Covid-19: Science and evidence on environmental transmission and modelling
Our three speakers gave a fascinating insight into how Covid-19 challenges the science and evidence around environmental transmission and modelling. All three are active participants in Sage (Scientific Advisory Group for Emergencies) and co-chairs of the Sage Environmental Modelling Group. The seminar was held on March 16, 2021, and was chaired by Mary Trainor of the Hazards Forum.
Andrew Curran, director of research at the Health and Safety Executive (HSE) presented the challenges of defining a ‘Covid-secure’ workplace and the need to understand the hierarchy of control.
Catherine Noakes OBE, professor of environmental engineering for buildings at the University of Leeds, explained how measuring airborne Covid-19 was like ‘finding a needle in a haystack’, which was dependent on multiple variables, themselves hard to measure.
Harry Rutter, professor of global public health at the University of Bath, explained how Covid-19 needed to be recognised as a ‘messy complex system level problem’ that could not be solved with a single intervention.
Andrew Curran summary
‘The pace at which we’ve learnt is incredible’. That was Andrew’s verdict on the journey to try to understand the transmission of Covid-19.
A Medline search, going back to 1917, just before the flu pandemic, shows you can find 134,000 papers published on influenza in the period from 1917 to 2020. A search for Sars-Cov-2 from the beginning of 2007 to the present day shows there are already 110,000 papers published.
This huge amount of information created risks and benefits. Andrew warned that a loss of rigour and a move to group think – with everyone ‘jumping into a particular space’– would have an impact on guidance, which in turn would have an impact on everything from cancer care to delivering net zero. However, the HSE had learnt from the challenge, he said, and the rapid evidence review was synthesising evidence very fast.
Challenges in assessing what was a ‘Covid-safe’ working environment had been very big but were built on the ‘firm foundations’ of the Health and Safety at Work Act 1974, which states that whoever creates the risk is responsible for it.
Heirarchy of control
Andrew explained how the hierarchy of control was proving useful to give defence in depth, starting with elimination of the threat of transmission, and working down through substitution, engineering controls, admin controls and finally PPE. He said everyone could use the process to identify simple practical solutions to reduce routes to transmission in work spaces.
Lateral flow tests have been mistakenly identified as elimination methods, whereas they are lower down the hierarchy as admin controls.
A key message was that all areas of the workplace and the working day should be considered, in the broadest sense. He said: ‘We need to think about more than the workplace itself. We need to think about canteens and smoking areas. We need people to think about what they are doing from the moment they leave the house to when they get home – not just when they are on a production line.’
ONS data shows a 90% rate for guidance being in place in workplaces, but non-related risk factors – such as canteens and changing areas – are where transmissions are still taking place.
He talked about the work of the Partnership for Research in Occupational, Transport and Environmental COVID transmission (PROTECT). Seventy researchers are working on a national core study, attempting to answer the big questions about how the virus moves around, to help policy makers. They’re taking evidence statements at a personal, workplace and national level.
Cath Noakes OBE summary
‘Hamsters don’t rub their eyes in the same way as people’ was one way Cath Noakes expressed the difficulties in understanding, measuring and mitigating airborne transmission of Covid-19. She explained how lab-based studies with hamsters, though useful, could not replicate the diversity of real environments.
‘The data is hard to unpick’, she said, going on to explain how there were many unknowns around how the behaviour of airborne Covid-19 particles and how likely different people were to spread them.
For example, it’s known that as singing or speaking gets louder, more particles are released. But, she said: ‘Some people breathe out far more particles than others and there’s a very big range in that… we can’t identify who’.
And the viral load of the particles isn’t predictable either, she said: ‘It is extremely hard to measure the amount of pathogens in aerosol. For influenza, there’s a range over four orders of magnitude, so some people produce almost nothing and other people produce incredibly high amounts – and we think it’s the same for COVID.’
An expert in airborne transmission, Cath set out the risk factors as:
- Environmental conditions
Though transmission can occur anywhere, Cath picked out cramped conditions where people worked with poor or no ventilation as the most vulnerable to super spreading events. She stressed the litres of air per second per person was an important measure, rather than just the air change rate. For example, huge sports halls being used as vaccination centres posed less of a risk of a transmission than a ventilated office environment.
She illustrated the behaviour of particles – how bigger ones might pack more of a punch and how smaller ones would head off around the room and around protective screens.
‘We can unpick what’s in the environment,’ she said. ‘It is stable in the air and on surfaces but in most sampling studies have been in hospitals – again we find it on touch sites, we find some in the air, it’s hard to accurately measure – we are learning how to sample from the air.’
When it comes to sampling in the community, she said: ‘It’s a bit like looking for a needle in a haystack … you will only find it in the air if somebody infected is in that space in that time and was emitting aerosol or was very shortly beforehand, If you wait any longer, it’s gone.’
She could conclude:
- close-range contacts carry the most virus
- far-field aerosol is likely to matter for longer duration exposure
- surfaces matter when you share a space with an infector, not when you pick something up in shop
But we need more evidence to understand the importance:
- the variation in viral load
- size of aerosols that contain virus
- dose-response and how it changes with route
- the impact of different mitigation strategies
Harry Rutter summary
‘If you wanted to find a messy, complex system level problem, you really couldn’t do any better than Covid-19,’ Harry explained.
He visualised the routes of transmission, to demonstrate their complexity, while stressing that they all needed to be considered. He asked how we mitigate transmission, as this would vary by context. Face masks were an example of an item that has different uses – it is a source control stopping the virus being spread, while also a protective measure for other people.
No silver bullet
He used his experience of working in obesity to talk about the danger of looking at solutions in isolation. He likened it to a wall of sandbags, each representing single approaches against disease – be it obesity or Covid-19 – that hold back the flood water when used together.
‘Any of these big challenges, you need the wall of sandbags,’ he said.
Although people know there’s no silver bullet for obesity, they look at the wall of sandbags and take out one to see what it will do in isolation – focusing just on school meals to solve the problem of childhood obesity for example.
‘Doing this we are setting ourselves up for failure… We pull out a simple aspect of messy complex problems. People pull out the single aspect of the problem because it’s so hard to think about how they get integrated in this way.’
Citing the ‘high risk strategy’ work of professor of epidemiology Geoffrey Rose, he stressed the need to think in terms of populations rather than pursuing interventions that were known to be effective but would benefit relatively few individuals.
To demonstrate the complexity of Covid-19, Harry contrasted the Saturn 5 Rocket with the Afghan War – the former is complicated, while the latter is complex.
With the rocket, if you know how to assemble and operate millions of parts, you can send three brave people to the moon, and know, within a few metres, where to collect the astronauts when they land.
The Afghan War however, is a different category of problem; however much one knows about insurgents or narcotics or popular support or tribal governance, there’s no one thing you can do to fix it …. it has interdependencies, it has feedback loops, it has emergent properties, it has adaptation.
And like the war, Harry argued scientists must accept Covid-19 is complex, not complicated. He stressed the need to accept that data would be flawed and incomplete, and to be transparent about that and work with others ‘to come up with a pragmatic response’.
‘I think we just have to accept that uncertainty is unavoidable when we’re dealing with a messy, complex problem like this …. Ignoring complexity does not make it go away, but sometimes appreciating it, understanding it, embracing it can help us to see through the fog. And I think perhaps most importantly of all, as scientists, we know that we will all make mistakes and we need to be humble about that.’
In closing questions, the panel was asked what could be done better at the start of another pandemic:
- ‘Spend more time thinking about it rather than jumping in.’ – Andrew
- ‘Don’t assume we know how something transmits based on what we’ve seen before … with this one that we assumed everything was droplets and surfaces. And it turns out that it’s not. – Cath
- ‘The single most important intervention is to understand that there is no single most important intervention. This is a messy, complex problem.’’ – Harry.
Though a tremendous amount has been done, at great pace, there is still a tremendous amount of work to be done.