covid-19

[RandBlade]

Using Lewk's data there has been a 10% increase in US fatalities: far, far more than any of the other nations outside of South America at the bottom of the table. Depressing.

And I'm far from convinced that we are comparing like-for-like. The UK is counting amongst its COVID deaths anyone who dies after being tested positive for any reason at all. Even if they die months later due to being hit by a bus. George Floyd's death would have been registered as a COVID death in this country (not that it would have happened here). I doubt whether the same is the case in the USA.

[Timbuk2]

Haven't posted the scoreboard recently:

Belgium 857.55
United Kingdom 678.59
Spain 608.17
Italy 579.45
Sweden 549.24
France 448.58
USA 422.33
Peru 394.35
Chile 389.23
Brazil 366.11

Needs to be cited. This data is either completely incorrect, woefully out of date, or using some weird metric. At the very least just numbers like this without any context cannot be interpreted.

The true picture, as of yesterday (19th July), is as follows:

Cases per capita:



Deaths per capita:



Cumulative cases:



Cumulative deaths:



Rolling average cases:



Rolling average deaths:



So the US may not have had the highest overall per capita infection rate since the virus first went global, but it has certainly fucked up its response.

European countries have had very high rates when the virus first appeared and not much was known about it, but have since controlled it and cases have fallen significantly.

Whereas the US has not learned a thing, it has the highest infection rate of those countries as of today, and the highest number of new cases daily. For supposedly such a capable country with the resources at its disposal and the time that has now passed, this is absolutely pathetic.

[Aimless]

The numbers are from statista. Curiously, he doesn't show figures for covid deaths or excess all-cause mortality.

[RandBlade]

The numbers are from statista. Curiously, he doesn't show figures for covid deaths or excess all-cause mortality.

What will need to be looked at to make reasonable comparisons is net excess mortality across the year. Its going to take a long time for that data to come in though, as it stands many European nations are probably running at negative excess mortality whereas I suspect at the moment the US is running at excess mortality but we can't see those figures yet.

What should not be done in making reasonable comparisons is look at the excess part of the graph alone and 'lock in' any excess above average while ignoring anything below the average.

[Aimless]

What will need to be looked at to make reasonable comparisons is net excess mortality across the year. Its going to take a long time for that data to come in though, as it stands many European nations are probably running at negative excess mortality whereas I suspect at the moment the US is running at excess mortality but we can't see those figures yet.

What should not be done in making reasonable comparisons is look at the excess part of the graph alone and 'lock in' any excess above average while ignoring anything below the average.

Why do you believe "negative" excess mortality is particularly interesting in assessing impact of and response to the pandemic? If people who might've died in September die in March, that's not a mitigating circumstance. You might see fewer deaths from accidents as well, but, again, how important is that in the grand scheme of things?

[RandBlade]

Why do you believe "negative" excess mortality is particularly interesting in assessing impact of and response to the pandemic? If people who might've died in September die in March, that's not a mitigating circumstance. You might see fewer deaths from accidents as well, but, again, how important is that in the grand scheme of things?

There will always in an average year be swings and roundabouts in the data, variability, that is why you use a net average, it smooths out the data.

By only counting the data above the line you are only looking at half the data, getting the swings but not the roundabouts.

What you are proposing would flatter Trump and the USA. Because they've consistently been above the line . . . but dragged out over a period of time . . . they have nothing to net off as they've never gone negative. But in other nations that have had peaks and troughs (South Africa springs to mind) the picture is different. In South Africa they had major negative data (disproportionately negative) when they went into lockdown before the COVID deaths started coming.

Net shows the whole picture. Only looking at one side of it is disingenuous.

[RandBlade]

Another way of putting it Aimless, lets say there are two examples (purely hypothetical)

Country A has 10 deaths per million above the line in part of the year, but negative 6 excess deaths per million below the line, so net over the year 4 deaths per million extra occurred.

Country B has 10 deaths per million above the line, never went negative, so overall finished the year with 10 deaths per million extra.

I would suggest B has done over the span of the year 2.5x worse than A. Country B would have 2.5x the net excess deaths by the end of the year than A has. But if you only look at the above the line period you would suggest they have done identically the same, despite A having had 60% fewer net deaths than B.

[wiggin]

RB, they are already guesstimating excess mortality numbers for countries with decent reporting. On a rolling basis it ain't negative:

https://www.ft.com/content/a26fbf7e-...3-955839e06441

Now you could imagine (as you do) that mortality will be a bit lower in the future because we already killed some of our oldest people. Which might be true (albeit in a macabre way) but the timing of a death matters as well; in QALYs if nothing else. You're essentially making the argument that it's not as bad as it looks because most of those people were going to die eventually anyways - which, in the long run, is true of all of us, I suppose.

So the US may not have had the highest overall per capita infection rate since the virus first went global, but it has certainly fucked up its response.

European countries have had very high rates when the virus first appeared and not much was known about it, but have since controlled it and cases have fallen significantly.

Whereas the US has not learned a thing, it has the highest infection rate of those countries as of today, and the highest number of new cases daily. For supposedly such a capable country with the resources at its disposal and the time that has now passed, this is absolutely pathetic.

What's fascinating (for good or bad) about the US is how different states and regions have very different looking curves. There are some nice comparators out there, but my individual state's curves look like they're from an epidemiology textbook - perhaps, in fact, because most of the authors of said textbooks live in my state and that governor actually listened to them. It just so happened that most of the earliest hit states (especially NY and MA) had decent public health capacity and decision making so they were able to arrest the epidemic - albeit at a huge human and economic cost, especially since they had less preparation and knowledge back in March when it really took off.

What you're seeing in the US is a superposition of many different regional curves that have had very different kinetics and magnitudes over the duration of this epidemic. It actually lends some (very minor) credence to the WH position that some of the decision-making should be done on a regional or state level rather than a federal level (not to discount the truly horrendous effects of the federal mismanagement of this crisis).

[Aimless]

For all practical purposes, covid doesn't save lives. Comparing excess mortality relative to average for specific time periods across the year helps answer questions about the impact of the covid pandemic; netting that against "negative" mortality averaged out over the year obscures information about the impact of the covid pandemic. The approach you describe has been used by some commentators to make Sweden's pandemic response look better than it is, by including mortality stats from before our outbreak in the calculations of excess mortality. The issue of natural fluctuation in mortality is something that can be addressed, and is usually addressed by comparing specific periods in a year with the average for the corresponding periods in previous years—not the averages for entire years. That way, you can glean more useful information about the impact of the pandemic. The approach you propose is the one favored by our ministry of public health, because they believe it will make them look better than they would've otherwise. However, it doesn't adequately answer the question that the investigation of excess mortality is supposed to answer, in this debate, namely: approximately how many people have died of covid in settings where we don't have 100% test coverage? The two approaches have two completely different objectives.

[RandBlade]

To be clear I'm not suggesting that it is OK that someone dies earlier in the year from COVID because they may have died later in the year anyway, that is not my point at all. Even if someone may have died anyway then simply dying peacefully from old age versus dying by drowning in your own fluids by COVID as you struggle to breath are not the same.

What I am saying - and you subconsciously made the same point with regards to "for specific time periods across the year" - is that in making comparisons between countries we need to look like-for-like over the same time period. EG if we want to look at say March to August then look at March to August consistently. If you're going to make comparisons then like-for-like comparisons over the same time period are appropriate was my point.

To take the example given earlier about someone dying in March who might have died in October, that is a tragedy in March. Now if we look at 2 different countries both of which have an excess death in March, but one has a negative one death in October (because the person who died earlier in the year didn't die then) while the second country has the same number of deaths as normal in October despite the fact the October person had already died in March . . . because this was a genuine excess death on its own that shouldn't have happened in October because country B hasn't stopped the pandemic yet . . . then it is reasonable to contrast that and say that B is suffering more still is it not?

If you "lock in" the death figures by looking at different time periods for different nations then you are distorting the figures. If you wish to compare how did nations A, B, C and D do over time period x to y then look at time period x to y for all 4 nations - don't look at their shaded above the line graph for each of them. Doing that flatters the countries that halted the virus later.

[RandBlade]

Twitter Link

Some good news!

[RandBlade]

Ironically the virus has been suppressed so much within the UK that the domestic later phases of the Oxford vaccine may struggle to demonstrate efficacy (since there's not going to be much infection within the placebo group either) so we need to see the results from the trial Oxford are taking in Brazil and elsewhere to see if this works fully before it can be delivered to people.

AstraZenica are so convinced this is going to work though they've already started producing the vaccine while the Phase III trial runs in Brazil so that if it does work the vaccine will be produced and ready to go the moment it gets a licence.

The UK has ordered 100 million doses of the vaccine. Not sure why we need so many.

[ar81]

After posting an increase of over 15,000 Florida now holds the record for most positive cases in a single day. We're winning!

Protests/riots killed mitigation.
Source: https://www.nber.org/papers/w27408.pdf
President does not kill Americans. People do.
Leftists are killing you with COVID.

[Ominous Gamer]

Protests/riots killed mitigation.
Source: https://www.nber.org/papers/w27408.pdf
President does not kill Americans. People do.
Leftists are killing you with COVID.

"Furthermore, we find no evidence that urban protests reignited COVID-19 case growth during the more than three weeks following protest onset. We conclude that predictions of broad negative public health consequences of Black Lives Matter protests were far too narrowly conceived."

Translated:
People protested with a concern for covid spread, and the protests caused all the scared white people who would have otherwise been out and about to stay home.

[RandBlade]

"Furthermore, we find no evidence that urban protests reignited COVID-19 case growth during the more than three weeks following protest onset. We conclude that predictions of broad negative public health consequences of Black Lives Matter protests were far too narrowly conceived."

Translated:
People protested with a concern for covid spread, and the protests caused all the scared white people who would have otherwise been out and about to stay home.

Indeed. That was a fascinating paper to read - clearly ar81 hadn't read it LOL!

[Loki]

Ironically the virus has been suppressed so much within the UK that the domestic later phases of the Oxford vaccine may struggle to demonstrate efficacy (since there's not going to be much infection within the placebo group either) so we need to see the results from the trial Oxford are taking in Brazil and elsewhere to see if this works fully before it can be delivered to people.

AstraZenica are so convinced this is going to work though they've already started producing the vaccine while the Phase III trial runs in Brazil so that if it does work the vaccine will be produced and ready to go the moment it gets a licence.

The UK has ordered 100 million doses of the vaccine. Not sure why we need so many.

If it's like the vaccine the US is working on, you need 2 doses per person.

[RandBlade]

Twitter Link

I think the Moderna vaccine and the Oxford vaccine are quite different and work through very different means . . . which is a good thing as since those are the two most advanced vaccines that they are different surely increases the chance that at least one of them will work? If the two shared some critical feature that failed it would rule out both at once, but looks optimistic that we should get a successful vaccine before long.

[Loki]

Stage 3 of vaccine testing generally lasts at least several months. It also takes months to produce a hundred million vaccines.

[Aimless]

Protests/riots killed mitigation.
Source: https://www.nber.org/papers/w27408.pdf
President does not kill Americans. People do.
Leftists are killing you with COVID.

Read the fucking paper you ridiculous toon.

[BalticSailor]

If you want to know how the world is doing vaccine-wise, I suggest reading Derek Lowe's blog In The Pipeline. The latest update covering the general state of global vaccine development efforts is here: https://blogs.sciencemag.org/pipelin...update-june-29

He has later blog entries devoted to new information regarding individual vaccine candidates, as it becomes available, as well as articles on general issues in vaccine development, such as production bottlenecks, various approaches and technologies (worth a read if you want to catch up on differences between various vaccine candidates researched currently, their advantages/disadvantages and likelihood of success, etc). The blog has categories list - obviously, "covid-19" is relevant here, as he blogs about completely unrelated stuff, too.

[RandBlade]

Stage 3 of vaccine testing generally lasts at least several months. It also takes months to produce a hundred million vaccines.

Not sure fully sure what Moderna are doing with their scheme but the Oxford team aren't messing around with this and are getting backed to the hilt by our government. The Phase III trial is already underway and rather than waiting for that AstraZeneca have already started producing hundreds of millions of vaccine doses. Our government have paid for a hundred million vaccine doses already for the UK and I believe AstraZenica have similar arrangements with other countries already too.

The logic is that at this moment money is virtually no issue and there is no point waiting for the results and then starting production. If this works and it gets licensed then the doses will already be produced and ready to go pretty much immediately. If it fails then the vaccines that have been produced can be binned but that cost will be inconsequential compared to the cost of the damage this virus is doing to the economy.

AstraZeneca could have the first doses ready to go by September: https://www.astrazeneca.com/media-ce...w-vaccine.html

[wiggin]

RB, I wish AZN and Moderna and all the rest the best of luck, but you should temper your optimism. So far all we've seen is very early data that safety profiles look okay and the vaccine has some desirable short term effects on the immune system. However, vaccine development is in no ways straightforward and the probability of any given approach being particularly successful (especially on the timelines we're working with) is unfortunately low.

The pharma community is currently trying to develop a vaccine so quickly that it would absolutely shatter existing records for how quickly mass-produced vaccines are brought to market. If it works, that would be extraordinary - both in the sense of being good and in the sense of being unusual.

It is not even clear whether this disease is a good target for vaccination, let alone whether the specific different approaches in development will be effective. Let us be cautiously optimistic, certainly, but not unreasonably so.

edit: I also want to mention that we've heard the most about the Oxford/AZN and Moderna stories because (a) they're Western and (b) they're attention hounds (Oxford because they're academics and Moderna because the company is all about buzz). There are very serious efforts underway by other companies but they've been a lot more circumspect with sharing their early (and often wildly preliminary) results - JNJ, Merck, Pfizer, Sanofi are just some of the deep pocketed big players. I wouldn't count them out just because they've been relatively quiet.

[RandBlade]

Oh I agree that we should temper our enthusiasm but two points. Firstly today's news was positive so its worth discussing as news. Secondly and specifically about timing I was replying to Loki's comments about timing and how it takes months to produce the vaccine . . . it does indeed but they've already begun so the vaccine will be ready for delivery from September.

That doesn't guarantee of course the vaccine will work but it does mean that the usual timeline of how long it takes to reach distribution will be reduced from normal when it first needs to get approval and then production starts. The few months delay for production that Loki referred to is eliminated.

If it doesn't work it doesn't work, as I said then it goes in the bin. C'est la vie. But if it does work it will be ready in months not years.

As for those two they're the only ones I know of not from China that have reached Phase III trials already so its not simply that they're getting publicity but they are genuinely further down the track as far as I can tell. Doesn't mean again that they'll be approved in the end but its reassuring to have them in Phase III trials already.

PS the Oxford team admit by themselves they're somewhat lucky, in as much as you can call anyone lucky right now. They were already working on a SARS vaccine when this hit and that research has given them a headstart into producing a vaccine for this.

[Hazir]

I'm wondering why we don't hear a peep out of India.

[wiggin]

Oh I agree that we should temper our enthusiasm but two points. Firstly today's news was positive so its worth discussing as news. Secondly and specifically about timing I was replying to Loki's comments about timing and how it takes months to produce the vaccine . . . it does indeed but they've already begun so the vaccine will be ready for delivery from September.

That doesn't guarantee of course the vaccine will work but it does mean that the usual timeline of how long it takes to reach distribution will be reduced from normal when it first needs to get approval and then production starts. The few months delay for production that Loki referred to is eliminated.

If it doesn't work it doesn't work, as I said then it goes in the bin. C'est la vie. But if it does work it will be ready in months not years.

As for those two they're the only ones I know of not from China that have reached Phase III trials already so its not simply that they're getting publicity but they are genuinely further down the track as far as I can tell. Doesn't mean again that they'll be approved in the end but its reassuring to have them in Phase III trials already.

PS the Oxford team admit by themselves they're somewhat lucky, in as much as you can call anyone lucky right now. They were already working on a SARS vaccine when this hit and that research has given them a headstart into producing a vaccine for this.

Yes, most pharma companies with decent candidates are ramping up production, often with government support (it's actually a rather more interesting list of which companies are willing to drop a billion or so dollars on ramping up production without government support). It's common sense.

Regardless, I'm not sure you should dismiss the 2 Chinese vaccines in Phase III trials, nor should you dismiss the Pfizer/BioNTech effort which is entering phase III within a week or two (about the same timeline as Moderna), or the other advanced work by e.g. JNJ. I think the real issue here is that people think that certain milestones (like an early Phase II data readout or entering Phase III studies) are super significant if they happen a few weeks earlier than another effort. This is not a sprint. There are many reasons why a slightly slower development program might end up working better as a usable vaccine in the real world - manufacturability, or duration/extent of immunity, or safety profile.

Most of the high profile efforts to date in the West (including all three I've mentioned) have used existing successful vaccination strategies (for a number of reasons), preferring newer methods that do not have a lot of clinical experience backing them up. There's reasons for doing so, and I hope they're successful. But maybe some older techniques will yield better results in the long run - or maybe none of the current approaches will work particularly well.

This is a very, very difficult problem with an extremely limited timeframe in which to solve it. It is in no way obvious that current efforts will be successful, or that a single company's reasonably decent Phase II early results are a game changer. We should note positive progress, certainly, but let's not make any plans for Christmas.

[RandBlade]

I dismiss the Chinese vaccines because I don't trust China. They're probably ahead of us on this and probably using vaccine trials on prisoners and/or the Uigher or something similar but that's a different story. Hence why I'm looking at only like-for-like comparisons.

I thought Moderna were in Phase III already like the Oxford one. Good news that Pfizer are going into Phase III too.

I agree its not a sprint but its good news to have vaccine trials in Phase III. A friend of mine in the industry gave a rule of thumb indication to me of how likely it is for vaccines to go from trial to being successful and approved . . . he put it at going into Phase I as 10% but going into Phase III as 70%. To have 3 different vaccines going into Phase III already then is entirely good news given that we've never before had a human coronavirus vaccine before (though to be fair animal coronavirus vaccines do exist and there's never been such a pressing need for a human one before).

[wiggin]

I dismiss the Chinese vaccines because I don't trust China. They're probably ahead of us on this and probably using vaccine trials on prisoners and/or the Uigher or something similar but that's a different story. Hence why I'm looking at only like-for-like comparisons.

I thought Moderna were in Phase III already like the Oxford one. Good news that Pfizer are going into Phase III too.

I agree its not a sprint but its good news to have vaccine trials in Phase III. A friend of mine in the industry gave a rule of thumb indication to me of how likely it is for vaccines to go from trial to being successful and approved . . . he put it at going into Phase I as 10% but going into Phase III as 70%. To have 3 different vaccines going into Phase III already then is entirely good news given that we've never before had a human coronavirus vaccine before (though to be fair animal coronavirus vaccines do exist and there's never been such a pressing need for a human one before).

Rules of thumb are, well, not super useful here. Yes, in general a drug in Phase III trials has a decent chance of making it to market. But two points: these are not normal circumstances, and a whole lot of the preclinical and phase I/II work that would normally have been done has been abridged. Furthermore, we're working with new methods and a new target, so the odds of success are lower than the general case.

It's also important to note that we might have several vaccine candidates that make it to the market and yet still not have good control of the pandemic. Efficacy is a sliding scale here and we might not be able to achieve a sufficient proportion of the population with a sufficiently high level of immunity for a sufficient period of time to be particularly effective.

It's very easy to place our faith in scientific advances, but this is a very tall order in a very short period of time.

[Ominous Gamer]

On metric after metric, the coronavirus pandemic has been worse in the U.S. than nearly any other country
https://www.washingtonpost.com/polit...other-country/

[Aimless]

American conservatives are fucking morons:

Twitter Link

Unbelievably stupid.

[GGT]

Trump resumes covid-19 "briefings": says the pandemic may get worse before it gets better....then says it will disappear.

Dr. Fauci was not invited, and Dr. Birx was hiding behind the curtain.