Fudging the Testing Figures

[ajohn]

ahh the usual benefit of hind sight and the ability to say everything they did only we'd do it better response.

That doesn't wash well really in this case. The Asian and so on is an established technique for dealing with epidemics as has been pointed out. The basic idea is to minimise infections. The hindsight aspect goes back well before cv19 - being prepared for similar. They weren't. Usual flue fine but something much worse no. Other countries were also poorly prepared. we had to be good at something - best unprepared maybe.

 

[SirGalahad]

I'll put that down as an "I don't know" then.

Poor sad filly.

 

[ajohn]

They have come up with a drug that reduces death rates on ventilators and also people on oxygen. Old cheap steroid in tablet form. Seems it's going to become standard treatment. Hope we don't find it's best to give it at some specific point in time during an illness and they use the wrong one.

The other inconsistency is masks. Mentioned a little in shops but needed on public transport. Not mentioned at all now more shops are open. I have wondered about spread. Do all go out partying in groups that regularly? What do all use - shops. I can't help wondering if that is a major factor other than public transport but is that one in such widespread use compared with London for instance.

 

[motorbiking]

Poor sad filly.

So no ideas then?

That doesn't wash well really in this case. The Asian and so on is an established technique for dealing with epidemics as has been pointed out. The basic idea is to minimise infections. The hindsight aspect goes back well before cv19 - being prepared for similar. They weren't. Usual flue fine but something much worse no. Other countries were also poorly prepared. we had to be good at something - best unprepared maybe.

The specific objection is that the preparation didn’t extend to enough economic analysis. I’d like to see evidence of someone making noise about this before the virus. It’s so easy to say we weren’t prepared, after the fact. It’s so easy to say we’d have done a better job after the fact.

As this thread proves, not many people were voicing better ideas at the time. Some can’t even come up with suggestions for improvement after the fact with the benefit of hindsight. But let’s blame the government anyway.

 

[Scarlet Pimpernel]

Hope we don't find it's best to give it at some specific point in time during an illness ...

Evidently, it is supposed to be given at some specific time during an illness.

About 19 out of 20 patients with coronavirus recover without being admitted to hospital.
Of those who are admitted, most also recover but some may need oxygen or mechanical ventilation.
And these are the high-risk patients dexamethasone appears to help.
https://www.bbc.com/news/health-53061281​

So it only potentially helps 1 in 20 people with the virus.

Then my maths fail me.

This is interpreted as a third?

Then this is interpreted as a fifth?

Added to this, they suggest that if this drug was used from the beginning of the pandemic, up to 5,000 lives may have been saved, that's about an eight of the 40,000 (or about a twelfth of the 60,000)

Are there any maths teachers out there who could help me out. Perhaps some elementary maths lessons for me (or Priti Patel, or whoever works out these statistics)?

 

[IT Minion]

Evidently, it is supposed to be given at some specific time during an illness.

About 19 out of 20 patients with coronavirus recover without being admitted to hospital.
Of those who are admitted, most also recover but some may need oxygen or mechanical ventilation.
And these are the high-risk patients dexamethasone appears to help.
https://www.bbc.com/news/health-53061281​

So it only potentially helps 1 in 20 people with the virus.

Then my maths fail me.

View attachment 196322​

This is interpreted as a third?

View attachment 196323​

Then this is interpreted as a fifth?

Added to this, they suggest that if this drug was used from the beginning of the pandemic, up to 5,000 lives may have been saved, that's about an eight of the 40,000 (or about a twelfth of the 60,000)

Are there any maths teachers out there who could help me out. Perhaps some elementary maths lessons for me (or Priti Patel, or whoever works out these statistics)?

The government has very little direct involvement in this. Which is exactly right, they've left it to the experts.

Of those on ventilators, we expect a certain proportion to die. Around 40%. If you give everyone going onto a ventilator dexamethasone then the number who die drops to around 30%. Which means you need to treat 8 people to save a life (so only 12.5% of people treated benefit), it provides a 10% reduction in the mortality rate (40->30) or roughly 25-30% relative reduction in mortality (30/40).

The 1 in 8 and 1 in 25 metrics are an interesting way of viewing the results. It makes sense when you're looking at a limited resource, like remdesivier, or an extortionate one like Tocilizumab. But with dexamethasone it's so cheap and available it doesn't make sense to me.

 

[SirGalahad]

So no ideas then?

Eddy asked me the same question and I provided a reply. Search for it.

 

[Vinty]

.

 

[Vinty]

Evidently, it is supposed to be given at some specific time during an illness.

About 19 out of 20 patients with coronavirus recover without being admitted to hospital.
Of those who are admitted, most also recover but some may need oxygen or mechanical ventilation.
And these are the high-risk patients dexamethasone appears to help.
https://www.bbc.com/news/health-53061281​

So it only potentially helps 1 in 20 people with the virus.

Then my maths fail me.

View attachment 196322​

This is interpreted as a third?

View attachment 196323​

Then this is interpreted as a fifth?

Added to this, they suggest that if this drug was used from the beginning of the pandemic, up to 5,000 lives may have been saved, that's about an eight of the 40,000 (or about a twelfth of the 60,000)

Are there any maths teachers out there who could help me out. Perhaps some elementary maths lessons for me (or Priti Patel, or whoever works out these statistics)?

Try Diane Abbot.

 

[Scarlet Pimpernel]

Try Diane Abbot.

I'd prefer to ask Priti Patel, then I would know her figures were wrong.

"three hundred thousand, and thirty four, nine hundred and seventy four thousand"
Priti Patel 11/42020​

 

[ajohn]

Going on PHE registered deaths just short of ~13000 could be saved if available from day one based on the comment that deaths on ventilators were reduced by 1/3.

That assumes all died on a ventilator. Most will have but not possible to split further. People are likely to have been in HDU before ventilation. More info needed really though. Age groups for for instance. 80+ have the highest number of registered cases and the highest number of deaths. Not looked at % as lots of numbers would need adding up. This suggests that 80+ are far more likely to have the symptoms that could get them into hospital.

Take you pick but it saves lives.

 

[Scarlet Pimpernel]

Going on PHE registered deaths just short of ~13000 could be saved if available from day one based on the comment that deaths on ventilators were reduced by 1/3.

That assumes all died on a ventilator. Most will have but not possible to split further. People are likely to have been in HDU before ventilation. More info needed really though. Age groups for for instance. 80+ have the highest number of registered cases and the highest number of deaths. Not looked at % as lots of numbers would need adding up. This suggests that 80+ are far more likely to have the symptoms that could get them into hospital.

Take you pick but it saves lives.

Have you read your own links, which were provided by both ceres and I before you?
Obviously not!

 

[ajohn]

Have you read your own links, which were provided by both ceres and I before you?
Obviously not!

Yawn. If you read the link you would have seen 1/3 mentioned.

 

[Scarlet Pimpernel]

Yawn. If you read the link you would have seen 1/3 mentioned.

Yawn, I'll explain it again, slowly.

The results during the trial were seen as giving one third results in the one case, and one fifth results in the other case.
It does not mean that those results will be replicated in the future! Nor that the one third figure will be replicated in all cases!

From those results, the projected potential achievement is 1 in 8 possibly saved in the one case, and 1 in 25 possibly saved in the other case.

From that projected figure, it is suggested that something in the region of 4,000 to 5,000 might have been saved, if the drug had been used from the start of the pandemic.
I assume that is some conglomeration of the 1 in 8 and the 1 in 25 to produce the figure that might have been saved, I assume from the 60,000 who actually died, from the two subsets of conditions.
I assume because I don't know how they worked out the projected figure from the results, and it seems no-one else on here can explain it.

 

[transam]

Yawn, I'll explain it again, slowly.

The results during the trial were seen as giving one third results in the one case, and one fifth results in the other case.
It does not mean that those results will be replicated in the future! Nor that the one third figure will be replicated in all cases!

From those results, the projected potential achievement is 1 in 8 possibly saved in the one case, and 1 in 25 possibly saved in the other case.

From that projected figure, it is suggested that something in the region of 4,000 to 5,000 might have been saved, if the drug had been used from the start of the pandemic.
I assume that is some conglomeration of the 1 in 8 and the 1 in 25 to produce the figure that might have been saved, I assume from the 60,000 who actually died, from the two subsets of conditions.
I assume because I don't know how they worked out the projected figure from the results, and it seems no-one else on here can explain it.

Well. That's cleared that up. Than