Scottish Government COVID-19 Advisory Group minutes: 21 January 2021

Attendees and apologies

Attendees

Advisory Group members:

• Andrew Morris
• Dave Caesar
• David Crossman
• Tom Evans
• Roger Halliday
• Nick Hopkins
• Jim McMenamin
• Jill Pell
• Stephen Reicher
• Jacqui Reilly
• Sheila Rowan
• Aziz Sheikh
• Gregor Smith
• Devi Sridhar
• Carol Tannahill
• Mark Woolhouse
• Angela Leitch
• Chris Robertson

Invited attendees:

• Cat Carver
• Gabe Docherty
• Matt Holden
• Andrew Rambaut

SG:

• Andrew Bracewell
• Rebekah Carton
• Richard Foggo
• Gary Gillespie
• Daniel Kleinberg
• Jason Leitch
• Audrey MacDougall,
• Alisdair McIntosh
• Arlene Reynolds
• Elizabeth Sadler
• Craig Thomson 

Secretariat: [REDACTED]

Items and actions

Welcome 

The chair welcomed group members, guests and SG observers.  

Minutes

The minutes of the previous meeting had been shared with group members for comments by close of play on 22 January. The Chair thanked Cat Carver for drafting the paper on Long Covid, following discussion at the previous meeting. The Chair offered some reflections on the most recent meeting of SAGE, earlier that day (see agenda item 7). 

Action:

• long Covid paper to be finalised and submitted to CMO, subject to any final comments received

New variants

The Vice Chair opened the discussion with a brief history of genomic sequencing of Covid-19 in Scotland, noting that regular changes in the virus had been observed from early days with new variants tracked, then compound variations had arisen. Data was now accumulating that e484k variants can display aspects of immune resistance. It was fortunate that the Kent/B117 variant had the S gene drop-out marker, which meant that this variant with increased transmissibility could be tracked without the need for sequencing, as Kent has meant in effect adjusting to a new virus due to the increased transmissibility. We don’t know for sure how these variants arose – it may be host defence attack, though transmission should be rare in such cases – but there are likely to be more down the line. We need to accept that this is a changing entity and we need to prepare for that.

Andy Rambaut and Matt Holden then spoke about latest development with the new variants. A striking feature of the three Variants of Concern (VOC) was the number of variations at one time – 22 mutations in the Kent variant within 1 month, with problematic mutations alongside those more neutral. The explanation is likely to be a chronically infected individual, possibly immune compromised, with the virus adapting to that host - though it is rare that these individuals transmit. The 681H mutation in the Kent variant is hypothesised to be significant in transmission and has been seen in one example of the ‘Brazilian’ variant, found in Finland.

The other two VOC in South Africa and Amazonia/Brazil are similar, with a large number of mutations including spike mutations. There is some concern that these mutations may be a new fitness peak, which is not usually possible through sequential mutation – likely to be caused by volume of cases. One mutation common to all three is not yet understood but may be important and there is some sign of the Kent variant acquiring the same mutation as the South African and Amazonian variants. There had been over 20 UK cases linked to South Africa, with four variants. RNA technology is well suited to produce vaccines targeted at variants, though there may be a need for some flexibility in licensing to allow these to be deployed without too much delay.

There is currently no evidence that the South African or Brazilian variants are more infectious, though the Brazil variant in particular appears to be spreading rapidly in an area where it is thought there was high existing seroprevalence. The COVID-19 Genomics UK Consortium (COG) are using genome sequencing to track variants. While Kent is trackable by S drop out a network of sequencing is important to identify/track others and link to Test & Protect. It was noted that sequencing is a tool that is valued greatly by local Public Health teams.

The group noted the desirability of real time sequencing and the need to check for immune and vaccine escape – were there lessons can we learn from other viruses e.g. flu? It was noted that there may be future mutations as the virus reacts to vaccination. There had been spread to mustelids, and possible lessons from that, but there would be concern if the virus was established in other animal populations. Carnivora appear susceptible to the virus but mink farms were a special case and that variant appears to have been eradicated. Domestic carnivores are hopefully in low enough density for problems to be unlikely.

The UK has world leading genomics capability (around half of all sequencing) but there is global input from US, Denmark & Netherlands (due to mink), less in the rest of Europe and Brazil sequences quite a lot. Japan sequences every quarantine case but variants are likely to be high prevalence to be detected there, though the Brazilian variant had initially been identified in a Japanese quarantine facility. Scale matters - UK has a chance of catching a new variant before it gets too common but elsewhere a new variant is likely to be already common before it’s detected. There are daily import of genomics data from international databases. The problem is knowing which variants may be of concern, though rapid growth is obviously a danger sign. 

Partial immunity is a concern as it puts selection pressure on virus. NERVTAG initial assessment of impact of new variant on children may have been affected by schools being open – current view is no evidence of increased severity for children. Still some uncertainty about Brazilian seroprevalence numbers based on blood donors as that may have biased the data, though equally no current evidence that it’s being driven by reinfection.

Sample processing and logistics is the biggest time limiting step. [REDACTED] There is an urgent issue with getting sequencing data [REDACTED] – equity of access for Scotland important. Priorities for the future are to increase coverage of sequencing and link to epidemiological data to help us understand what we’re seeing. Speed of sequencing also important – need to get results in days/no more than a week to make most of links to epi data – data flows are important. Improving the access and tracking of samples is the key to improving the speed of sequencing process. The ability to sequence all incoming travellers in quarantine could be a helpful way of detecting significant variants. The linking of real time sequencing to epidemiology is of huge importance across public health.

Action:

• testing sub-group to look at impact of new variants and whether technology may be able to help speeding up sequencing. Look to improve flow of data [REDACTED] and report back to next meeting

Travel and borders

The SAGE paper on travel circulated earlier that day would add to the Group’s responses to the questions posed. The aim was to discuss the issues now and come to composite position over the next week, with a paper to be agreed at the next meeting. 

The key questions considered in the discussion were:

• the question for the advisory group is what measures will help protect Scotland and get numbers as low as possible? What did we learn from wave 1 that would help achieve this?
• border control is needed when prevalence is low, to avoid importation causing outbreaks. It is also needed at other times to prevent new harmful variants entering the country
• anything less than total closure of borders (without other measures) only delays importation but total closure is not a sustainable option as new variants will arise continually. [REDACTED]
• no country has supressed the virus without harsh travel restrictions. What’s needed is an enduring strategy – we will need to manage borders until we have a cure. Effective control needs a multi-layered system, which can be adjusted in light of progress with vaccination and any new variants of concern. There is a need to ensure that any relaxation is not too early, which would permit the virus to take off again
• we should look to learn from countries with effective border controls in place, including Australia and New Zealand but a multi-layered system is likely to comprise pre-departure testing, screening of new arrivals and more effective isolation/quarantine measures
• in addition to new measures, there is a need to improve performance of current systems and improve data e.g. on compliance with isolation. We also need more data on test & release effectiveness. Living with the virus means living with some restrictions for the foreseeable future but Covid not as transmissible as flu and is controllable with NPIs, though we need to look at effectiveness in light of the new variant.
• probably inevitable that there will be vaccine escape but it is possible to update vaccines. [REDACTED] RNA vaccines in particular should be easy to update – licensing needs to respond flexibly. Maximising vaccine coverage for the current strain and combining with very tight Test & Protect and effective frequent sequencing will enable us to develop an ongoing feedback loop for mutating strains and enable an effective response
• need to be conscious of inequalities and what approaches provide the most support and protection for those who need it most, including in relation to non-Covid harms

[REDACTED]

Preparation for deep-dive

The session was introduced by CMO, Gary Gillespie and Richard Foggo who noted the extreme pressures on the NHS currently. This is not sustainable. If pressures exceed NHS capacity the impact on mortality will be severe. How do we avoid that now, avoid a repeat in future, reduce incidence and prevalence, maximise impact across genomics, isolation etc., get to a controllable level and keep it there? How do we prepare early for next winter? 

School return is a priority and we will deploy vaccines at scale but we need a strategy to keep new variants at bay. What are the key factors relevant to policy decisions – what are the factors that we need to know to make decisions? The virus has had a severe impact on the economy and current restrictions further reduce the economy but best scenario for economy longer term is low virus – if we know how to get there we can then model economic impacts and look to recovery. 

The Group discussed these questions. It was suggested that there were a series of phases to consider - actions now to reduce virus to approaching low levels; longer term measures - NPIs, Vaccines etc. - which offer a means to control the virus and might be improved over time; and finally how to manage ‘opening up’ domestically and to the rest of the world. There would be a need to consider the levels system, as there was some evidence that lower levels may not be effective – need to look hard at the lessons from that evidence. 

The group commented positively on the work on scenario planning and the depiction of various harms, not just public health. There remained significant unknowns about the impact of the vaccination programme which made it impossible to predict what will happen beyond three months with any certainty. The vaccine trials covered symptoms and safety but we don’t know how well it will block transmission and don’t know whether vaccination can achieve herd immunity in the light of the more infectious new variant. Human challenge studies will shortly start which will help address effects on natural infection/vaccination and on transmission/viral levels etc. but that will take some time to inform the models. Uptake currently high but vaccines refusal might impact more in future, though indication from Israel were that vaccine hesitancy has reduced as the vaccine has been rolled-out. New mutations were inevitable and may have implications for vaccination. Mitigations would remain a challenge in the light of the new variant and elimination might not be realistic; certainly there is no prospect currently of eradication at a global level. If an annual injection is needed for covid as it is for flu and it takes till autumn to finish round one, then we will effectively then need to immediately start round two.

It was noted that the point of the scenarios is to depict plausible futures in the face of the many unknowns. While the pandemic was the main consideration, there are also other factors e.g. the impact of climate change and travel. Behaviour will be differentiated e.g. on vaccine take-up and there is a need to support compliance and good mental health. Inequalities will continue to be an issue – ethnic minorities, the impact of deprivation, gender and also urban/ rural differences. There are inequalities within the economy – the impact of businesses being open/closed, the ability to work at home or requirement to be in workplaces are not neutral and income and skills biases may entrench existing inequalities. 

A key question is what will the ‘new normal’ look like – how do we reconfigure to live in Covid safe ways e.g. in schools and in workplaces - taking a longer term view opens up the possibility of changing the physical environment in the way that a short term view does not encourage. We now know much more than we did 9 months ago and should work to highlight the science that allows us to optimise measures – e.g. the evidence around testing and behavioural science that will maintain benefits and gains, improve the public health response, avoid optimism bias and promote safer travel domestically and internationally.

Action:

• convene meeting next week to complete preparations for Deep Dive, with input on scenario modelling, which will be developed to draw up rich pictures and develop assumptions – need to comment on plausibility of assumptions and what else we need to know/think about on the science

Subgroup updates

The group noted references to the work of the Testing sub-group during the discussion. The Education and Children sub-group were due to meet on Tuesday to consider advice on ELC and the potential for phased return of particular school groups and blended learning. The Chair of the Nosocomial sub-group would share a note of the meeting the following day.

SAGE update

While the R number for Scotland had decreased, the new variant meant that there was still the risk of pressure even if it was below 1 – incidence & prevalence estimates were substantially higher than testing numbers. Scottish analysis of EAVE data had been commended and group members were working to improve data flows in Scotland.

[REDACTED] There had been good uptake of vaccinations, though there were indications of vaccine hesitancy amongst minority ethnic populations in England and advice against vaccination while pregnant appeared to be causing concern for some women of child-bearing age. There had been targeted anti-vaccine propaganda aimed at care homes and some indication of some coming from outside the UK. The vast majority of those in care homes in Scotland had now been vaccinated but it was important to communicate that vaccination isn’t immediately effective and that NPIs remain important.