When will this shit end?

[stuartbert two hats]

35 minutes ago, Toilet Duck said:

Had a look through the report, brevity is not one of my strong suits! So, to save you reading any further, the long and short of it is that it builds on the Imperial report, but essentially reaches the same conclusion and justifies the approach taken so far. Basically, what the report shows is that using non-pharmaceutical interventions alone, repeated, stringent measures need to be introduced over a prolonged period to keep ICU admissions below capacity and avoid the consequent spike in mortality associated with a healthcare system that cannot cope. They model out to the end of 2021 with reintroduction of various levels of restrictions and conclude that repeated, stringent control measures are required to keep fatalities down (glimmers of hope will be offered at the end of the next part, if you want to skip to that!).

For anybody who is interested...with the caveat that I'm not an epidemiologist, nor a mathematician (though I work with both regularly):

Like the Imperial model before it, this report examines the effect of different public health interventions on the transmission of the virus, what impact this will have on hospitalisations and what proportion of those will require ICU admission. By extension, this is then used to predict the amount of people who will become infected and the amount of people who may die. 

The main differences here compared to the Imperial report (the one that resulted in a change of strategy earlier in March) are that in this model, age stratification is introduced and additional public health measures are also modelled (such as school closures, but care for schoolchildren by grandparents). They also looked at nationwide implementation of measures versus local controls. The age stratification is an interesting addition, as in the Imperial model, the case fatality rate (CRF) was estimated at 1%, which was not unreasonable, but the analysis here is somewhat more nuanced (with CFRs ranging from 0% for the 0-9 age group, right up to 7.68% for the over 80s, though these are adjusted CFRs  and I can't exactly figure out how they adjusted them!...they say they used hospitalisation/mortality rates from the Wuhan outbreak and adjusted using the Diamond Princess outbreak, but both are examples of spread among close contacts in confined spaces (lockdown in Wuhan was pretty severe), so I don't know how relevant a method for adjusting CRFs this is if community transmission is the principal way the virus is spreading (which is certainly the case here in Ireland)). The other key addition to the model is the inclusion of estimates of asymptomatic cases and pre-symptomatic (sub-clinical) cases, adjusted for how infectious they might be (and the numbers they use are close to those described in a paper that was published in Science a couple of weeks ago estimating how infections asymptomatic cases might be, so it seems reasonable). 

Key messages from the report are that individual measures (school shutdown, banning large gatherings, working from home, case isolation, shielding/cocooning high risk groups and so on) are ineffective on their own. Only in combination do they have the required impact on case numbers, ICU admission and ultimately on fatalities (and even in combination, "lockdown" at the current level is required to make them completely effective. They looked at different combinations of these, such as closing schools only,  but even as little a 1 contact per week between schoolchildren and grandparents wold negate the impact of the measure. There are a few odd weightings in some of their conclusions (for example, under a lockdown scenario, symptomatic individuals are assumed to be 65% as infectious as they would be under a free for all, but almost all of that infectiousness is weighted by home contact, which remains at 100% no matter what intervention strategy they model...this assumes that everybody in a household will become infected if there is a symptomatic case in the house, but that is at odds with all available data (it's entirely possible to be in contact with somebody who is infectious and not catch the disease if you take the correct precautions)..but the key part of this is that it doesn't contribute massively to the R0 of the virus as numbers of contacts are reduced significantly and that lockdowns reduce it to below 1, which is the target for "flattening the curve". However, "lockdown" can't go on indefinitely, so they look at what happens when restrictions are relaxed and when they are reintroduced. They, like the imperial report before them, use ICU admissions as triggers for the implementation of more stringent public health measures. In this scenario, there's not a huge difference between triggering "lockdown" at 1000 bed occupancy Vs 2000, both stay close to ICU capacity, whereas higher triggers (5000 beds) will quickly overwhelm capacity. But the conclusion from this part of the modelling is peaks and troughs of 2 month lockdowns with about 1 month in between them till the next one (all the way out to December 2021). With a 1000 bed trigger, this predicts just over 5k admissions in the peak weak of each surge, with 1.4k fatalities. The final conclusion drawn is stated as "we estimated that a scenario in which more intense lockdown measures were implemented for shorter periods may be able to keep projected case numbers at a level that would not overwhelm the health system"...

So, as per the Imperial report, pretty grim reading at first glance. But, some of the same glimmers of hope at that stage, still exist here. All of the models assume no pharmaceutical intervention and also do not model behavioural change. Both of which I really believe could have significant impact on how things play out. The CDC is currently revising its advice on mask wearing. Previous advice was that surgical masks were in short supply, offered little protection and should be left for frontline healthcare workers. I entirely agree that PPE for frontline healthcare staff is essential and the general public bulk buying any they can get their hands on is only going to make matters worse. However, the type of PPE under discussion here is different. A surgical mask (or any face covering) is used to protect the other person, not primarily the wearer (surgeons don't mainly wear them to stop themselves from picking up an infection from their patient, they wear them to protect you from picking up and infection from them when they open you up). The type of PPE required on the frontline to protect the wearer is different and not necessary for the general population when restrictions are lifted and we start to move about again. But in situations where people are in close contact (i.e. public transport), any kind of face covering reduces transmission rates and I'd like to see more discussion on this as part of an exit strategy. Hand hygiene too and individual protective measures are also not factored into the models (they measure populations not individuals, so we can all take personal responsibility to reduce our risk). As I pointed out last time wrt to the Imperial report, pharmaceutical intervention will shift the entire model substantially. APN01, an actual coronavirus therapeutic, has just entered Phase II trials having been deemed safe at Phase 1. It's based on the receptor the virus uses to get into our cells and we'll see how effective it is. More an more of this will happen, in addition to proper assessment of drugs that have demonstrated anecdotal success. I would be much more confident that treatment for at least some patients would be available in the short-term (with vaccination a longer term goal). And then there's the serology tests...these will give a much clear picture of how extensive infection rates are, what the actual asymptomatic case rate is and inform the models even more accurately (plus maybe allowing at least frontline staff to get back to work). The promise of tests within days has run into the problem of accuracy (often things look great in the lab, but not so great when you see how they perform in actual patients!), so we may have to wait a little bit on this, but it won't take for ever. They may even form part of aggressive testing and contact tracing between peaks that can also shift the model and spread the peaks. 

Anyway, a long post to basically say that the new model doesn't change things a whole lot! 

Wasn't it saying something alarming about the title of the thread, I.e. when does this shit end? I.e. when are we going to be able to lift lockdown?

[zahidf]

Fucking idiots 

 

 

[Smeble]

7 minutes ago, zahidf said:

Fucking idiots 

 

 

What is it about 5g that particularly riles the morons? 4g is fine but 5g is giving us all cancer and covid19? Morons.

[Toilet Duck]

15 minutes ago, stuartbert two hats said:

Wasn't it saying something alarming about the title of the thread, I.e. when does this shit end? I.e. when are we going to be able to lift lockdown?

Based on the models (and on personal communication from colleagues in China), 7-9 weeks is required. So, I'd expect stepping down sometime in May, greater relaxation into June then it's down to how measures are implemented to change the model predictions...it's either rinse and repeat, or we manage to alter it in some way (drugs, new tests to more accurate identify cases, better contact tracing, altered public behaviour to enhance social distancing without having to shut everything down...any number of things. getting buy in from the public should be easier when we've seen what the alternative is). 

[steviewevie]

Number of deaths up 684 to 3605.

10,000 by this time next week?

[Lizzim]

1 minute ago, Smeble said:

What is it about 5g that particularly riles the morons? 4g is fine but 5g is giving us all cancer and covid19? Morons.

My brother in law sent a you tube video of some guy explaining it.  I thought it was a spoof it was so ridiculous, especially when he started quoting revelations.  But my bro in law believes all that shit.   

[Deaf Nobby Burton]

8 minutes ago, steviewevie said:

Number of deaths up 684 to 3605.

10,000 by this time next week?

That’s 1000 deaths a day, it’s possible but not at the current rate of change 

Edited April 3, 2020 by Deaf Nobby Burton

[stuartbert two hats]

24 minutes ago, Toilet Duck said:

Based on the models (and on personal communication from colleagues in China), 7-9 weeks is required. So, I'd expect stepping down sometime in May, greater relaxation into June then it's down to how measures are implemented to change the model predictions...it's either rinse and repeat, or we manage to alter it in some way (drugs, new tests to more accurate identify cases, better contact tracing, altered public behaviour to enhance social distancing without having to shut everything down...any number of things. getting buy in from the public should be easier when we've seen what the alternative is). 

Did that report specifically imply a date when lockdown could be lifted?  

[Madyaker]

The likes of India and Pakistan are going to make uk numbers look tiny I fear. 

[whitehorses]

35 minutes ago, Smeble said:

What is it about 5g that particularly riles the morons? 4g is fine but 5g is giving us all cancer and covid19? Morons.

No, they also think 3G gave us SARS (2003) and 4G gave us Swineflu (2009). 
 

True story. Got a lunatic on my fb spouting this stuff all the time 

Edited April 3, 2020 by whitehorses
Sp

[steviewevie]

9 minutes ago, Madyaker said:

The likes of India and Pakistan are going to make uk numbers look tiny I fear. 

This.

...also Africa, poorer areas of middle east etc. UN/WHO need to come up with a plan to help these places.

[steviewevie]

Do people think 5G is part of a Chinese conspiracy to give us covid19?!

[Matt42]

19 minutes ago, whitehorses said:

No, they also think 3G gave us SARS (2003) and 4G gave us Swineflu (2009). 
 

True story. Got a lunatic on my fb spouting this stuff all the time 

Very strange innit. A lot of people believe this stuff pretty intensely.

[guypjfreak]

2 hours ago, Toilet Duck said:

Had a look through the report, brevity is not one of my strong suits! So, to save you reading any further, the long and short of it is that it builds on the Imperial report, but essentially reaches the same conclusion and justifies the approach taken so far. Basically, what the report shows is that using non-pharmaceutical interventions alone, repeated, stringent measures need to be introduced over a prolonged period to keep ICU admissions below capacity and avoid the consequent spike in mortality associated with a healthcare system that cannot cope. They model out to the end of 2021 with reintroduction of various levels of restrictions and conclude that repeated, stringent control measures are required to keep fatalities down (glimmers of hope will be offered at the end of the next part, if you want to skip to that!).

For anybody who is interested...with the caveat that I'm not an epidemiologist, nor a mathematician (though I work with both regularly):

Like the Imperial model before it, this report examines the effect of different public health interventions on the transmission of the virus, what impact this will have on hospitalisations and what proportion of those will require ICU admission. By extension, this is then used to predict the amount of people who will become infected and the amount of people who may die. 

The main differences here compared to the Imperial report (the one that resulted in a change of strategy earlier in March) are that in this model, age stratification is introduced and additional public health measures are also modelled (such as school closures, but care for schoolchildren by grandparents). They also looked at nationwide implementation of measures versus local controls. The age stratification is an interesting addition, as in the Imperial model, the case fatality rate (CRF) was estimated at 1%, which was not unreasonable, but the analysis here is somewhat more nuanced (with CFRs ranging from 0% for the 0-9 age group, right up to 7.68% for the over 80s, though these are adjusted CFRs  and I can't exactly figure out how they adjusted them!...they say they used hospitalisation/mortality rates from the Wuhan outbreak and adjusted using the Diamond Princess outbreak, but both are examples of spread among close contacts in confined spaces (lockdown in Wuhan was pretty severe), so I don't know how relevant a method for adjusting CRFs this is if community transmission is the principal way the virus is spreading (which is certainly the case here in Ireland)). The other key addition to the model is the inclusion of estimates of asymptomatic cases and pre-symptomatic (sub-clinical) cases, adjusted for how infectious they might be (and the numbers they use are close to those described in a paper that was published in Science a couple of weeks ago estimating how infections asymptomatic cases might be, so it seems reasonable). 

Key messages from the report are that individual measures (school shutdown, banning large gatherings, working from home, case isolation, shielding/cocooning high risk groups and so on) are ineffective on their own. Only in combination do they have the required impact on case numbers, ICU admission and ultimately on fatalities (and even in combination, "lockdown" at the current level is required to make them completely effective. They looked at different combinations of these, such as closing schools only,  but even as little a 1 contact per week between schoolchildren and grandparents wold negate the impact of the measure. There are a few odd weightings in some of their conclusions (for example, under a lockdown scenario, symptomatic individuals are assumed to be 65% as infectious as they would be under a free for all, but almost all of that infectiousness is weighted by home contact, which remains at 100% no matter what intervention strategy they model...this assumes that everybody in a household will become infected if there is a symptomatic case in the house, but that is at odds with all available data (it's entirely possible to be in contact with somebody who is infectious and not catch the disease if you take the correct precautions)..but the key part of this is that it doesn't contribute massively to the R0 of the virus as numbers of contacts are reduced significantly and that lockdowns reduce it to below 1, which is the target for "flattening the curve". However, "lockdown" can't go on indefinitely, so they look at what happens when restrictions are relaxed and when they are reintroduced. They, like the imperial report before them, use ICU admissions as triggers for the implementation of more stringent public health measures. In this scenario, there's not a huge difference between triggering "lockdown" at 1000 bed occupancy Vs 2000, both stay close to ICU capacity, whereas higher triggers (5000 beds) will quickly overwhelm capacity. But the conclusion from this part of the modelling is peaks and troughs of 2 month lockdowns with about 1 month in between them till the next one (all the way out to December 2021). With a 1000 bed trigger, this predicts just over 5k admissions in the peak weak of each surge, with 1.4k fatalities. The final conclusion drawn is stated as "we estimated that a scenario in which more intense lockdown measures were implemented for shorter periods may be able to keep projected case numbers at a level that would not overwhelm the health system"...

So, as per the Imperial report, pretty grim reading at first glance. But, some of the same glimmers of hope at that stage, still exist here. All of the models assume no pharmaceutical intervention and also do not model behavioural change. Both of which I really believe could have significant impact on how things play out. The CDC is currently revising its advice on mask wearing. Previous advice was that surgical masks were in short supply, offered little protection and should be left for frontline healthcare workers. I entirely agree that PPE for frontline healthcare staff is essential and the general public bulk buying any they can get their hands on is only going to make matters worse. However, the type of PPE under discussion here is different. A surgical mask (or any face covering) is used to protect the other person, not primarily the wearer (surgeons don't mainly wear them to stop themselves from picking up an infection from their patient, they wear them to protect you from picking up and infection from them when they open you up). The type of PPE required on the frontline to protect the wearer is different and not necessary for the general population when restrictions are lifted and we start to move about again. But in situations where people are in close contact (i.e. public transport), any kind of face covering reduces transmission rates and I'd like to see more discussion on this as part of an exit strategy. Hand hygiene too and individual protective measures are also not factored into the models (they measure populations not individuals, so we can all take personal responsibility to reduce our risk). As I pointed out last time wrt to the Imperial report, pharmaceutical intervention will shift the entire model substantially. APN01, an actual coronavirus therapeutic, has just entered Phase II trials having been deemed safe at Phase 1. It's based on the receptor the virus uses to get into our cells and we'll see how effective it is. More an more of this will happen, in addition to proper assessment of drugs that have demonstrated anecdotal success. I would be much more confident that treatment for at least some patients would be available in the short-term (with vaccination a longer term goal). And then there's the serology tests...these will give a much clear picture of how extensive infection rates are, what the actual asymptomatic case rate is and inform the models even more accurately (plus maybe allowing at least frontline staff to get back to work). The promise of tests within days has run into the problem of accuracy (often things look great in the lab, but not so great when you see how they perform in actual patients!), so we may have to wait a little bit on this, but it won't take for ever. They may even form part of aggressive testing and contact tracing between peaks that can also shift the model and spread the peaks. 

Anyway, a long post to basically say that the new model doesn't change things a whole lot! 

To much time on your hands.... 

Aaaaaa@aaagggggggggg 

Fuck 

[Toilet Duck]

29 minutes ago, stuartbert two hats said:

Did that report specifically imply a date when lockdown could be lifted?  

Nope! They show the models and the fluctuation in cases based on imposition and relaxation of restrictions. I don't think anybody can say with any certainty when restrictions might be lifted, but 2-3 months is all anybody is willing to venture (and based on the available models, they will have to be reintroduced unless we are able to do something different to control cases). Extrapolating out what anyone is willing to say publicly, from mid march when things started to be introduced, we are looking at mid-May before we start to think about lifting restrictions (restrictions started a bit later in the UK, so these dates might be more applicable to where I am in Ireland, probably closer to the end of May in the UK). We've been told in Ireland that the current restrictions are until April 12th, but in recent days there are various ministers saying they may need to be in place for longer (they will). Our own modelling in work doesn't project us being back in before the summer and we are now in contingency planning for students not being able to start in September/October. 

 

Graph below is taken from the report, pink shading is public health restrictions, shaded parts in the pink block are intensive controls (i.e. lockdown), blue shading is scheduled school/university closures. Basically, it shows two month blocks with a month/6 weeks between each..bold black line is actual ICU capacity, broken line is double capacity should this be possible. This is with the lowest admission trigger, which keeps ICU admissions within existing capacity...(there's an odd peak after relaxation of social distancing at the end that doesn't occur with higher trigger points...why this is, I don't know and they don't get into it in the report...possibly greater "herd immunity" at higher trigger points?...who knows, I haven't thought about it enough) 

 

Higher trigger (2000 ICU admissions) leaves longer gaps between the lockdowns and shorter lockdowns...still within ICU capacity and fatality rate slightly higher, but not massively so...

 

With a much higher trigger (5000 ICU admissions), lockdowns periods are shorter again, with bigger gaps between them, but even double ICU capacity is overwhelmed and CFR increases a lot. 

 

So, based on that, increasing ICU capacity, introduction of therapeutics that work (this will reduce ICU time hopefully), and additional measures between the peaks to spread them out are all ways in which the projected models can be influenced and limit the amount of time we all spend indoors! 

[Toilet Duck]

17 minutes ago, guypjfreak said:

To much time on your hands.... 

Aaaaaa@aaagggggggggg 

Fuck 

😀

Actually, I wrote it on my lunch break! but I started to think about a therapeutic approach last night, so have been spending some time reading about it today (and any and all information about SARS-CoV-2 is useful to me). 

Also, don't read the next post after yours! 😁

[steviewevie]

So...from those graphs it will be a case of..."Ok, everyone out!"......"Ok, everyone back inside!"......repeat for 12 months or so...

?.

[steviewevie]

I wonder if they'll close the parks next week after all the front page  pics of people out and about over the weekend.

[Toilet Duck]

9 minutes ago, steviewevie said:

So...from those graphs it will be a case of..."Ok, everyone out!"......"Ok, everyone back inside!"......repeat for 12 months or so...

?.

If that's all we do, then yes (in theory). But better testing, better contact tracing, better case isolation, better personal risk mitigation measures, better drugs, and better capacity in the healthcare system all influence how long the gaps are...we also have no idea what impact climate has (or any prior immunity to other coronaviruses), and we are only guessing about the prevalence of asymptomatic cases based on what we can observe in transmission, so really until we know how many people have really been exposed to it, we have no idea about when levels of infection will reach a point where transmission naturally slows. What they do tell us is that leaving it without any intervention would be a nightmare and that it can in theory be brought under a degree of control that doesn't come close to that nightmare. 

[Deaf Nobby Burton]

Let’s just hope the vaccine is ready before next June, I don’t want to be talking about double ticket roll overs in a years time!

[Crazyfool01]

8 minutes ago, steviewevie said:

I wonder if they'll close the parks next week after all the front page  pics of people out and about over the weekend.

its ok everyone is going to listen to Queenie when she tells us off from her palace or castle on Sunday .... presume she's cooking her own beans on toast these days 

[steviewevie]

The massive problem with the vaccine will be how it is distributed globally.

[Deaf Nobby Burton]

6 minutes ago, steviewevie said:

The massive problem with the vaccine will be how it is distributed globally.

Yes it won’t be readily available to everyone, but if we can get enough to vaccinate the most vulnerable then that will be a hugely significant help. Not everyone takes the flu jab, I think the most vulnerable are written to about it and of them about 75% take it up.

[st dan]

‘Good’ news is that Italy certainly appears to have reached its peak both in terms new cases and deaths. As we are a number of weeks behind them, we can use their example of how they exit the lockdown and see if their strategy works or not. 

In ‘bad’ news - USA is really heading for a horror show at this rate. 

[steviewevie]

1 hour ago, crazyfool1 said:

its ok everyone is going to listen to Queenie when she tells us off from her palace or castle on Sunday .... presume she's cooking her own beans on toast these days 

She's not on till 8pm...so after a nice day out with friends and family can go home and watch the Queen tell us to stay indoors.