Transparency data

SAGE 49 minutes: Coronavirus (COVID-19) response, 30 July 2020

Published 11 December 2020

Forty-ninth SAGE meeting on COVID-19, 30 July 2020.

Held via Video Teleconference.

Summary

1. In England, R is estimated at 0.8 to 1.0, with a daily growth rate of -4% to -1%. However, since estimates of R rely on lagged data and because of increases in other ‘lead’ indicators, R is likely to be above 1 across England. Due to considerable heterogeneity, we think that R is above 1 in certain locations.

2. During the application of NPIs nationally, there were substantial support packages available. SAGE advised that local areas of intervention will require similar – and probably more – support to facilitate adherence, particularly in disadvantaged communities.

3. There is more likely to be a positive response to interventions from the public if the reasons behind changes are fully explained and understood. Clear communications and engagement are required to avoid interventions being seen as arbitrary or discriminatory. They should emphasise care rather than punishment and involve co-creation with and delivery through trusted community voices.

4. SAGE again recommended that the proportion of confirmed cases which can and cannot be linked to known clusters should be monitored. This is an important indicator of transmission and of the effectiveness of the test and trace system.

Situation update

5. The case isolation period has been extended from 7 to 10 days from symptom onset, or from a positive test for asymptomatic cases.

6. SAGE approved R and growth rate estimates. The latest estimate of R for the UK is 0.8 to 0.9, while the daily growth rate estimate is -4% to -1%. These estimates mask wide variations across the country of case numbers and trends.

7. In England, R is estimated at 0.8 to 1.0, with a daily growth rate of -4% to -1%. However, these estimates of R rely on lagged data (for example number of deaths) and increasing proportion of positive tests are being reported. When lead indicators are used, R is likely to be above 1.

8. As previously, SPI-M does not have confidence that most regional R estimates are sufficiently robust to inform decisions, since they are based on low numbers and/or are dominated by clustered outbreaks. Heterogeneity means that it is likely that R is above 1 in certain regions.

9. SPI-M’s short-term forecasts will start to look again at hospitalisations and also at proportion of positive tests, in addition to deaths.

10. PHE work on secondary attack rates in households found little variation in transmission or susceptibility by age. Estimates of the proportion of asymptomatic to symptomatic cases continue to vary widely, although more data are becoming available on this. The household study suggested a near 50% risk of infection in households, and that about 75% were symptomatic.

11. Discussions are underway around the design of a CoMIX study focused on BAME groups.

Actions

• NERVTAG to review latest evidence on proportion of cases which are asymptomatic and pauci-symptomatic

Local interventions

12. SAGE endorsed SPI-B’s consensus paper in response to the MHCLG commission, subject to minor amendments.

13. The term ‘lockdown’ connotes separation and punishment, and does not empower nor capture the responsibility of individuals to make good choices. ‘Areas of intervention’, from the government COVID-19 Contain Framework, is better terminology; this would be even better if it could describe support or care.

14. During the application of NPIs nationally, there have been substantial support packages available. SAGE advised that local areas of intervention will require similar – and probably more – support to facilitate adherence, particularly in disadvantaged communities, as they will be experiencing restrictions for a second time.

15. There is a need to motivate people to engage and adhere to behaviours they may perceive to have failed previously. Young males currently have the poorest adherence with NPIs.

16. There is more likely to be a positive response to interventions from the public if the reasons behind changes are fully explained and understood. Clear communications are required to avoid interventions being seen as arbitrary or discriminatory. Communications should emphasise care rather than punishment and be culturally appropriate. They should be co-created and delivered with trusted community voices to maximise engagement and make it clear that interventions are for the benefit of, and endorsed by, the community, rather than the result of external enforcement.

17. Measures may need to be applied multiple times in some areas. Co-production of national guidelines that are attuned to local communities will help retain public support. This is best achieved by community engagement.

18. Care is also needed to avoid discrepancies between national and local government messaging in areas of intervention.

19. Risk of non-adherence can potentially be mitigated by ensuring interventions are implemented at the correct spatial scale. More work is needed to understand this.

Actions

• James Rubin to make suggested amendments to SPI-B consensus paper on local interventions by 3 August; SAGE secretariat to circulate paper to MHCLG, PHE, JBC, DHSC, CO and DAs for policy consideration, with a covering note explaining its relevance and importance, by 3 August
• SPI-M, SPI-B, CSA MHCLG, JBC and DAs to meet to consider how to identify the most appropriate geographic definitions and scales for interventions, by 6 August

Reasonable worst-case scenario (RWCS)

20. SAGE endorsed the RWCS from SPI-M and the accompanying narrative, subject to minor amendments to the latter around public behaviours and the impact of a cold winter.

21. SAGE strongly recommended the accompanying narrative – which covers a range of possible challenging scenarios, of which the RWCS is just one – be read alongside the RWCS. These scenarios are not predictions or forecasts.

22. The RWCS has been developed by SPI-M for government planning purposes and covers the whole of the UK.

23. There is variance in the models, including some which now separate infections in care homes from community infections (although care homes are modelled as a single entity rather than individual homes). In future, it would be helpful to model hospital infections separately as well.

24. A small shift in age distribution of infection has a very significant impact on modelled deaths. SAGE once again highlighted the importance of measures to protect older people.

25. The accompanying narrative highlights the difficulties which would be created if R rapidly increases past 1 with high regional variation. While it might be possible to suppress local outbreaks when R is around 1, it will be much more difficult when R has risen above 1.

26. Testing is critical. The initial phases of exponential growth will appear as a gradual increase in cases, so any such increase must be gravely monitored.

27. It would be highly undesirable for the NHS to need to make the same surge capacity available, as it did in March, at the expense of other activities, as this has significant consequences for healthcare provision.

Actions:

• James Rubin to provide input to SPI-M narrative paper; SAGE secretariat to circulate reasonable worst case scenario and revised narrative by 3 August
• NHS Medical Director to advise SPI-M chairs on length of stay and proportion of patients being treated in ITUs by 3 August

Joint Biosecurity Centre (JBC) trigger points

28. SAGE again recommended that the proportion of confirmed cases which can and cannot be linked to known clusters should be monitored. This is an important indicator of transmission and of the effectiveness of the test and trace system.

29. SAGE cautioned that data from the KCL ZOE app may not be a reliable indicator, owing to the high levels of ‘noise’ (very low proportion of suspected cases confirmed when tested). This needs to be monitored and evaluated.

30. SAGE agreed the value of making public the indicators and triggers for moving between different national alert levels. Transparency around these is very important to gain public trust. It is also important to test public understanding of triggers.

31. SAGE noted that care is needed when setting NPIs in relation to alert levels. If a change in alert level immediately led to a lifting of NPIs (such as mask wearing or social distancing) there could be a yoyo effect, with a subsequent rise in cases and a requirement to switch back to a higher alert level.

32. Differentiation is required between NPIs which constrain activities (such as closing pubs) and those that rely on individual behaviours (such as mask wearing).

Action:

• SAGE participants to send any further comments on JBC triggers paper to James Benford by 3 August

List of actions

• NERVTAG to review latest evidence on proportion of cases which are asymptomatic and pauci-symptomatic
• James Rubin to make suggested amendments to SPI-B consensus paper on local interventions by 3 August; SAGE secretariat to circulate paper to MHCLG, PHE, JBC, DHSC, CO and DAs for policy consideration, with a covering note explaining its relevance and importance, by 3 August
• SPI-M, SPI-B, CSA MHCLG, JBC and DAs to meet to consider how to identify the most appropriate geographic definitions and scales for interventions, by 6 August
• James Rubin to provide input to SPI-M narrative paper; SAGE secretariat to circulate reasonable worst case scenario and revised narrative by 3 August
• NHS Medical Director to advise SPI-M chairs on length of stay and proportion of patients being treated in ITUs by 3 August
• SAGE participants to send any further comments on JBC triggers paper to James Benford by 3 August

Attendees

Scientific experts

• Patrick Vallance (GCSA)
• Alan Penn (CSA MHCLG)
• Andrew Curran (CSA HSE)
• Angela McLean (CSA MoD)
• Brooke Rogers (KCL)
• Calum Semple (Liverpool)
• Catherine Noakes (Leeds)
• Charlotte Watts (CSA DFID)
• Graham Medley (LSHTM)
• Ian Boyd (St Andrews)
• Ian Young (CMO Northern Ireland)
• James Benford (JBC)
• James Rubin (KCL)
• Jeremy Farrar (Wellcome)
• Jim McMenamin (Health Protection Scotland)
• John Aston (CSA HO)
• John Edmunds (LSHTM)
• Jonathan Van Tam (dCMO)
• Maria Zambon (PHE)
• Mark Walport (UKRI)
• Michael Parker (Oxford)
• Nicola Steedman (dCMO Scotland)
• Peter Horby (Oxford)
• Rob Orford (Health CSA Wales)
• Robin Grimes (CSA MOD Nuclear)
• Steve Powis (NHS)
• Venki Ramakrishnan (Royal Society)
• Wendy Barclay (Imperial)
• Yvonne Doyle (PHE)

Observers

• Vanessa MacDougall (HMT)

Secretariat

• Simon Whitfield
• Stuart Wainwright
• Tim Cowen

Total: 57

1 scientific expert, 8 observers and government officials and 15 Secretariat members redacted.