The tail after the peak

1finny

1finny

Well-known member
Just some simple maths
Nearly 2000 new cases today
Current death rate is around 14% of cases
Means in a week/two weeks we will still be announcing deaths of around 250 - 280

What am I missing
Other than, if I’ve got it right our ‘tail’ is pretty long
 
1finny said:
Just some simple maths
Nearly 2000 new cases today
Current death rate is around 14% of cases
Means in a week/two weeks we will still be announcing deaths of around 250 - 280

What am I missing
Other than, if I’ve got it right our ‘tail’ is pretty long
Click to expand...
Nothing. It's a long way to go. Possibly 6 weeks to get to a point where numbers are down to 100 per day. We'll possibly be treading water as things open up with track and trace keeping R steady.
 
The University of Washington projections have us at 100 deaths around 22 June and 50 deaths around the 6th July, I think the success of track and trace is going to be very important to reach those figures
 
It’s too early to say, but the data for the deaths last 2 days are higher than last week, suggest that things have taken a turn. Hope they are just an anomaly
 
1finny said:
Just some simple maths
Nearly 2000 new cases today
Current death rate is around 14% of cases
Means in a week/two weeks we will still be announcing deaths of around 250 - 280

What am I missing
Other than, if I’ve got it right our ‘tail’ is pretty long
Click to expand...

I think the difference is that the majority of deaths took place when we were only testing those most seriously ill (Pillar 1). Therefore, the 14% rate is really only applicable to that sub-set.

Now we are testing in far higher numbers, with the majority of positive tests coming from key workers and others (Pillar 2), who tend to be much less ill. In fact, from its peak, the number of positive Pillar 1 tests has fallen from an average of approx. 4,500 per day to approx. 900 per day now.

If you apply the 14% rate to the 900 figure, you'd expect approx. 125 deaths per day in around two weeks' time. That is broadly consistent (allowing for deaths in other settings) with our current fortnightly rate of decline (40%), which would take us from an average of 250 per day at the moment to approx. 150 per day in two weeks' time.
 
Billy Horner said:
I think the difference is that the majority of deaths took place when we were only testing those most seriously ill (Pillar 1). Therefore, the 14% rate is really only applicable to that sub-set.

Now we are testing in far higher numbers, with the majority of positive tests coming from key workers and others (Pillar 2), who tend to be much less ill. In fact, from its peak, the number of positive Pillar 1 tests has fallen from an average of approx. 4,500 per day to approx. 900 per day now.

If you apply the 14% rate to the 900 figure, you'd expect approx. 125 deaths per day in around two weeks' time. That is broadly consistent (allowing for deaths in other settings) with our current fortnightly rate of decline (40%), which would take us from an average of 250 per day at the moment to approx. 150 per day in two weeks' time.
Click to expand...

Actually, thinking about it, we should probably exclude positive Pillar 2 tests from the figures, as they've only been conducted for a relatively short period of time. If we look at the death rate as a proportion of just the positive Pillar 1 tests, this increases it to 20%.

Apply that proportion to the 900 per day average and you would expect approx. 180 deaths per day in a fortnight. That would be approx. 30% lower than the current average, so still lower than today but perhaps a slight slowing in the decline.
 
GazC_MFC said:
It’s too early to say, but the data for the deaths last 2 days are higher than last week, suggest that things have taken a turn. Hope they are just an anomaly
Click to expand...

Hopefully it's the long bank holiday weekend anomaly Gaz. I do think the 7 day rolling average is the best and most indicative metric.
 
I feel that the best thing to look at is the excess deaths stats. Although there is lag I feel that it is best to wait. Then judge by looking at these stats.

Plus wait longer for corrections.
 
Every day that passes our doctors are getting better at dealing with this virus, so the mortality rate is declining. I think the numbers will be more like what Billy Horner states, but at the lower end of that scale.
 
If we are running at 9000 new infections a day now and a fatality rate of 0.75% that's around 70 deaths per day say 3 weeks into the future. I assume an average gap of 3 weeks between confirmed infection and death. Possibly we can get the fatality rate down with time with better equipment and better knowledge. The problem we have in this country is that the virus is all over the place and our lock down has been limited in its impact. If the lock down was strict and strictly followed with excellent PPE we would have very low figures by now and we have not. One example in this country is that people generally don't wear masks so if they have it they are spreading it much more. We are effectively on "herd immunity lite"
 
Your maths is flawed, the reason it’s 14% is based on the early positives we’re frompeople already in hospital, the majority in bad way, now anyone can get tested and don’t have to be in hospital.

The agreed mortality rate is nearer 1% and probably lower when you include the asymptomatics.

It was said the other day around 15% of Londoners had had it and 5% of the rest of the country earlier this month. At 14% mortality there would be 150000 dead in London and another 400000 from everywhere else in the uk, when it’s less than a tenth of that including all of the additional deaths.
 
gravy173 said:
Your maths is flawed, the reason it’s 14% is based on the early positives we’re frompeople already in hospital, the majority in bad way, now anyone can get tested and don’t have to be in hospital.

The agreed mortality rate is nearer 1% and probably lower when you include the asymptomatics.

It was said the other day around 15% of Londoners had had it and 5% of the rest of the country earlier this month. At 14% mortality there would be 150000 dead in London and another 400000 from everywhere else in the uk, when it’s less than a tenth of that including all of the additional deaths.
Click to expand...

You're mixing up your fatality/mortality rates. The 14% figure is the current Case Fatality Rate (CFR) in the UK as of May 28. It had been higher (15.75% on May 19, for example). As the graph below mentions, this is the percentage of people with confirmed cases, that went on to die. However, as it also says, CFR is not really a measure of the mortality risk from the disease to the population as a whole.

IMG_20200529_114756.png
For instance, The latest figures I could find for the total of confirmed cases for London was 26,883. Using a case fatality rate of 14% would mean 3,763 deaths not 150,000.
 
Last edited:
GazC_MFC said:
View attachment 3514
Click to expand...

Thanks Gaz. I've read the PHE report, but they don't publish the underlying data for some reason, so it's impossible to analyse.

It does seem to fly in the face of other data available. The total number of people in hospital with Covid-19 has continued to fall in all regions (including London), and although there were a couple of days in the past week where there were small rises in London, these were vastly outnumbered by the falls on the other days.

According to the Intensive Care National Audit and Research Centre, the number of people in ICU has fallen every day (with one minor blip) since 16th April. However, that data only runs until 22nd May and covers the whole of England, Wales and Northern Ireland, so it would be impossible to pick out a recent uptick in the London region alone.

I think we'll just have to wait until PHE publish next week's report to see if this was a blip or the beginning of a new trend. Bit frustrating though.
 
You must log in or register to reply here.
Back
Top