Naked Science Forum

Life Sciences => Physiology & Medicine => COVID-19 => Topic started by: katieHaylor on 13/05/2020 16:04:10

Title: How is R calculated?
Post by: katieHaylor on 13/05/2020 16:04:10

Roddie says:

I understand what the R number purports to show, but am unclear about how it is calculated. Not looking for an explanation of complicated formulae, rather what data is required as inputs?

Am I correct to assume that the more testing results available, the more robust the number would be?  If so, how confident can the current government scientific advisers be about the figure being below 1 in UK, as is being claimed?

What do you think?

Title: Re: How is R calculated?
Post by: vhfpmr on 13/05/2020 17:14:26

It's calculated from the growth of the infection rate:

R large: fast growth
R small: slower growth
R=1: infection rate stays the same
R just below 1: epidemic shrinks
R well below 1: epidemic shrinks faster

Bigger samples give better data accuracy, but only up to a point, you don't need to test the whole population. Like you don't need to toss 65 million coins to find out the probability of heads or tails.

Title: Re: How is R calculated?
Post by: Bored chemist on 13/05/2020 17:24:59

It's calculated from a  model. They model the way in which the number of cases changes with time  and then see which value of R most closely represents the real data.