measurements or even weekly averages may not be able to capture and represent the true variability and may not be representative of actual rates. Consequently, exposures that are related to ventilation may not be properly estimated and may not reflect the actual exposures. They may simply be too low or too high compared with the actual levels.
Barring feats of ventilation engineering, it would appear misguided therefore to expect natural ventilation to reliably meet a minimum ventilation rate, except when backed up with the safeguard of mechanical systems in mixed mode designs.
There is also evidence that occupants tend to close or permanently block natural vents. A Scottish study of 40 naturally ventilated airtight homes found that 63% of vents in bedrooms and 63% of vents in living rooms were stated to be “closed and never opened”. Meanwhile, a 2010 study of natural ventilation in 22 airtight English homes found that 60% of vents were closed, while a 2005 BRE study of 37 homes found that trickle vents were fully open in only four of the study homes, and fully closed in 13 homes.
The Scottish study, which was conducted from January to March 2014, calculated that the average per person overall ventilation rate was 3.1 litres per second per person and the range was from 0.9 l/s/p to 6.0 l/s/p. The BRE study involved monitoring ventilation rates in 37 naturally ventilated, leaky homes during winters and summers over two years. The measured air change rates were considerably lower in winter: 0.44 air changes per hour in winter, versus 0.62 air changes per hour in summer. The main reason given was increased opening of windows and trickle vents in summer months. Homes where windows were open ‘most or all of the time’ had significantly higher ventilation rates than other homes, both in winter and summer. House type was also an apparent factor: flats had lower ventilation rates than other house types.
Bearing all of this in mind, there is reason to fear that the conditions will be ripe for the airborne spread of Covid-19 as winter approaches, due to people spending more time indoors and the near ubiquitous reliance on natural ventilation via background ventilators and window opening – ventilation strategies that many people are inclined not to use during winter.
To view the CIBSE guidance on emerging from lockdown, including ventilation guidance, visit: www.cibse.org/coronavirus-covid-19/emerging-from-lockdown.