Thread by @LawDavF

The paper I tweeted yesterday was not the one to which Channel 4 referred (apologies) but one actually discussed at SAGE (which is why i assumed one mentioned). I have done a bit more research to work out how Prof Riley’s paper fitted into pattern of decision-making. 1/

https://twitter.com/LawDavF/status/1270784406060437510

This is what Channel 4 said.
channel4.com/news/uk-govern…
Professor Riley’s paper warned that is measures were not taken then the UK faced a Covid-19 catastrophe. Channel 4 say they don’t know how it was evaluated and they don’t say why it was written. 2/

Professor Riley, one of the Imperial College team, was not the only modeller urging action. John Edmunds’ group at the London School of Hygiene and Tropical Medicine had been warning of the scale of the coming epidemic since February.3/

At issue was not whether a UK epidemic would be bad but the best responses and when they should be introduced. There was never an intention not to respond.4/

It is possible to trace development of the government’s response to Covid-19 by using materials in the public domain. SAGE has done a good job making its papers available so there is now a reasonably full record of not only its deliberations but the papers it considered./5

An Excel file on SAGE website lists these papers (this page links in to all the published material). (gov.uk/government/gro… Prof Riley’s paper was not tabled at SAGE. As he is a member of the modelling group (SPI-M ) it was presumably discussed by that group./6

The SAGE meeting of 10 March was not the first time that a combination of social distancing measures had been discussed. The first suggestion I have found in the minutes that they might be an effective response to the virus was 25 February. /7
gov.uk/government/pub…

This led to a paper prepared by SPI-M for 26 February that was discussed at the next meeting. /8
assets.publishing.service.gov.uk/government/upl…

This paper describes as 'a political decision whether ‘it is preferable to enact stricter measures at first, lifting them gradually as required, or to start with fewer measures and add further measures as required.’ SAGE ended up pushing the second option./9

THE SAGE minutes for 27 Feb contain a worst case scenario in which ‘80% of the UK population became infected with an overall 1% fatality rate’. So this suggests (as a worst case) that over 50 million people could get infected with about 500,000 dying. /10
assets.publishing.service.gov.uk/government/upl…

It added that modelling suggested that ‘earlier and/or combined interventions will have more significant impact’. They asked SPI-M to ‘develop illustrative scenarios showing the plausible impact of combinations of interventions’. This is when proposition first addressed./11

This led to this paper ./12
assets.publishing.service.gov.uk/government/upl…

Further discussion at meetings on 3 and 5 March. Minutes show readiness to start the interventions but caution when it came to banning large gatherings and school closures. The issue with lockdown was getting sufficient compliance and sustainability./13
assets.publishing.service.gov.uk/government/upl…

The paper I tweeted yesterday was a new iteration of the paper tabled for the 5 March dated 9 March./14
assets.publishing.service.gov.uk/government/upl…

The impact is seen in 10 March SAGE minutes with start of a shift in thinking./15
assets.publishing.service.gov.uk/government/upl…

Now noted that ‘tiered approach to social distancing might reduce its impact on the epidemic curve and on mortality’. In ‘theory maximum efficacy from all interventions would be achieved through simultaneous introduction’. By the end of the week this was direction of policy./16

The problem with the Channel 4 ‘scoop’ is that gave Prof Riley’s paper no context and suggested it had no antecedents. /17

As I have argued elsewhere the problem with SAGE’s approach was its gradualism and the belief that it had more time than it did. While we can see an important shift at the meeting on 10 March it was not sufficient. /18
google.com/search?q=IISS+…