The roadmap should be transparently and explicitly linked to data

The much-anticipated roadmap out of lockdown was announced and published on Monday. While the prospect of heading to the disco on 21 June is exciting, the focus on dates, as opposed to data, was a shame.

Also disappointed was the Tony Blair Institute. Their “lockdown lessons” paper, published last week, had suggested explicitly linking the easing, or tightening, of restrictions to the data. They proposed a framework for linked tiers and alerts to triggers – see image below.

While basic (no account for NHS capacity or vaccination efforts), it seems like a good start.

There are many scientific nuances to consider before removing restrictions, but not having an open and transparent framework linked explicitly to public data is more than just a shame. Explicit measures set expectations that reinforce the intended outcome. They avoid “the politics” taking control of what should be rational decisions. A bit like how inflation-targeting by the Bank of England helped remove destabilising inflation.

To see how a framework might have helped, the chart below shows infection rates per 100,000 over the past few months mapped to the TBI’s proposed framework. Level 1 is noticeably not reached – even in the summer. Also noticeable is the delay in ramping up restrictions in mid-December.

Would decision-makers have been happier to ramp up restrictions before Christmas if there had been agreed trigger points based on public data? Would knowing the direction of travel of case numbers have led to more cautious behaviour in the population? Would being more cautious have helped stop the need for additional restrictions?

We may never know what might have happened if Lockdown 3 had started in mid-December. Perhaps we would be a nation of rulebreakers! Equally, the modelling that we did for the Telegraph in December suggested that a proper pre-Christmas lockdown could have led to only 6,300 deaths in January – significantly lower than the ~28,000 actual deaths in January.*

Looking forward, based on the current rate of transmission (with a slight uplift for schools returning), we have mapped the expected case rate onto the case rate framework – see chart below. This chart shows the dates when our data suggest the Government could reduce the restrictions (e.g. hospitality open from 30 April).

There is some alignment between these data dates and the first two steps of the Government’s plan. It is also doubtful that there isn’t a huge amount of modelling behind the proposed dates. So why the apparent secrecy or lack of willingness to provide a more transparent framework ahead of time?

The vaccination programme’s success is something to be optimistic about, and the TBI framework doesn’t explicitly take this into account.** It may also just be me, but I feel conditioned to expect the worst from this pandemic and its lockdown dancing partner. Perhaps the vaccination programme will be like the awkward friend that sends you home early from the disco, but indeed there are still reasons to be a bit cautious on the dancefloor?

————
Sign up for our newsletter here.
Published 26 February 2021

————
* Deaths are modelled weekly, so these numbers will not align perfectly. It is important to note that our projections were significantly lower than 28,000, possibly due to more mixing than expected and the highly transmissible B117 variant even in our worst-case scenario.
** Success with the vaccinations would, of course, reduce the amount of transmission, so would reduce case numbers.