The Swiss government issued an ordinance stating that restaurants can be kept open when R < 1, and must close when R > 1 at the cantonal level. Worth having some reflections about how R values should influence policy decisions. A thread. 1/n admin.ch/opc/de/classif…
In Switzerland, these policy decisions will be made based on point estimates of R that are provided with two (!) decimal places. For example, R is currently given as 0.98 in the canton of Valais. 2/n covid19.admin.ch/en/repro/val?t…
To be honest, the ordinance sounds a bit like a technocratic idiocy to me. But let’s have a look at how @GOVUK uses R as a guide during the pandemic response. 3/n gov.uk/guidance/the-r…
"Estimates for R and growth rates are shown as a range, and the true values are likely to lie within this range." The current range in the UK is reported as 1.1-1.2. Note the careful attention to uncertainty in these estimates. 4/n
"The UK estimates of R and growth rate are averages over very different epidemiological situations and should be regarded as a guide to the general trend rather than a description of the epidemic state.” 5/n
“Given the increasingly localised approach to managing the epidemic, particularly between nations, UK level estimates are less meaningful than previously." 6/n
"When the numbers of cases or deaths are at low levels and/or there is a high degree of variability in transmission across a region, then care should be taken when interpreting estimates of R and the growth rate." 7/n
"Different modelling groups use different data sources to estimate these values using mathematical models that simulate the spread of infections. Some may even use all these sources of information to adjust their models to better reflect the real-world situation.” 8/n
“There is uncertainty in all these data sources so estimates can vary between different models, so we do not rely on just one model; evidence from several models is considered, discussed, combined, and the growth rate and R are then presented as ranges.” 9/n
“The most likely true values are somewhere within the ranges." 10/n
Bottom line: Considering the amount of uncertainty that comes with estimates of R, don’t base your policy decisions solely on values of R. 11/n
• • •
Missing some Tweet in this thread? You can try to
force a refresh
