I don’t really understand the R rate

[Fettuccinecarbonara]

As I understand it, the R rate is the rate of infection; average number of people that one infected person can expect to pass the coronavirus on to if no containment/protective measures are introduced. Am I right?

This seems to mean, that if I had coronavirus back in April, I would most likely infect 3 other people, despite us all being in lockdown.

If I have coronavirus now, I’m likely to infect one, maybe less, person. This is before protective measures are introduced.

So, how, exactly? Does the virus become less infective? Please can you explain?

@Fettuccinecarbonara

As I understand it, the R rate is the rate of infection; average number of people that one infected person can expect to pass the coronavirus on to if no containment/protective measures are introduced. Am I right?

This seems to mean, that if I had coronavirus back in April, I would most likely infect 3 other people, despite us all being in lockdown.

If I have coronavirus now, I’m likely to infect one, maybe less, person. This is before protective measures are introduced.

So, how, exactly? Does the virus become less infective? Please can you explain?

Hi, not quite, the r rate is very affected by what you do, it hasn't changed, we have changed.

Lockdown is what reduced the r rate. Ours would go back up if we all started hugging and and sharing glasses and going to the theatre and big dinners.

The number the government reports is only an average.

So each disease has a r rate for normal conditions.

If you have 100 people in Britain who each give it to 1 person (r rate = 1) you don't have a massive problem. If you have 1000 people who give it to 1 each you have a bigger problem. If you have 100 people who give it to 2 each, on the next measure you don't have a massive problem - but you have a growing problem.

Day 1 1000 X 1 = 1000
Day 4 1000 X 1 = 1000
Day 7 1000 X 1 = 1000 etc.

Vs

Day 1 100 X 2 = 300
Day 4 300 X 2 = 600
Day 7 900 X 2 = 1800
Day 11 900x 2 = 3600 etc.

So the two numbers that matter are number of cases, and number of people they infect.

Sorry, maths error, day 11 should be 1800x 2 = 3600

Oh my god, my whole things is a maths error - I started with threes and changed to 2s and didn't correct. I'm a shit proofreader.

These are right now I think:
Day 1 100 X 2 = 200
Day 4 200 X 2 = 400
Day 7 400 X 2 = 800
Day 11 800 X 2 = 1600 etc

I actually work with figures too thank goodness this is anonymous

There's also some evidence to suggest that it's not quite as simplistic as that - a lot of people will infect no-one, then you'll have 'super spreaders' who will infect 10's of people. I think one outbreak was traced to a single person who infected 70-odd other people.

Also, it's less useful as the cases decline as the numbers can be skewed by a single outbreak. This has happened in Germany, where an outbreak in a meat processing plant pushed the R above 1 for a short time - however by locking down the meat plant and surrounding area, they were able to prevent it reverting to exponential growth and meaning that cases in the actual community were contained.

If R is less than 1, it means that some people who haVe Covid infect no one else as it is impossible to infect 0.8 of a person. But some people must infect one or two others.
OP I think it looks high inApril because of a lag in the figures and the time it takes between infection and a positive test. In April, tests were in short supply too meaning more of a lag.
If R is lower than 1 numbers of infected people are falling. If it's higher, they're going up.

The strategy of isolating people without symptoms is to reduce the number of people they infect and therefore to reduce R. So one person gets symptoms and confirmed as infected, their contacts in preceeding days are asked to stay home; when they get to the most infectious period (the day before and day of symptoms) they're at home with less people to potentially infect. Even if not everyone sticks to it in aggregate the strategy reduces the rate of infection so numbers start to come down.

85% of cases early on were thought to be from super-spreaders, individuals who infected lots of people through their workplace / sporting event / whatever. If there are precautions about meeting in public and removal of noisy areas that have people shouting, those events become less likely. Where there are outbreaks at workplaces, the UK are isolating all staff and now testing asymptomatic people too, that will all hopefully bring down infection rates too.

This seems to mean, that if I had coronavirus back in April, I would most likely infect 3 other people, despite us all being in lockdown.

The R is the average number of people an infected person will infect.

Before lockdown, it was thought to be about 3. After lockdown in April it was dramatically lowered to below 1.

1 is the critical number... If it’s above 1, numbers increase (because the average person infects more than one other person). If it’s below 1, numbers decrease.

With an R of 3... 1 person infects 3, they infect 9, and they infect 27, and it very, very quickly gets out of hand. Even with a R not much above 1, numbers can increase significantly over quite a short period.

So it worries me when people think “numbers are really low - it’s all over”, as things could escalate very quickly if we all went back completely to normal. At the end of January numbers were lower than today... Only six weeks later things were out of control.

Also, R is the average... Evidence seems to suggest that some people infect dozens, while many, perhaps most, infect no one, with this being due to a combination of how much virus an individual with Covid is shedding and the extent to which they are mixing with others.

In April, tests were in short supply too meaning more of a lag.

Yes, back in March and early April you had to be ill enough to be admitted to hospital to be tested... the vast majority just had to stay at home and get better. It’s plausible that 100,000+ a day we’re getting infected before society started serious social distancing in mid to late March.... which was a week or so before actual lockdown.

As I understand it, the R rate is the rate of infection; average number of people that one infected person can expect to pass the coronavirus on to if no containment/protective measures are introduced. Am I right?

As far as I understand it, that's how it's generally used to show the difference in infection rates between diseases eg in normal conditions measles has a very high r rate, chicken pox is a bit lower, MERS is very low.

For the purposes of the pandemic and lockdown restrictions, its being used to refer to the current rate of infection, taking into account the measures. That's my understanding.

If we all wash our hands, wear a mask, socially distance, don’t go out to pubs, restaurants etc then positive COVID-19ers will infect fewer people, including the ones who don’t know they have it, And fewer healthy people will catch it. So overall the R number needs to be less than 1 or you start to see growth in hotspots etc.

As other people said, except that the R rate was still technically quite high during April due to what was happening in care homes and hospitals. That's why it took so long for the infection rate to dramatically reduce in mid May. In the community, it has been sub 1 or far less than that since the end of March.

I think you are thinking of the R number as something linked to a specific person.

But if the R=3 then it's not that every person will infect 3 more. It's that there is an average of 3 infections per person but the reality is that some will infect none and others will infect more.

How many are infected isn't just linked to how infectious the disease is, it's also down to what limits society has placed on infection spread. Social distancing means that it is harder to the disease to move between people, this bringing down the average infected. BUT if we all went back to behaving normally it would go back to where it was. The disease is just as infectious as it always was, but we're now behaving in ways that makes in harder for it to spread from one person to another.

I understand the mathematics of the R being average number of people an infected person will infect.

Without compulsory and constant (daily/weekly) repeat systematic mass and compulsory accurate testing how is the real "live" local R rate actually accurately predicted?

You will no doubt have some super spreaders with off the scale R rates that say infect the whole work place factory in places like Leicester which in turn creates a rapid hot spot of other super spreaders with none of them bothering to wear any PPE not stop the spread.

On the light community spread scale you have others who only inadvertently infect their household and neighbours etc.

Is it not true that the evidence from numerous overseas nations that unlike the UK implemented a normal hard enforced lockdown with mass (compulsory) test, track, trace and isolate and strict international boarder controls. suggest that the R rate cannot not be easily measured? But is a guesstimate relative figure and dependent on the number of overall detected infected cases and fatalities.

I understand many asymptomic infected spreaders can not necessarily be detected when tested due to biological timelines when the virus first enters the victim.

I am no expert but the UK Covid tally of 50k plus fatalities probably suggest our Covid strategies and lack of public support and adherence is much to be desired?

Even now after half a year is there finally some common sense acknowledgement on the protective functionality of basic personal protective equipment (PPE) with face coverings systematically worn by the whole outside of this country for obvious reasons and possibly hence totally different R rates and overall excess Covid fatalities.

There are various R rates.

At the very beginning, you have an R0 - the basic reproduction rate that applies to a population which is all susceptible to the infection, and no mitigation in being undertaken. There are estimates that the R0 for Covid is around 3 - 4, but it will differ between environments. It refers to the average number of people one infected person will infect, but research suggests that for CV we have a number of superspreader events where one person might infect many more than that whereas others infect few if any.

Once the infection starts to spread and/or mitigation is undertaken, it's not an R0 anymore.

As we go on through the pandemic, it becomes Re - effective reproduction rate in the environment as it is at that point. This is also variable, as suppression has an affect and (possibly) the population build some level of immunity.

R rates are calculated by a variety of different mathematical models, and usually in retrospect, so you will get different estimate from different modellers. At very low levels of infection, smallish outbreaks can make Re swing wildly; in Germany, it swung up to 2.88 for a day or so a couple of weeks ago due to a specific outbreak.

Do different countries have different r rates anyway, depending how much close contact or socialising is the norm in a society?

Thank you all!

And patriciaholme particularly for making me finally realise it was a retrospective decision!