Coronavirus Statistics: Cases, Mortality, vs. Flu

[Mendel]

Arabic countries have lots of cases, but claim extremely few deaths.

a) Your comparison is skewed, the median European country as per https://www.ecdc.europa.eu/en/cases-2019-ncov-eueea is Finland with 47 deaths per 1000 cases. Iceland has the low end at 5.5.

b) The Arab countries are in the exponential phase of the spread. Deaths lag case reporting by about two weeks, if the case number doubles in a week, you are underestimating by a factor of 4.

c) Demographics plays a huge role, the death count depends highly on how many people over 70 get infected. If you can protect the elderly, you're good; if early spread is among active, mobile people, you'll have few elderly. (Germany used to have a 0.5% CFR early on for those reasons.)

d) I don't know how many women in these counries cover their face for religious reasons, but a Niqab would work like a community mask. It'd be interesting to see infection rates by gender for these countries.

 

[DavidB66]

I think these countries (Saudi Arabia, Bahrain, etc) are in an earlier stage of the epidemic than Europe. Their daily new infections only really started taking off around mid-April. People infected with Covid don't usually die quickly, so there is a time lag of at least a few weeks before a surge of infections turns into a surge of deaths. A better comparison might be with Iran, where the epidemic took hold in early March. The death rate as a proportion of infections there is about 6%. (Incidentally, Iran seems to be having a second wave of infections.)
There are a lot of oddities in the international data which I'm sure will keep epidemiologists in work for years to come. Why is the death rate so much higher in Belgium than the Netherlands, for example? I know the Belgians point out that their figures include all deaths, not just those in hospitals, but that doesn't seem nearly enough to explain the difference. A thought just crossed my mind that the Netherlands has euthanasia, so some of their potential Covid fatalities may already have been helped over the finishing line!
Another possible factor that is now gaining attention is Vitamin D. Since Vitamin D status depends on a combination of diet, exposure to sunlight, and skin pigmentation, it all gets quite complicated. There is a new article on the subject by Fiona Mitchell in The Lancet.

 

[Mendel]

Another possible factor that is now gaining attention is Vitamin D. Since Vitamin D status depends on a combination of diet, exposure to sunlight, and skin pigmentation, it all gets quite complicated.

I only have a third-hand source for this, but I heard that vitamin D insuffiency can be caused by kidney problems. We know that Covid-19 can involve the kidneys, so maybe the vitamin D link may be a matter of a common cause, i.e. kidney problems preciting bad outcomes for Covid-19, and vitamin.D would just be a marker for that and not be directly involved.

 

[qed]

When one considers deaths per million, is the USA really doing that badly? They seem to be doing better than the UK, Spain, Italy, France, Belgium, the Netherlands, Sweden and Ireland?

Is this argument flawed?

 

[Mendel]

When one considers deaths per million, is the USA really doing that badly? They seem to be doing better than the UK, Spain, Italy, France, Belgium, the Netherlands, Sweden and Ireland?
View attachment 40924

Is this argument flawed?

Yes, because you are comparing small countries with an entity the size of a continent. The EU as a whole has ~295/1M deaths, and I believe the "new deaths" numbers are dropping faster than the US's.
Belgium is the worst-hit country in the EU, and it still looks better than New York State. The median country in the EU is Hungary with 49 deaths/1M. The median state in the US is Missouri with 111/1M, and 4 US states are worse off than Spain and Belgium.

Article:

Source: https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases

(The UK is doing about as badly as the US on terms of case numbers not trending down. I think it's because they're getting the same misinformation.)
If you take the US, and subtract NY, NJ, MA and CT from the daily new cases, is the graph going up or down?

P.S.:

Article:

Our estimates suggest that the epidemic is not under control in much of the US: as of 17 May 2020, the reproduction number is above the critical threshold (1.0) in 24 [95% CI: 20-30] states. Higher reproduction numbers are geographically clustered in the South and Midwest, where epidemics are still developing, while we estimate lower reproduction numbers in states that have already suffered high COVID-19 mortality (such as the Northeast). These estimates suggest that caution must be taken in loosening current restrictions if effective additional measures are not put in place.

Source: https://statmodeling.stat.columbia.edu/2020/05/22/new-report-on-coronavirus-trends-the-epidemic-is-not-under-control-in-much-of-the-us-factors-modulating-transmission-such-as-rapid-testing-contact-tracing-and-behavioural-precautions-are-cr/

 

[qed]

I don't think I get that. What does country population size have to do with deaths per million? What is the correlation and why?

And can we say at least say that Europe and and the USA are doing similarly well?

 

[deirdre]

I think it's because they're getting the same misinformation.

large cities skew predominately left politically. so where would they be getting this misinformation? I think now its most likely because people are tired of being cooped up after 2 plus months of lockdowns. and/or they need to work after 2 plus months of lockdowns.

 

[Mendel]

large cities skew predominately left politically. so where would they be getting this misinformation? I think now its most likely because people are tired of being cooped up after 2 plus months of lockdowns. and/or they need to work after 2 plus months of lockdowns.

Article:

According to a new Yahoo News/YouGov poll, 44 percent of Republicans believe that Bill Gates is plotting to use a mass COVID-19 vaccination campaign as a pretext to implant microchips in billions of people and monitor their movements — a widely debunked conspiracy theory with no basis in fact.

The survey, which was conducted May 20 and 21, found that only 26 percent of Republicans correctly identify the story as false.

In contrast, just 19 percent of Democrats believe the same spurious narrative about the Microsoft founder and public-health philanthropist. A majority of Democrats recognize that it's not true.

Source: https://news.yahoo.com/new-yahoo-news-you-gov-poll-shows-coronavirus-conspiracy-theories-spreading-on-the-right-may-hamper-vaccine-efforts-152843610.html

Misinformation hits both sides of the aisle, this particular kind of misinformation just hits Republicans harder. (And that may just be a result of the educution level of the respective constituencies, Republicans are currently appealing more to the less educated.)
It spreads because of the common language: Trump says "we don't have the virus, it will go away", then Bill deBlasio says "we don't have the virus, go ride the subway", and the UK people see that and they're then not willing to take precautions either.

And they're getting it from social media, see my post on the "current events" thread: https://www.metabunk.org/posts/239278/

 

[deirdre]

According to a new Yahoo News/YouGov poll, 44 percent of Republicans believe that Bill Gates is plotting to use a mass COVID-19 vaccination campaign as a pretext to implant microchips in billions of people and monitor their movements

so you think people in the UK and America are exposing themselves to coronavirus because they think Bill Gates wants to microchip them? Is the thinking that if they are naturally immune (antibodies) then they won't need his microchip vaccine? hhmmmm...

Trump says "we don't have the virus, it will go away"

and he did not say that. (speaking of spreading misinformation...)

 

[Mendel]

so you think people in the UK and America are exposing themselves to coronavirus because they think Bill Gates wants to microchip them? Is the thinking that if they are naturally immune (antibodies) then they won't need his microchip vaccine? hhmmmm...

That is the CT I have data on, I assume it correlates with "the epidemic is a hoax", what Cynthia Brehm made the news with yesterday.

and he did not say that. (speaking of spreading misinformation...)

I'm paraphrasing, obviously.

Article:

Jan 22 "We have it totally under control. It's one person coming in from China, and we have it under control. It's going to be just fine."
Feb 10 "Looks like by April, you know, in theory, when it gets a little warmer, it miraculously goes away. I hope that's true. But we're doing great in our country."
Feb 13 "In our country, we only have, basically, 12 cases, and most of those people are recovering and some cases fully recovered. So it's actually less."
Feb 26 "When you have 15 people, and the 15 within a couple of days is going to be down to close to zero, that's a pretty good job we've done."
Mar 5 "It's going to all work out. Everybody has to be calm. It's all going to work out."
Mar 10 "This was unexpected. ... And it hit the world. And we're prepared, and we're doing a great job with it. And it will go away. Just stay calm. It will go away."
Mar 11 "The vast majority of Americans, the risk is very, very low."

Mar 16 "If everyone makes this change or these critical changes and sacrifices now, we will rally together as one nation and we will defeat the virus."

Mar 24 "Easter is a very special day for me. And I see it sort of in that timeline that I'm thinking about. And I say, wouldn't it be great to have all of the churches full?"

Source: https://www.npr.org/sections/goatsandsoda/2020/04/15/835011346/a-timeline-of-coronavirus-comments-from-president-trump-and-who

The npr timeline I have sourced these quotes from contrasts them with the WHO communications over the same time span.

 

[deirdre]

I'm paraphrasing, obviously.

you're reaching. and trying to hard to prove a speculation. you're trying to say something Trump said March 10th is the reason cases in the us and uk are going up NOW, even though on March 10th noone understood the severity of the disease and deaths that would follow. Now people do know. So the question is 'what is the reason cases keep going [up] NOW'. I stand by my original theory.

 

[Mendel]

you're reaching. and trying to hard to prove a speculation. you're trying to say something Trump said March 10th is the reason cases in the us and uk are going up NOW, even though on March 10th noone understood the severity of the disease and deaths that would follow. Now people do know. So the question is 'what is the reason cases keep going [up] NOW'. I stand by my original theory.

Deirdre, you are losing the context, and that's why you misunderstand what I'm trying to say. My original statement was,
(The UK is doing about as badly as the US on terms of case numbers not trending down. I think it's because they're getting the same misinformation.)
I'm saying here that misinformation is a factor in behaving recklessly, and that the US and the UK are exposed to the same sources of misinformation.
You reply, large cities skew predominately left politically. so where would they be getting this misinformation?
I'm making the point that misinfomration is not exclusive to a political orientation, I'm saying that this misinfomration spreads via social media, and I gave examples of misinformation to illustrate the effect it might have: Trump says "we don't have the virus, it will go away", then Bill deBlasio says "we don't have the virus, go ride the subway", and the UK people see that and they're then not willing to take precautions either. Note that I used a politicians of both alignments in my example.
You then challenge my Trump example with and he did not say that.
So I give you quotes where Trump actually said "it will go away", and downplayed the amount of virus spread that was going on in the US at the time.

I am making the case that there are a lot of people out there who believe "the virus is a hoax", and cited Cynthia Brehm, who still believes that. I'm convinced that this belief causes reckless behaviour, and that misinformation strengthens this belief. It comes from many sources. Trump and deBlasio are just examples, there are many others, including right-wing media such as Fox News and others.

-----

Instead of acknowledging that Trump said that, and that your assertion that "he did not say that" was wrong, you are now shifting the argument to something that is even more offtopic, but equally wrong:

on March 10th noone understood the severity of the disease and deaths that would follow

But my NPR source (cited in my previous post!) directly contradicts this assessment because it contains the many severe warnings that the WHO issued. I have also showed another NPR timeline here on this forum.

Article:

March 11, 2020
"Bottom line, it's going to get worse."— Dr. Anthony Fauci

Source: https://www.npr.org/2020/04/21/837348551/timeline-what-trump-has-said-and-done-about-the-coronavirus

On March 9, Prime Minister Giuseppe Conte quarantined all of Italy.
On March 7, a participant of the Red Dawn mailing list wrote:

The outbreak in the US is looking more like Italy but without the aggressive actions [including cordon sanitaire of
50,000 people, closing schools and universites , and canceling mass gatherings taken by Italy as soon as they
identified their first death . I pulled the numbers of cases and deaths reported by themedia at the end of each (
data for today is preliminary /morning data ).
Interesting to compare the two countries and align the outbreaks (4th slide by the date of first reported death). The
US cases include the Princess Diamond cases of repatriated passengers as well as Americans evacuated from
Wuhan. I was unable to estimate the number of tests performed by Italy compared to the US. The US case count
seems to be lagging what Italy observed. The US appears to be about a week behind Italy. Time will tell.

The emails can be downloaded via https://www.nytimes.com/2020/04/11/us/politics/coronavirus-red-dawn-emails-trump.html

Nancy Messonier of the CDC said on February 26:

Article:

I understand this whole situation may seem overwhelming and that disruption to everyday life may be severe. … I told my children that while I didn't think that they were at risk right now, we as a family need to be preparing for significant disruption of our lives."
— The CDC's Messonnier, who warned that the government's containment strategy would likely have to move to mitigation measures such as closing businesses and schools

Source: https://www.npr.org/2020/04/21/837348551/timeline-what-trump-has-said-and-done-about-the-coronavirus

Every competent public health official in the United States understood "the severity of the disease and the deaths that would follow" on March 10th. The data from February's Wuhan outbreak had shown the lethality of the disease and the age groups and other risk factors involved; Italy was showing that it could hit Western countries just as hard as it hit Wuhan.

To inform the public otherwise constituted misinformation. What Donald Trump and Bill deBlasio communicated was misinformation. Provably so.

I don't enjoy telling you that you got easily verifyable facts wrong. I wonder if you tried to verify them before you posted your statements.

 

[deirdre]

Instead of acknowledging that Trump said that, and that your assertion that "he did not say that" was wrong,

it wasn't wrong. trump did not say what you admitted after was a paraphrase.

I'm convinced that this belief causes reckless behaviour, and that misinformation strengthens this belief. It comes from many sources.

and that is fine if that is your opinion. and I stand by my original opinion.

I don't care if you think Trump was/is the only one saying {my words} 'it might go away with warmer weather.' or if you think covid-19 will be with us forever. I already know how you feel. I disagree that old bad information has much to do with new cases NOW. I still think now, a bigger influence is people needing to get back to work or just sick of being cooped up for months on end. I don't understand why it is so horrible if I hold a different opinion to you. quoting me things from 2-3 months ago before the big outbreaks, isn't going to help change my mind.

 

[deirdre]

New info from CDC reported

• 1/3 asymptomatic
• 40% transmit virus to others before they feel sick
• .4% mortality rate for those with symptoms (subject to change)

(CNN) — In new guidance for mathematical modelers and public health officials, the US Centers for Disease Control and Prevention is estimating that about a third of coronavirus infections are asymptomatic.

The CDC also says its "best estimate" is that 0.4% of people who show symptoms and have Covid-19 will die, and the agency estimates that 40% of coronavirus transmission is occurring before people feel sick.

The agency cautions that those numbers are subject to change as more is learned about Covid-19, and it warns that the information is intended for planning purposes. Still, the agency says its estimates are based on real data collected by the agency before April 29.

https://boston.cbslocal.com/2020/05/22/coronavirus-cdc-symptoms-asymptomatic-mortality-rate/

 

[Agent K]

CDC's "best estimate" "based on data received by CDC prior to 4/29/2020"
https://www.cdc.gov/coronavirus/2019-ncov/hcp/planning-scenarios.html

R0: 2.5
Percent of infections that are asymptomatic: 35%
Infectiousness of asymptomatic individuals relative to symptomatic individuals: Same
Symptomatic Case Fatality Ratio: 0.4%
Symptomatic Case Hospitalization Ratio: 3.4%
Percent admitted to ICU among those hospitalized: 0-49 years: 21.9%, 50-64 years: 29.2%, ≥65 years: 26.8%
Percent on mechanical ventilation among those in ICU: 0-49 years: 72.1%, 50-64 years: 77.6%, ≥65 years: 75.5%
Percentage of transmission occurring prior to symptom onset: 40%
Time from exposure to symptom onset: ~6 days (mean)
Time between symptom onset in an individual and symptom onset of a second person infected by that individual: ~6 days (mean)
Time to seek care (outpatient): 3–7 days: 50%
Mean number of days from symptom onset to hospitalization: 6-7 days
Mean number of days of hospitalization among those not admitted to ICU: 4-6 days
Mean number of days of hospitalization among those admitted to ICU: ~10 days
Mean number of days of mechanical ventilation: 5.5 days
Mean number of days from symptom onset to death: ~14 days
Mean number of days from death to reporting: ~7 days

I highlighted some interesting things, like the fact that hospitalized 50-64-year-olds are more likely to be admitted to ICU than those over 65, and deaths are reported after a delay of about a week.
The symptomatic Case Fatality Ratio of 0.4% appears to account for undiagnosed cases, so it's more like an Infection Fatality Rate. Otherwise, if you divide the ~2000 daily deaths by ~30,000 daily reported cases in April, you get a 6.7% CFR.

If the symptomatic IFR is 0.4%, and 35% of infections are asymptomatic and presumably not fatal, then the overall IFR is 0.26%, which would mean that 99K deaths resulted from 38 million infections, which would be 23 times greater than the 1.67 million reported cases. It would mean that 11.5% of the U.S. population has already been infected despite lockdowns, which is a higher percentage than the ~4% I've seen in various serology studies and projections, except in hot spots like NYC, where 27% may have been infected.

 

[Agent K]

• 40% transmit virus to others before they feel sick

The CDC called it "Percentage of transmission occurring prior to symptom onset," so it's more like 40% of infections were caused by pre-symptomatic people.

 

[Mendel]

New info from CDC reported

• 1/3 asymptomatic
• 40% transmit virus to others before they feel sick
• .4% mortality rate for those with symptoms (subject to change)

The CDC is actually using ranges for these models, so it'd be appropriate to say
* 20%-50% asymptomatic
* 0.2%-1% symptomatic fatality rate

The fact that they are actually using these values for models means they're not ruling any of these values out yet.

The 40% value of "transmission before symptom onset" needs to be put in context with some other values.
Let's assume the "best estimate" of R0=2.5, Percentage of infections that are asymptomatic 35%, Infectiousness of asymptomatic individuals relative to symptomatic individuals 100%, from table 1, scenario 5 at https://www.cdc.gov/coronavirus/2019-ncov/hcp/planning-scenarios.html#table-1

Now let's take 100 infected persons. 35 of them are asymptomatic and infect 2.5*35=88 other people. 65 of them are symptomatic and infect 2.5*65=162 other people; 40% of these infections (0.4*162=65) occur before symptom onset, 97 occur afterwards. Note that in this scenario, on average, each symptomatic person infected one other person before they had symptoms!

In this scenario, 153 infections occur from indviduals that don't have symptoms, and only 97 occur from individuals that do. In this scenario, 61% of all transmissions come from individuals who don't feel sick!

These assumptions all apply in the "natural" state where the virus spreads in an unaware population. Once we became aware, for example, infections from symptomatic individuals probably dropped because people are now more careful when they have symptoms. Social distancing helps reduce asymptomatic spread; tracing and isolating contacts of symptomatic cases makes asymptomatic contacts aware that they might have the virus and hopefully keeps them from spreading it.

 

[Mendel]

35% of infections are asymptomatic and presumably not fatal

Death is a symptom.

Your calculations neglect the age demographics. Basically, from the death count you can fairly well estimate how many people in the 65+ age group were probably infected, but for the younger age groups, we just don't know. And the serological studies mostly measure people from those younger age groups, so there's a disconnect there. The upshoot is, you can't take the death figures and compute the "true" number of infected unless you consider the age demographics. (Well, obviously you can, but it won't be reliable.)

The Diamond Princess symptomatic fatality rate for 65+ was 7/200=3.5% at the time of the study https://www.medrxiv.org/content/10.1101/2020.03.05.20031773v2 on March 8 (disregarding subsequent symptoms or deaths). Given that people with worse health stay in care homes and don't go on cruises, I'm doubting that the CDC scenarios with symptomatic fatality rates for that age group of under 3% are realistic.

A difficulty with the asymptomatic rate is that I don't know if an asymptomatic infection always causes a person to develop a specific antibody response. If that does not happen in many cases, the serological surveys won't count these people. These are studies that the Navy would be doing with their recent outbreak in a confined population that has been thoroughly PCR-tested, they should have a detected a large group of asymptomatic cases, and seroconversion in these cases should be observable.

 

[Agent K]

In this scenario, 153 infections occur from indviduals that don't have symptoms, and only 97 occur from individuals that do. In this scenario, 61% of all transmissions come from individuals who don't feel sick!

These assumptions all apply in the "natural" state where the virus spreads in an unaware population. Once we became aware, for example, infections from symptomatic individuals probably dropped because people are now more careful when they have symptoms. Social distancing helps reduce asymptomatic spread; tracing and isolating contacts of symptomatic cases makes asymptomatic contacts aware that they might have the virus and hopefully keeps them from spreading it.

Right, this assumes that 40% of transmission "prior to symptom onset" only counts pre-symptomatic and not asymptomatic transmission, so adding asymptomatic transmission increases that percentage. Recognition of asymptomatic and pre-symptomatic transmission is what convinced public health authorities to advise or require everyone to cover their faces in public even if they don't feel sick.

 

[Mendel]

Recognition of asymptomatic and pre-symptomatic transmission is what convinced public health authorities to advise or require everyone to cover their faces in public even if they don't feel sick.

That's the logic behind social distancing: that almost every spreader has spread the virus while not (yet) knowing they're infected, so we must require everyone to exercise caution because we can only stop the spread if the people who don't know they're infected are cautious.

The highest risk is extended meeting (15 mins) at 6 ft or closer (~5% attack rate under these conditions in a workplace setting has been documented in Munich), which is why reduction in contacts and distance is effective. Hand washing may not contribute much, but it doesn't cost anything and is generally beneficial.
My personal opinion is that the use of community masks or surgical masks is mostly psychological: people want to feel that there is a positive activity they can do to influence the epidemic even if it plays only a minor role.

Ultimately, the safe way out of the lockdowns is via contact tracing, as the WHO has been preaching and many countries and states have been implementing: close contacts of known infected people are precisely the persons who might be engaging in asymptomatic spreading, and if the local health officials can trace and isolate most of them, a lessening of "lockdown" measures does not lead to exacerbated spread. That is why NY state has made the availability of sufficient contact tracing teams a prerequisite to reopening for regions.

In a region where the case numbers are rising, that could indicate that contact tracing isn't working to stop the spread. But it could also mean that contact tracing detects outbreaks more thoroughly, so that will depend on the situation; and cases should go down eventially anyway. So that's the second set of criteria: new cases going down to levels that the contact tracing teams can handle, and are handling successfully.

I believe anything less risks creating second waves (or in some places, first waves).

 

[Agent K]

Top U.S. causes of death between February and May.

 

[qed]

Why are US daily new infections and daily new deaths not following the same trajectory? They did in the initial stage. Is this good news?

 

[Mechanik]

IMHO, it is good news, with a caveat.

As has been discussed, if not here then on other forums, the US started off by testing only those suspected of Covid infections and so the ratio of positive tests to deaths was frighteningly high. As time goes on and testing is expanded to those showing no symptoms, positive tests increase, but deaths no longer track to the same rations as those found in the initial months. The good news is that the death rate from infection is lower than initially calculated. Again, in my humble opinion, your chances for dying from Covid in 2020 are still the same as they were in March, but the published death rates are much more accurate today than they were a few months ago.

The caveat is that the infection is still spreading so additional people will still be infected and die from it. I'm personally encouraged that the limited openings so far have not led to a large surge in deaths even while confirmed infections continue to rise.

 

[Z.W. Wolf]

Death should no longer be the only metric. This is not the flu. It's not a matter of suffering for a while, then recovering.

https://news.berkeley.edu/2020/07/08/fr ... ing-marks/

One thing we didn't anticipate was that the virus seems to accelerate a great deal of scarring in the lungs. And if lung tissue is replaced with scar tissue, it is no longer functional as regular pulmonary tissue, which translates to poor gas exchange. What we really fear is long-term shortness of breath that could extend anywhere from being very mild to severely limiting. There is also a disturbing report looking at computerized tomography (CT) scans of asymptomatic people that found they were left with some scar tissue. So, this could even be happening on a subclinical level.

Another area is the heart. There is evidence now that the virus can directly attack heart muscle cells, and there's also evidence that the cytokine storm that the virus triggers in the body not only damages the lungs, but can damage the heart. We don't know what the long-term effects of that may be, but it could be that we will have a population of people who survive COVID-19 only to go on and have chronic cardiac problems.

The third organ system that we're now pretty clear about is the central nervous system. There is evidence of direct involvement of the virus with neurons, and also the cytokine storm and inflammatory mediators can cause damage to the central nervous system. This is manifesting itself not only in neurologic clinical findings, but also psychological findings. We're seeing patients post-discharge struggling with psychological challenges, almost like PTSD. And, we're also seeing some cognitive defects in some people that are very disturbing.

We've also seen damage to the kidneys from the cytokines, and there is also evidence that the virus can bind to receptors in the liver, though we haven't seen significant liver disease yet in patients. Finally, the gastrointestinal tract itself has virus receptors, and about 15% of people, especially children, present with gastrointestinal symptoms. But, so far, there is no evidence that this will cause persistent symptoms.

Finally, it has become clear that infection with SARS-CoV-2 triggers abnormal clotting of the blood in some people. This has led to pulmonary emboli, which are blood clots that travel to and damage the lungs, and strokes, which are blood clots in the vascular system of the brain. Both pulmonary emboli and strokes may have long-term consequences for these two organs.

Then, in pediatrics, there's this multi-system inflammatory syndrome in children, which appears to occur not typically with acute infection, but following the acute infection by a short period of a few weeks. This is where multiple systems are involved with inflammation, including skin, joints, kidneys, lungs and heart. And some of these kids can be very sick, with rare deaths.

 

[Agent K]

Why are US daily new infections and daily new deaths not following the same trajectory? They did in the initial stage. Is this good news?

In the early stage, testing was reserved for the sickest and most high-risk patients, and the graph reflected an increase in testing as well as an increase in cases. Now younger and lower-risk people are getting infected and tested.
Here's an estimate of the actual number of infections based on the number of deaths.
https://covid19-projections.com/us

 

[qed]

How do we explain the difference between France's massive exponential increase in new cases since August not being reflected in a similar increase in new deaths. This seems very significant. This seems to be the case in other European countries too. Is this good news? https://www.worldometers.info/coronavirus/

 

[Oystein]

qed, three factors that certainly play a role:

1. Tests are probably far more available now - Cases were more under-counted in March than they are now
2. Doctors have learned a lot on how to treat patients to avoid death more often (dexamethson, ...)
3. Current level of measures against spread protect the elderly, particularly residents of care facilities, relatively better than young people (or vice versa: young people worse relative to the early lockdown measures), and thus a smaller percentage of the infected today is high-risk.

 

[Mendel]

@qed What is your source? Death statistics are delayed in many countries, is that the case in France? and by how much?
France used to not count cases that weren't hospitalized or occurring in care homes (INSS), are they counting "private" cases now?
French hospitals were unprepared and overrun in the first wave, I think (from a cursory glance at Google) the situation has been better this time around.

Eyeballing your charts, an average of 60 deaths/day in September would match ~6000 cases/day in August, for a 1% case fatality, which is on par with other countries that count all diagnosed cases. If France is doing that now, their figures are reasonable.

 

[Mauro]

The same thing happens in Italy: we had a high infections numbers and a very high mortality rate in March/April. Then infections decreased to just a few until the end of July, then started increasing again and they are increasing at the moment. Today we are about at half the infection rate as of March/April, but about 1/30 of the deaths. By the way, we have one of the worst overall death rates in the world, a very sad 13.49% as of today, mostly due to the March/April period where peak fatality rate was much higher, about 40% (yes, amost one patient in two):

[Source: https://www.worldometer.info]

I agree with Oysten:

qed, three factors that certainly play a role:

1. Tests are probably far more available now - Cases were more under-counted in March than they are now
2. Doctors have learned a lot on how to treat patients to avoid death more often (dexamethson, ...)
3. Current level of measures against spread protect the elderly, particularly residents of care facilities, relatively better than young people (or vice versa: young people worse relative to the early lockdown measures), and thus a smaller percentage of the infected today is high-risk.

I'd add:

4. Interpreting the numbers is difficult. Beyond the number of tests, the testing policy can have a major effect on the infections rate. Ie.: am I testing only people with symptoms? or tracing contacts? or doing blanket tests? One should also factor in the type of testing kits used, which I guess have all different sensitivity/specificity values, which can change (usually getting better) over time. The reporting of the cause of death is also variable and imprecise due to a variety of causes. All in all it's a statistic nightmare.. it will take time before it will be sorted out, and don't expect small error bars.
5. At least in the case of Italy most of the March/April infections were concentrated in just one region: Lombardy, and within Lombardy infections were concentrated in two provinces (Bergamo and Brescia). This overwhelmed the sanitary system (which by itself is rather good, unfortunately not up to German standards) which, also, was unprepared. It's sad to say this, but a lot of people with serious symptoms was left home instead of being hospitalized until it was too late, simply because there were no places available in hospitals or ICU units. Also, I guess having a much higher local concentration of cases than what the Italy-wide figures suggest has caused more elderly people to be exposed. Mismanagements and underestimation at elderly care facilities compounded the problem: in Lombardy, during the worst weeks, Covid patients were even sent to elderly care facilities to quarantine. Today instead infections are spread out all over Italy, hospitals have a good spare capacity and elderly residences are much more conscious of the risks.

 

[deirdre]

How do we explain the difference between France's massive exponential increase in new cases since August not being reflected in a similar increase in new deaths

sept 24

When infection rates first began to rise in France in July, the new cases were predominantly people aged 15-44 and the hospital wards stayed relatively empty of Covid patients.

However its last few weekly reports, Santé Publique France has warned about a worrying rise in the proportion of new cases among elderly, and the over-75s were the age group seeing the biggest increase in new cases over the seven days prior to September 15th.
Hospitals worry that ICU numbers will increase as more elderly people become infected, but the incubation period causes a time lag of around 14 days between new infections and eventual hospital admissions.

.....

Of the total patients admitted to intensive care units since the beginning of pandemic, 71 percent were aged over 75.

But between September 8th and September 15th - the newest detailed data available - over 75s represented just 25 percent of the total ICU patients.

That week, 35 percent of patients were aged 45-64 and 31 percent were aged 65-74, a considerably younger profile than the early days of the epidemic.

https://www.thelocal.fr/20200924/an...sly-ill-covid-19-patients-are-getting-younger

 

[Mendel]

Summarizing the previous posts, death rates depend highly on which age groups get sick; if they're managing to protect the elderly better now, there are less fatalities to be expected.

 

[qed]

Nearly half a million new cases world wide yesterday. https://www.worldometers.info/coronavirus/

 

[Mendel]

Nearly half a million new cases world wide yesterday.

Yes. This global increase is driven by the surge in Europe.

https://www.ecdc.europa.eu/en/geographical-distribution-2019-ncov-cases

 

[Mendel]

Source: https://twitter.com/GregWEaton/status/1320808540680921088?s=19

 

[Jeff Semenak]

Source: https://twitter.com/GregWEaton/status/1320808540680921088?s=19

Sorry, but I dispute this as; facts not in evidence. Cases, implies those seeking treatment for an ailment. This is not true with Covid. A Novel way to report.These are positive for a PCR test which is unable to differentiate between, any virus at all; and a meaningful viral load. If, you have 3 Positives, then 2 negatives; you now have 3 new 'cases'. This distortion of language, is a symptom of a corrupt system.

 

[Mendel]

Sorry, but I dispute this as; facts not in evidence. Cases, implies those seeking treatment for an ailment. This is not true with Covid. A Novel way to report.These are positive for a PCR test which is unable to differentiate between, any virus at all; and a meaningful viral load. If, you have 3 Positives, then 2 negatives; you now have 3 new 'cases'. This distortion of language, is a symptom of a corrupt system.

If one and the same person is tested 5 times, they're still only one case. Do you agree?

If a person is positive for SARS-NCoV-2 in a PCR test, the virus has infected their body and replicated in it. Do you agree with that?

The "positivity rate" compares the numbers of positive tests. This may include re-tests. It does not count cases. The advantage is that this rate is more readily available, as labs can just report on the number of tests they do; but it doesn't really reflect the state of epidemic without a measure of how many people got tested. Generally, a high positivity rate (>10%) means that a significant amount of cases is not tested at all, and not detected anywhere.

In the case of this tweet, the average daily new case numbers for cities in Oklahoma are compared. This means they're actually counting people, not a positivity rate; with the "positivity rate" data, you can't distinguish new cases and old cases. It's also highly likely that because all of these cities are in the same state, the people getting tested are selected according to the same set of rules. The cities are considered to be very similar in all respects except for the mask mandate, and therefore the case number comparison is meaningful; and even the positivity rate comparison is meaningful.

I would like to see from you evidence of re-tests of the same infection being counted as new cases, as you seem to claim. I have not seen any evidence of that happening.

Article:

case [kās]
a particular instance of a disease or other problem; sometimes used incorrectly to designate the patient with the disease.

Source: https://medical-dictionary.thefreedictionary.com/case

The "other problem" is "SARS-CoV-2 infection", commonly referred to as "coronavirus infection", but meaning that specific Coronavirus type. The case counts derived from the PCR tests are cases of this infection. I agree that referring to them as "Covid Cases" would be inaccurate; Covid-19 is technically the disease caused by this infection, and the most common usage would be to talk of "Covid patients", i.e. people actually having symptoms or undergoing treatment.

Some people have become "coronavirus cases" twice by getting infected with the virus on two separate occasions, typically with a genetically different strain of the virus. That usage is within the definition I cited, and it's useful for tracking the spread of the virus.

I find that medical professionals are very careful with the language that they use.

 

[Mendel]

Note that the Oklahoma difference between cities with and without mask mandates extends to the death rate:

Article:

Source: http://patricklivingood.oucreate.com/okcovid/mask-mandates/

That blog post includes a very detailed analysis sourced from publicly available data.


Elsewhere on the same site:

Article:

Source: http://patricklivingood.oucreate.com/okcovid/

With the current surge in Europe, I expect the G7 average to rise again.

 

[Jeffrey Orling]

Covid19 finds a way to attack any number of ways leading to death. You will die, for example, from heart failure... but it can be caused by C19. I am thinking when there are "co-morbidities" the stat is counted as C19 not the direct cause of death... I believe AIDS was similar... people died from things AIDS enabled.

 

[Mendel]

Here's a conundrum:

Article:

The numbers are calculated from the number of deaths by month in the SCB database for the years 1860 - 2019 and for 2020 the running updates from SCB, which lags two weeks and now are solid for September with the October 19 updated Excel sheet. This is divided by the yearly population by year for 1860 - 2019 from the SCB database and for August 2020 for the year 2020.

For the year 2020 so far, ie January - September, the number of deaths per capita is the third lowest ever.

Deaths January - September 1860 - 2020 in Sweden per 100 000 population. Data: SCB

The number of deaths per 100 000 population in Sweden was 675 for January - September 2020. 2019 January - September have the record low at 636 and 2016 had the record before that at 673.

Source: https://cornucopia.cornubot.se/2020/10/september-2020-least-deadly-month-ever.html

This seems counterintuitive, because Euromomo shows a clear spike in deaths:

Article:

Source: https://www.euromomo.eu/graphs-and-maps

But Euromomo does not display actual numbers, but rather the deviation from a statistical baseline, called Z-score. And that baseline is continually adjusted.

The ECDC data shows 5938 Covid deaths in Sweden (source: https://www.ecdc.europa.eu/en/cases-2019-ncov-eueea ); Sweden has approximately 10 million inhabitants, so that'd be ~60 deaths per 100 000. I've marked that 60 on the last column in the previous graph:

Obviously, the implication is now the same as for the Euromomo graph: there is a long-term downward trend in deaths, but the excess deaths from the first virus wave do show up as an abnormal spike in the data.

Conundrum solved!

 

[Mendel]

Here's simultaneous animation of the trajectory of the epidemic so far for every country on Earth in terms of daily deaths and cases per capita. Since deaths trail cases by a few weeks, a wave of the epidemic produces a loop in its track. Note the logarithmic scale of the axes! After the simultaneous animation, the tracks of some countries are highlighted, this is the US:

Source: https://m.youtube.com/watch?v=eeiguFaRil0