advertisement


Coronavirus - the new strain XXI

Status
Not open for further replies.
43 941 cases, 207 new deaths and 865 admissions.

Jeez, this is simply not looking good (once more)
Then again, BJ & Co will do precisely nowt (once more)

& the minister for total nonsense Mogg says all is fine if you know the folks around you.
So what can possibly go wrong.
 
I’m not liking these new daily figures.
Last Wednesday there were 49139 new cases, so it does seem that the number of new cases is dropping quickly, albeit still big numbers.

Number of new admissions looks to have reached its peak, but the number of deaths, as I stated yesterday, is unfortunately very likely to continue rising as it follows the number of new admissions around a week ago.
 
Number of new admissions looks to have reached its peak, but the number of deaths, as I stated yesterday, is unfortunately very likely to continue rising as it follows the number of new admissions around a week ago.

That’s good to know. As stated upthread I tend to ignore the ‘cases’ figure as it is such a questionable dataset, the admissions and deaths are the real indicator.
 
Last Wednesday there were 49139 new cases, so it does seem that the number of new cases is dropping quickly, albeit still big numbers.

Number of new admissions looks to have reached its peak, but the number of deaths, as I stated yesterday, is unfortunately very likely to continue rising as it follows the number of new admissions around a week ago.

There have been a few false flats that have caught people out over the past few months
 
There have been a few false flats that have caught people out over the past few months
Clearly they are the least reliable of the three datasets, but changes in the week to week variations have often given a good indication as to what then happens to the number of admissions and the deaths. So currently I am seeing the case data as a good indicator that number of admissions are likely to drop.
 
Clearly they are the least reliable of the three datasets, but changes in the week to week variations have often given a good indication as to what then happens to the number of admissions and the deaths. So currently I am seeing the case data as a good indicator that number of admissions are likely to drop.

I think they will too but I also think there's likely to be high case numbers for a long time. The reported case numbers are a very unreliable measurement - especially now PHE is no more
 
I think they will too but I also think there's likely to be high case numbers for a long time. The reported case numbers are a very unreliable measurement - especially now PHE is no more
I am hoping that this is the start of something like the BBC article that @Andrew C! posted in #880 :

https://www.bbc.co.uk/news/health-59039739

At least for the rest of the year. Next year the figures look a bit odd as cases and deaths are high and there is no clear set of reasons as to why this could happen.
 
I am hoping that this is the start of something like the BBC article that @Andrew C! posted in #880 :

https://www.bbc.co.uk/news/health-59039739

At least for the rest of the year. Next year the figures look a bit odd as cases and deaths are high and there is no clear set of reasons as to why this could happen.
Has to do with assumptions regarding waning immunity, behaviour change and their interaction, all very uncertain apparently. This is a helpful thread on the latest SPI-M-O modelling, looking at Warwick, Imperial and Bristol models as well as LSHTM:

https://twitter.com/edmhill/status/1451684475901972480?s=21

The big wave in ‘22 shown in the graph in the BBC article assumes a slow return to normal behaviour: assuming a speedier return results in a flatter but longer wave. Sage conclude though that slower is better as it gives us more time to develop our understanding of several variables and to make evidence-based decisions.

Twitter-based modeller James Ward is concerned that a big drop in cases now may be building up trouble for January and Feb.

https://twitter.com/jamesward73/status/1453405370811097090?s=21
 
Who'd a thought it - the Torys have become a tax raising, public spending government (for the next three years at least) but they appear to have taken this course of action begrudgingly and only in response to the pandemic.
 
Has to do with assumptions regarding waning immunity, behaviour change and their interaction, all very uncertain apparently. This is a helpful thread on the latest SPI-M-O modelling, looking at Warwick, Imperial and Bristol models as well as LSHTM:

https://twitter.com/edmhill/status/1451684475901972480?s=21

The big wave in ‘22 shown in the graph in the BBC article assumes a slow return to normal behaviour: assuming a speedier return results in a flatter but longer wave. Sage conclude though that slower is better as it gives us more time to develop our understanding of several variables and to make evidence-based decisions.

Twitter-based modeller James Ward is concerned that a big drop in cases now may be building up trouble for January and Feb.

https://twitter.com/jamesward73/status/1453405370811097090?s=21
Thanks for posting the links.

Many good points made (including the very relevant changes in data collection due to half term and fewer children being tested), but it really shows that whilst we have all this data and have developed statistical models, their remains a large uncertainty in any predictions, and this gets worse as a function of time into the predicted future.

I think that I have been overly optimistic as to the quality of the statistical models. With the enormous amount of data available (albeit with varying levels of confidence an relevance)I have been expecting the use of kalman or Bayesian filters etc to help determine core simulation parameters more effectively than has been evidenced.

Underlying the predictions there is a core pattern of lower cases now and higher cases later, or higher cases now and lower cases later. Almost as if most of the measures that have been and are being put in place just move the cases around to different age groups and different times of the future. Which makes me think more and more that (apologies to all who have had enough of me mentioning this) that ventilation of buildings is extremely important in all this, as it reduces transmission, no matter the behaviour of individuals or groups and has other significant benefits to individuals, communities and companies.
 
Which makes me think more and more that (apologies to all who have had enough of me mentioning this) that ventilation of buildings is extremely important in all this, as it reduces transmission, no matter the behaviour of individuals or groups and has other significant benefits to individuals, communities and companies.

Absolutely! I don't think single doses for kids will reduce school transmission a great deal but boosters for the over-50s should do, providing the uptake is high enough - I'm not too optimistic on this after being told that my GP is not running clinics.
 
Covid measures give us choice. They are not restrictions on British life

"...What is telling here is not only the decision not to act, but the way in which the government’s framing of the issues serves to justify that decision. Covid measures are characterised as Covid restrictions or even a lockdown. Those calling for action can thereby be characterised as fanatics who want to remove our liberty while the government presents itself as defender of our freedoms, preserving “normality”.

This has been an effective strategy. After all, who wants lockdown? So if you are labelled a lockdown supporter, you have probably lost the argument. And the reason why this finger-pointing strategy has gained traction is because – probably more as a result of thoughtlessness than collusion – it is echoed by most of the media."

https://www.theguardian.com/comment...-restrictions-science-freedom-infections-safe
 
Thanks for posting the links.

Many good points made (including the very relevant changes in data collection due to half term and fewer children being tested), but it really shows that whilst we have all this data and have developed statistical models, their remains a large uncertainty in any predictions, and this gets worse as a function of time into the predicted future.

I think that I have been overly optimistic as to the quality of the statistical models. With the enormous amount of data available (albeit with varying levels of confidence an relevance)I have been expecting the use of kalman or Bayesian filters etc to help determine core simulation parameters more effectively than has been evidenced.

Underlying the predictions there is a core pattern of lower cases now and higher cases later, or higher cases now and lower cases later. Almost as if most of the measures that have been and are being put in place just move the cases around to different age groups and different times of the future. Which makes me think more and more that (apologies to all who have had enough of me mentioning this) that ventilation of buildings is extremely important in all this, as it reduces transmission, no matter the behaviour of individuals or groups and has other significant benefits to individuals, communities and companies.
Well beyond me as to whether the Do-You-Want-Your-Deaths-Now-Or-Later pattern reflects limitations of the models or the fundamental realities of disease dynamics as we move into endemicity, but one way or another I'm sure it's informing policy, for better or worse. Easy to imagine the rhetorical power of some of these graphs if you wanted to make an argument for doing nothing now.

One obvious limitation of the whole modelling arrangement (reflecting the larger politics-science arrangement), as we've discussed before, is that there doesn't seem to be a way for long term, structural features of the situation - such as ventilation and better support for isolation - to be included as variables (as far as I know - haven't read the reports themselves). They're obviously pertinent but the whole decision-making apparatus seems blind to them, for whatever reason.

Thinking about what is and isn't politically permissible to model: is the NHS being overwhelmed included somewhere in the modelling?
 
This week's surveillance report

Surveillance indicators suggest that at a national level COVID-19 activity increased in some
indicators, while remaining stable in others in week 42 of 2021. Laboratory indicators suggest
that influenza activity is very low.

Overall COVID-19 case rates remained stable in week 42. Case rates increased in some age
groups, regions and ethnic groups, while remaining stable or decreasing slightly in others. Case
rates decreased in those aged under 19. Overall Pillar 1 and Pillar 2 positivity remained stable
compared to the previous week.

The overall number of reported acute respiratory incidents in the past week decreased
compared to the previous week. SARS-CoV-2 was identified in the majority of these.

COVID-19 hospitalisations increased in week 42. Deaths with COVID-19 remained stable in the
most recent week.
 
Status
Not open for further replies.


advertisement


Back
Top