COVID-19 #13 or: How I Learned to Stop Worrying and Love the Disease

[Fury Resurrected]

29 minutes ago, Triskele said:

Not sure how different the US system is from others, but they definitely did use it on Ebola.  It just would have been done then as an investigational drug rather than an approved drug.  The main difference I think is just that it had to be used as part of research and couldn't just be prescribed.  But once you have FDA approval for any reason then a doctor can prescribe it outside of a research setting even in a so-called off-label use.

 

HC, because it is FDA approved for Lupus, can be prescribed for COVID now if you can convince your doctor to do it.  Remdesivir is still only investigational, so you can't do that yet.  

As someone with lupus, I cannot stress enough that you do NOT want to be taking Plaquenil unless you have no other choice. You really really don’t

[The Marquis de Leech]

People here seem to be treating Lockdown Level 3 as the end of Lockdown. Which it really isn't...

[Jeor]

2 hours ago, The Marquis de Leech said:

People here seem to be treating Lockdown Level 3 as the end of Lockdown. Which it really isn't...

Yeah, I hope Australia and NZ don't undo the good work by everyone rushing out. I don't think they will, but the chances of a second wave are not negligible. While we haven't seen much community transmission, there are undoubtedly some mild/asymptomatic cases still out there, and in the hands of a super-spreader it's possible a second wave could be reignited. I think we're going to have to stick with social distancing for a long time yet, though hopefully with more shops and services open.

If they're allowed to open, it'll be a difficult thing for restaurants to figure out, mind you. Do you still open, knowing that you're only get half the revenue you got before (and potentially much worse?). Arts organisations are also going to continue to be decimated under extended social distancing.

 

[Raja]

There was also a Lancet study released on Remdesivir yesterday, this is the WHO study that had leaked early but now the full paper is out. I think this blog gives a very decent review that is easy to understand for non-medical folk, I'd recommend reading the blog as opposed to just the tweet!

Hopefully we get to read the NIAID study soon.

 

[Rippounet]

Quote

 

https://www.theguardian.com/society/2020/apr/29/more-cases-of-rare-syndrome-in-children-reported-globally

Doctors around the world have reported more cases of a rare but potentially lethal inflammatory syndrome in children that appears to be linked to coronavirus infections.

Nearly 100 cases of the unusual illness have emerged in at least six countries, with doctors in Britain, the US, France, Italy, Spain and Switzerland now reported to be investigating the condition.

The first cases came to light this week when the NHS issued an alert to paediatricians about a number of children admitted to intensive care units with a mix of toxic shock and a condition known as Kawasaki disease, an inflammatory disorder that affects the blood vessel, heart and other organs. So far 19 children have been affected in the UK and none have died.

The French health minister, Olivier Veran, said on Wednesday that the country had more than a dozen children with inflammation around the heart, and while there was insufficient evidence to prove a link with coronavirus, he said the cases were being taken “very seriously.”

 

[The Anti-Targ]

The interesting thing about testing Remdesivir on patients already at a stage of needing hospitalisation is that this tells us nothing about whether or not the drug might be more effective while symptoms are mild. The animal study cited at the beginning is later described in more detail at the end, noting that the animal trial used the drug on animals early on in the course of infection, sometimes within hours of being exposed to the virus. So it seems the mechanism of action (reducing virus replication) only works when viral load is lower. So the value of Remdesivir might be in contact tracing. Somone comes down with COVID-19, catch that person early in their symptomatic phase and maybe they won't get too sick, but track down as many people as possible who were in contact with the patient within say 4 days of the person getting symptoms and give those in contact people Remdesivir for maybe a week and perhaps they never get sick and even never become infectious. The the probability is that those pre-symptomatic people may have never got enough of a virus hit for any kind of immune response to occur and they they remain as susceptible to a second exposure as their first.

I don't know if it's more expensive to give prophylactic anti-virals to in-contact people to do serial testing first to see if they actually get infected and only treat them if they test positive for the virus. I heard that a research facility in Kenya is trying to develop a breathalyzer test for the virus, if that can be done cheaply with high sensitivity then it might be a good diagnostic tool, esp if Remdesivir does prove to be an effective treatment early in the infection as it appears to have been proven to be in animal studies.

Scratch the studies for use in severely ill patients and get on to studying patients more recently infected and with mild symptoms.

Quote

In prior animal studies of remdesivir, investigators felt that early administration of remdesivir would be critical to the drug’s efficacy.  Thus, remdesivir was used either prophylactically or within 12-24 hours of virus inoculation. 

Of course, in real-world clinical practice, remdesivir was given at a far later timepoint:

• There is a time delay between infection and the development of symptoms (the incubation period).
• There is a subsequent delay between the development of symptoms and presentation to the hospital (a median of ten days in this study).

This leaves open the possibility that earlier administration of remdesivir could be more effective.  In the subgroup of patients treated within <10 days of symptoms, there was a non-significant trend towards faster clinical improvement in patients treated with remdesivir.  However, subgroup analysis still found no differences in viral load.

Up to 10 days after symptoms develop is very far into the infection compared to the prophylactic or use within 24hrs of exposure in animals. 9 days after symptoms first appear could be up to 23 days after infection

[Zorral]

https://www.nytimes.com/2020/04/30/opinion/coronavirus-warm-weather-mutation.html?

Will Warm Weather Slow Coronavirus?
It’s hard to know yet. But there are many things we can do to flatten the next wave of the contagion.

By John M. Barry, author of “The Great Influenza: The Story of the Deadliest Pandemic in History” and a professor at the Tulane University School of Public Health and Tropical Medicine.

Quote

 

[....]
Mutation was also an important factor. It probably accounts for the timing of the third wave in 1919: It seems likely that by then, the virus had changed enough that any immunity to the initial virus didn’t protect well against its mutated form. This hypothesis is supported by the fact that exposure to the first wave provided up to 89 percent protection against second wave illness (the best vaccine in the last 15 years provided 62 percent protection), but neither first nor second wave exposure protected against that third wave.

What does all this mean now?

Nothing is certain and little is known about Covid-19, but a few things are likely.

First, modelers estimate that the true number of infected persons is up to 20 times the reported number, which still leaves about 95 percent of the population susceptible. If, as in 1918, susceptibility proves more important than seasonal influences, hot weather will not give as much relief as hoped for. By the same token, that would mean the expected seasonal surge when colder weather arrives might not be as large as feared.
[....]

 

 

[Which Tyler]

FT have now included excess deaths in their stats, not just UK:
https://www.ft.com/coronavirus-latest

 

[Fragile Bird]

19 minutes ago, Which Tyler said:

FT have now included excess deaths in their stats, not just UK:
https://www.ft.com/coronavirus-latest

 

That article has a chart showing more than 10,000 excess deaths in Ecuador, yet they "only" have reported 24,675 coronavirus cases and 883 deaths.

[Buckwheat]

We are allowed to go to other communes/municipalities as of today, but there is no detailed information about it - how far does that mean (just to the next city or across the country), for which purposes etc. There has been some weird contradicting information about what public services are going to open when, for example libraries and museums. But things are slowly opening again.

[Which Tyler]

17 minutes ago, Fragile Bird said:

That article has a chart showing more than 10,000 excess deaths in Ecuador, yet they "only" have reported 24,675 coronavirus cases and 883 deaths.

It's one of the reasons "excess deaths" is the stat that matters. You can't politic your way around it by only counting those who've been tested, or those who died in hospital etc etc (not to mention that it also counts deaths by secondary causes such as the lack of ICU for a stroke victim etc). Not perfect (relies on the average from previous years, doesn't take into account an expected reduction for other causes due to lockdown etc) of course, but way better than anything else.

[Fragile Bird]

4 minutes ago, Which Tyler said:

It's one of the reasons "excess deaths" is the stat that matters. You can't politic your way around it by only counting those who've been tested, or those who died in hospital etc etc (not to mention that it also counts deaths by secondary causes such as the lack of ICU for a stroke victim etc). Not perfect (relies on the average from previous years) of course, but way better than anything else.

Oh, I know, I know, but the difference is so huge!

I'm just waiting to see real news out of Russia. Rioting in the streets, people are hungry, more than 100,000 cases and only a 1,000 deaths.

[Mr. Chatywin et al.]

5 minutes ago, Fragile Bird said:

Oh, I know, I know, but the difference is so huge!

I'm just waiting to see real news out of Russia. Rioting in the streets, people are hungry, more than 100,000 cases and only a 1,000 deaths.

My half assed guess is that more than half the world cannot be tested with any reliability. I've used South America as an example many times before. And horror stories are starting to come out of Africa. 

[The Anti-Targ]

8 hours ago, Which Tyler said:

It's one of the reasons "excess deaths" is the stat that matters. You can't politic your way around it by only counting those who've been tested, or those who died in hospital etc etc (not to mention that it also counts deaths by secondary causes such as the lack of ICU for a stroke victim etc). Not perfect (relies on the average from previous years, doesn't take into account an expected reduction for other causes due to lockdown etc) of course, but way better than anything else.

And also needs to take account of possible / probable reduced death from causes associated with a normal pattern of life (traffic accidents). I would like to get a bit of a sense of murder rates too (not that these are a significant cause of deaths in most countries). But reports are saying domestic and sexual violence is increasing, but is that translating into increased murders or is the murder rate going down because there is less overall criminality of the type that tends to lead to murder?

ref the spread of Covid 19, if you want to get a rough sense of what your country's R0 is doing keep an eye on your active cases graph. So long as testing is remaining fairly consistent in its application so that there is no change in the testing programme's ability to detect new cases, if the number of active cases is rising then the R0 is >/=1. If the number of active cases is consistently decreasing then the R0 will be <1. You can't know what the R0 is from looking at graphs but you can tell if your country is > or < 1. There will be a lag given the mean 6 day incubation period. So the trend today reflects the R0 1 to 2 weeks in the past so to be certain there is a sustained R0<1 there would probably need to be at least a week or more of consistent decrease in active cases.

So, as of 2 weeks ago for the top 10 infected countries the R0 is >/=1 for: USA, UK, Russia, Brazil (may be a R0 approaching 2 as new cases is on an exponential curve and active cases is very steep, almost exponential). The R0 is <1 for Germany and Iran. The R0 is turning a corner (ie. probably <1 but too early to tell) for: France (active cases is flat), Turkey (has had 6 days of active cases reducing so it's looking promising), Italy (active cases has been decreasing for almost 2 weeks but very very slowly), Spain (active cases are dropping but only for about 6 days).

I would say beware if your country starts to relax control measures if the active case graph is still trending upwards. If your active case graphs has been dropping for at least 2 weeks then cautious relaxation of control measures is probably justified. And if your country has recently relaxed control measures wait 2 at least weeks before you decide whether the relaxed control measures has not lead to an increase in R0.

[williamjm]

7 minutes ago, The Anti-Targ said:

So, as of 2 weeks ago for the top 10 infected countries the R0 is >/=1 for: USA, UK, Russia, Brazil (may be a R0 approaching 2 as new cases is on an exponential curve and active cases is very steep, almost exponential). The R0 is <1 for Germany and Iran. The R0 is turning a corner (ie. probably <1 but too early to tell) for: France (active cases is flat), Turkey (has had 6 days of active cases reducing so it's looking promising), Italy (active cases has been decreasing for almost 2 weeks but very very slowly), Spain (active cases are dropping but only for about 6 days).

I think it can be too dependent on testing frequency to be reliable. The UK is now (after many delays) testing more than twice the number it was testing a week before which makes it difficult to compare active cases graphs. It might work better for countries where the number of tests being done was roughly constant.

In random statistics, the Worldometer figures are now showing more than 1 million recoveries worldwide (there will be a lot more, particularly since some countries like the UK aren't providing recovery statistics).

South Korea seem to be continuing to do well, they apparently reported no new local cases for the first time in months (4 cases were apparently reported at airports). The time when they were one of the worst-hit countries by the pandemic seems a long time ago now.

[The Anti-Targ]

1 hour ago, williamjm said:

I think it can be too dependent on testing frequency to be reliable. The UK is now (after many delays) testing more than twice the number it was testing a week before which makes it difficult to compare active cases graphs. It might work better for countries where the number of tests being done was roughly constant.

In random statistics, the Worldometer figures are now showing more than 1 million recoveries worldwide (there will be a lot more, particularly since some countries like the UK aren't providing recovery statistics).

South Korea seem to be continuing to do well, they apparently reported no new local cases for the first time in months (4 cases were apparently reported at airports). The time when they were one of the worst-hit countries by the pandemic seems a long time ago now.

It is definitely not reliable for any kind of precise assessment of where R0 is. But to give a rough indication of whether your R0 is above or below 1, as long as testing is not being scaled back so that new infection is less likely to be picked up there is still value in looking at the trend of active cases to get a sense of it. Ramping up testing will artificially increase the apparent R0 but not by a substantial amount. If your R0 is just below 1 it will make the graph flatten out rather than trend downwards until there is a period of adjustment (probably 2+weeks), but it is unlikely to make the graph start rising by much. If your R0 is substantially below 1 then the active cases graph should still remain on a downward trajectory. In the UK's case the rate of increase in active cases has remained linear after the increased testing, this means that while the graph may have started to flatten if testing rate had remained the same it would still be trending upwards and so the R0 remains roughly at 1 or a smidge above, as of 7-14 days ago.

On Hydroxychloroquine and Chloroquine, it appears to me that the more robust the scientific method applied to investigating it's usefulness vs COVID-19 the less beneficial it appears to be.

[ljkeane]

8 hours ago, The Anti-Targ said:

If your R0 is substantially below 1 then the active cases graph should still remain on a downward trajectory. In the UK's case the rate of increase in active cases has remained linear after the increased testing, this means that while the graph may have started to flatten if testing rate had remained the same it would still be trending upwards and so the R0 remains roughly at 1 or a smidge above, as of 7-14 days ago.

Er what? The UK isn't reporting recoveries so there isn't really a publicly available measure of 'active cases'.

I assume it's probably a bit of a guesstimate but apparently the team trying to model the UK's R value at Imperial College, presumably with better access to data, currently have it at 0.7.

[Prue]

8 hours ago, The Anti-Targ said:

On Hydroxychloroquine and Chloroquine, it appears to me that the more robust the scientific method applied to investigating it's usefulness vs COVID-19 the less beneficial it appears to be.

Remdesivir seems to be more promising, with far fewer side effects.

[rotting sea cow]

10 hours ago, The Anti-Targ said:

And also needs to take account of possible / probable reduced death from causes associated with a normal pattern of life (traffic accidents). I would like to get a bit of a sense of murder rates too (not that these are a significant cause of deaths in most countries). But reports are saying domestic and sexual violence is increasing, but is that translating into increased murders or is the murder rate going down because there is less overall criminality of the type that tends to lead to murder?

ref the spread of Covid 19, if you want to get a rough sense of what your country's R0 is doing keep an eye on your active cases graph. So long as testing is remaining fairly consistent in its application so that there is no change in the testing programme's ability to detect new cases, if the number of active cases is rising then the R0 is >/=1. If the number of active cases is consistently decreasing then the R0 will be <1. You can't know what the R0 is from looking at graphs but you can tell if your country is > or < 1. There will be a lag given the mean 6 day incubation period. So the trend today reflects the R0 1 to 2 weeks in the past so to be certain there is a sustained R0<1 there would probably need to be at least a week or more of consistent decrease in active cases.

So, as of 2 weeks ago for the top 10 infected countries the R0 is >/=1 for: USA, UK, Russia, Brazil (may be a R0 approaching 2 as new cases is on an exponential curve and active cases is very steep, almost exponential). The R0 is <1 for Germany and Iran. The R0 is turning a corner (ie. probably <1 but too early to tell) for: France (active cases is flat), Turkey (has had 6 days of active cases reducing so it's looking promising), Italy (active cases has been decreasing for almost 2 weeks but very very slowly), Spain (active cases are dropping but only for about 6 days).

I would say beware if your country starts to relax control measures if the active case graph is still trending upwards. If your active case graphs has been dropping for at least 2 weeks then cautious relaxation of control measures is probably justified. And if your country has recently relaxed control measures wait 2 at least weeks before you decide whether the relaxed control measures has not lead to an increase in R0.

Why in the seven hells the number of cases in Spain and Italy is going down so slowly despite strong quarantine measures??? Germany and Austria with far fewer restrictions reported a decrease of cases at the expected scale. 

Where do these cases in Italy coming from?

[Jeor]

21 minutes ago, rotting sea cow said:

Why in the seven hells the number of cases in Spain and Italy is going down so slowly despite strong quarantine measures??? Germany and Austria with far fewer restrictions reported a decrease of cases at the expected scale. 

Where do these cases in Italy coming from?

I also scratch my head thinking that Italy and Spain should have been coming down quicker than this. My only thought is that demographics might be playing a role. Not just that Italy and Spain might have older populations, but I get the impression that perhaps they have populations with generally higher risk factors (smoking, high blood pressure?) or greater household size or population density? Just guesses.