Project news

Contamination of workplaces with SARS-CoV-2 virus

We have searched through the scientific literature to find out what’s known about contamination of workplaces with the SARS-CoV-2 virus. This is important information for us so that we can compare our measurements with those made by other scientists, and to help us calibrate the model we are building to estimate the risks of infection in different situations.

We found 35 articles that had useful data. These mostly describe measurements made in hospitals with a small number of datasets from public transport settings. The studies were carried out in a diverse group of countries including China, UK, Italy, Spain, USA, Singapore and Iran. The contamination of the air and surfaces were assessed, and in those studies that had data for both it was clear that they were linked, i.e. higher air concentrations were generally associated with higher virus surface loads. We interpret this as suggesting that the surfaces become contaminated by fine droplets from the air.

Typically, around 6% of air and surface samples in hospitals were positive for the SARS-CoV-2 virus, although there was a wide range of results from the different locations. However, most of the studies did not report their results in terms of virus concentrations, which complicates that interpretation of the data. Our best estimate of typical air concentrations in healthcare settings is around 0.01 SARS-CoV-2 virus copies per cubic metre of air. This is low, although in some circumstances the air concentration was more than 10,000 virus copies per cubic metre, and at this concentration a worker might inhale around 100,000 virus particles during a working day.  

The standard of reporting in the published studies was poor with it difficult to be sure there was consistency between researchers in the classification of positive samples. There was also little consistency in the measurement methods used. This is a real problem for our understanding of how much virus contamination is present in hospitals and other workplaces. We feel that there should be concerted efforts to standardise the methods used for measuring SARS-CoV-2 and other respiratory viruses in work environments, and that government agencies and the World Health Organisation should take the lead in this initiative.

You can read a full account of our review in a paper we have recently posted on the internet.

Engineering Novel interventions

Sunlight and UVC decontamination

Whilst the US president, Donald Trump’s suggestion that light could be brought “inside the body” was unhelpful, he stimulated renewed attention to the use of light to control SARS-CoV-2 outside the body. This continues a well established line of research, beginning with Downes and Blunt, who in 1877 found that sunlight limited microbial growth. In this post we will show how sunlight and UVC decontamination could be used to slow the spread of COVID-19 in hospital settings in 2021 and beyond.

Sunlight and UVC

Light is measured by it’s frequency (Hz), wavelength (nm) and energy (eV). The electromagnetic spectrum describes the full range of light, from gamma rays to radio waves.

Electromagnetic spectrum (Credit: NASA’s Imagine the Universe)

Sunlight includes visible and ultraviolet light (UVA and UVB). UVC is blocked by Earth’s atmosphere and therefore does not form any part of sunlight reaching us. This is lucky because it would cause severe DNA damage if we were exposed to it. Inside buildings, UVC irradiation of air and surfaces can be achieved through specialised equipment in the absence of anyone that could be exposed to it.

How does sunlight and UVC affect SARS-CoV-2?

Dabish and his colleagues, found that a combination of higher sunlight, temperature and relative humidity sped up the decay rate of the Corona virus (2020). Based on this research a calculator was then developed by the Department of Homeland Security to show how these conditions change the time taken for 50, 90 and 99% of the virus to be inactivated. For example, during a clear mid-summer day in Edinburgh (UV index 5, temperature 20°C and 70% relative humidity), 99% of the virus will be inactivated in 28 minutes. If it was extremely overcast (UV index 0), but otherwise similar weather conditions to above, it would take 6 times longer for the virus to be inactivated. The time for the virus to be inactivated using UVC is much faster. Kitagawa et al., found that UVC (222 nm in wavelength) at 0.1 mW/cm2  resulted in a 99.7% reduction after 30 seconds (2020).

Where do sunlight and UVC fit within the engineering controls being considered by SAGE?

The UK Government advisory committee, the SAGE – Environmental and Modelling group, reviewed the evidence of sunlight and UVC decontamination in their “Transmission of SARS-CoV-2 and Mitigating Measures” as part of their report on the 4th of June, 2020. Evidence was rated on the control efficacy and real-world effectiveness, and their confidence in it. We digitised the data and visualised the results in interactive bubble plots. Bubble size relates to how many of the 14 raters considered the control efficacy/ effectiveness/ confidence to be high or very high. The figure below shows the engineering controls that were considered and how they compare in an overall rating that combines the scores above.

Sunlight and UVC decontamination both show potential as COVID-19 controls – being in the middle of the 10 measures assessed. Some of the most suitable engineering controls are now being routinely applied, including hand wash stations, screens/partitions and provision of fresh air. It makes sense to look down the list to additional interventions, like sunlight and UVC decontamination, that score higher on efficacy, but lower on effectiveness and confidence (i.e. there is a lack of research on their implementation).

Using sunlight to prevent infection in hospital care settings

World Health Organisation guidelines mention the use of sunlight to reduce hospital acquired infections, especially for pathogens that are airborne. Given that COVID-19 is shown to be predominantly transmitted by aerosol in hospital settings (57% in a recent study by Dr Rachel Jones) then it is worthwhile to consider controls that have a focus on this route. The guidance states that in addition to negative air pressure and ventilation of 6-12 air-changes per hour, patients should be in single rooms with sunlight. Glass completely blocks UVB so ideally the window should be open. However, opening a window more than 10 cm is not allowed in British hospitals (Department of Health, 2013). There may be opportunities develop acrylic sheets to be inserted in open windows to allow UVB to pass through.

Using UVC decontamination to prevent infection in hospital care settings

UVC decontamination is seen as a good alternative to ventilation, when the latter cannot be improved (SAGE-EMG, 2020). UVC decontamination can take 20-45 minutes (depending on room size), typically in an unoccupied room. The equipment should only be used by trained staff with risk assessments and controls in place. The effectiveness is dependent on the distance from the equipment to the contaminated air/surfaces and the impact of objects that cause shadowed areas. A promising avenue of research is far UVC (207 to 222 nm), which does not seem to harm mammalian skin (Welch et al., 2018), but further research is needed to prove the safety of using this light in occupied rooms.

Some products on the market have attempted to address the limitations mentioned above. One product is the UVD Robot (image under the blog post title), which takes 10 minutes to decontaminate a typical patient room and is completely autonomous, removing safety concerns for operators and human error in application. Another product is Sodeco’s UVC air purifier, a low noise unit that can be used in occupied rooms and provides the equivalent of 6.5 air changes per hour in a 500 m3 room.

Testing the effectiveness of sunlight and UVC decontamination interventions

We have adapted a model for understanding COVID-19 transmission in hospital care settings to test the effectiveness of sunlight and UVC decontamination. We will determine the probability of infection and the dominant route of transmission (i.e. through contact with surfaces, inhaling contaminated aerosol or intercepting cough droplets) before and after applying the interventions mentioned above. The results will be shared in a peer-reviewed publication and on this blog, so stay tuned!

Mark Cherrie


Impact of Covid interventions on the workforce

Most of the articles on this blog consider the positive and important advantages of particular interventions in controlling our exposure to SARS-CoV-2 and reducing the risk of developing Covid-19. We have discussed the benefits of ventilated headboards in minimising concentrations of airborne virus particles within a hospital ward or ICU area. However, this blog piece looks ahead to 2021 and aims to flag up some of the potential unintended consequences of interventions that have been essential in the course of tackling Covid in 2020 whilst the virus has been on the rise. In the future – hopefully not too late in 2021 – the balance of benefits and risks of these preventive measures may begin to change.

Managing risk is often about balancing population-level benefits with potential disadvantages or costs. And for much of 2020 the dangers of uncontrolled spread of Covid and the potential stress on the healthcare system and the mortality from the disease have rightly been given priority. Population level interventions such as lockdowns and targeted restrictions in opening particular venues (such as pubs and restaurants) or movement around the country have been the focus of much of the debate around public health benefits versus financial and societal costs, including loss of employment through to wider mental health impacts. Government decisions have been difficult and often finely balanced.

More direct interventions and measures within the workplace have been less openly examined in terms of benefits and costs, beyond the financial costs associated with structural changes. Few would argue that perspex barriers to protect supermarket workers are not a ‘good idea’ and that their relatively low cost is worthwhile to protect them and members of the public and help prevent Covid transmission. Effective cleaning regimes in our schools, shops and hospitals are similarly sensible protective measures. But what of other interventions in the workplace?

Perhaps the commonest impact of Covid-19 interventions for most workers has been the requirement to ‘work from home where possible’. Across the country office workers have set up workstations to allow them to work with laptops propped up on books on dining room tables, hunched over poorly-lit screens in places never designed to minimise the ergonomic risks of musculoskeletal or repetitive strain injuries. Pre-Covid, office workers had regular Display Screen Equipment (DSE) assessments to check that their seating, screens and desks were appropriate and not likely to lead to long-term back, arm and hand injury. These mandatory assessments were largely ignored in the early weeks of restrictions in March, before being partially acknowledged with the provision of basic online guidance for home-workers as the weeks turned to months, but it seems likely that many employers have not been able to fulfil their responsibilities to protect their workers from such injuries during much of 2020. Will we have a cohort of workers who have chronic neck, back and arm problems when Covid begins to fade from memory?

There may be other unintended consequences of Covid-control measures. Alcohol-based hand gels are now placed at the entry, exit and many other positions of most workplaces and guidance recommends frequent hand sanitisation with these materials. Before Covid we knew that frequent hand-washing or use of alcohol hand gels could increase the risk for workers of developing occupational irritant contact dermatitis: an often debilitating disease with life-long consequences. It seems plausible that the increased use of these materials over an extended period of time could produce an additional proportion of the workforce with long-term dermatitis. The numbers of extra workers who will suffer from dermatitis and the severity of their disease is unknown, and has tended not to be considered in the risk-benefit decisions of the ‘clean hands’ part of the Covid intervention guidance. Workplace interventions to protect the population from the risk of uncontrolled transmission of Covid, mortality and unsustainable burdens on our health services, have been essential and, taken together, have been a huge public health success in 2020. However, as occupational health experts it is worth our thinking of what, if any, the longer-term unintended consequences may be from the measures we’ve deployed to help overcome the risks. Measuring and tracking these impacts will be important for public and occupational health researchers in 2021 and beyond.

Sean Semple


Virus protection from face shields

Face shields are widely used in healthcare to protect workers from virus contamination in droplets (i.e. large visible drops of mucous) and aerosols (i.e. very small particles that can be inhaled) from patient coughs, but how effective are they?

Research by the US National Institute for Occupational Safety and Health (NIOSH) helps us understand what protection they can provide. The researchers simulated a coughing patient using machinery that generates a droplet spray and placed a breathing mannequin in front of it. They varied the distance between the coughing and the breathing headforms and had either large (8.5 μm) or small diameter (3.4 μm) aerosol produced with each cough. The experiment did not measure the protection from droplets splattering into the face, which we can assume was 100% In the experiments the visor was very good at protecting against the initial larger aerosol exposure, where more than 95% was diverted away by the visor. The face shield also reduced the surface contamination of a respirator worn behind the visor by a similar amount. However, it was less effective for the smaller diameter aerosol where it blocked around 70% of the initial cough aerosol and 76% of the respirator surface contamination.

The visor was much less effective at protecting the worker in the period after the cough, and in the 30 minutes following a cough the aerosol dispersed through the room and the face shield only reduced aerosol inhalation by 23%. This underlines the need to wear the visor in combination with an effective respirator to achieve effective protection. The authors concluded that the:

‘…results show that health care workers can inhale infectious airborne particles while treating a coughing patient. Face shields can substantially 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 more easily to be inhaled. Thus, face shields provide a useful adjunct to respiratory protection for workers caring for patients with respiratory infections. However, they cannot be used as a substitute for respiratory protection when it is needed.


Lindsley WG, Noti JD, Blachere FM, Szalajda JV, Beezhold DH. (2014) Efficacy of face shields against cough aerosol droplets from a cough simulator. Journal of Occupational and Environmental Hygiene; 11: 509-18.

Project news

Starting with health care

The IOM are working for Covid Safe Workplaces, starting with health care.

As part of our project Evaluation of the Effectiveness of Novel Workplace Interventions in Protecting Healthcare Workers from Virus Infection (Covid-HCW), funded by the Chief Scientist’s Office in Scotland, we have implemented a mathematical model of infection risk to health care workers from their work environment. The model incorporates estimated levels of virus in the air and surfaces, and the frequency of workers contacting contaminated air and surfaces to estimate their risk of infection.

We are collecting samples of the SARS-CoV-2 virus in hospital settings, both in the air and on surfaces, along with details of the tasks that various workers in these settings perform to provide data inputs for this model. We are also collecting information about the effectiveness of various novel interventions that could be applied in these settings to test the potential for infection risk reduction in hospitals. While we are currently focusing on the health care sector, we hope to be able to adapt this evaluative model for other workplace settings to help inform risk management decisions.

Blog written by Dr Miranda Loh


Ventilated headboards

Hospital patients with Covid-19 can contaminate the room air and surfaces with the virus. Healthcare workers have had to rely on personal protective equipment (PPE) to try to protect themselves from being infected. One alternative way to reduce the risks for workers is to use a ventilated headboard on the bed.

Ventilated headboard
The prototype ventilated headboard

These headboards are designed to extract air contaminated with Corona virus from behind the patients head. The air is passed through a high efficiency filter before being discharged. The original design was published by the US National Institute for Occupational Safety and Health nearly ten years ago, but the system has still not been commercialised. However, the researchers provided a set of DIY instruction to build the system.

The researchers list a number of advantages for their design, including:

  • Proven design that successfully captured and removed over 99% of airborne infectious-sized aerosol in a laboratory test
  • Cost-effective system, where the cost per isolated patient is much less than traditional airborne infection isolation rooms
  • Healthcare workers operate outside the “hot zone” of infectious aerosol
  • Easy patient access
  • Scalable from one to many units
  • Highly adaptable to fit most sizes of hospital beds
  • Quick and easy installation
  • Easily dismantled for storage

Watch a short video about the system on YouTube.

While the researchers demonstrated the system has a very high effectiveness we believe that it is unlikely to achieve such performance in real healthcare situations. It seems more realistic to assume that a ventilated headboard could reduce SARS-CoV-2 aerosol emissions from an infected patient in hospital by around 90%


Dungi S, Ghia U, Mead K, Gressel M. (2015) Effectiveness of a Local Ventilation/Filtration Intervention for Health-Care Worker Exposure Reduction to Airborne Infection in a Hospital Room. Published by ASHRAE.