PlagueWatch 2020-23 - Coronavirus

[Jeffrey]

Probably a little early for much data. Most of these countries only reopened their schools in May and June. See the following (WSJ is paywalled):

Is It Safe to Reopen Schools During Covid-19 Pandemic? Europe Is About to Find Out (WSJ May 14)
Is It Safe to Reopen Schools? These Countries Say Yes (WSJ May 31)
Reopened schools in Europe and Asia have largely avoided coronavirus outbreaks. They have lessons for the U.S. (WaPo July 11)

Or, the July 14th WSJ article...

“Israelis Fear Schools Reopened Too Soon as Covid-19 Cases Climb
Outbreaks in schools had infected at least 1,335 students and 691 staff by Monday since the reopening in early May”

https://www.wsj.com/articles/israeli...mb-11594760001

[Jeffrey]

I will be happy to do so after the kid goes to bed. Workday is done so Mrs CDu wants me off the computer and tending the honeydo list

Is the boss still working you hard?

[Bluedog]

My local high school district (in IL) announced on Monday a hybrid in-person/remote learning that allowed approx 1/4 of the students in the building at any one time. Today, they changed their mind and announced 100% remote learning at the beginning of the school year. Quick reversals indeed...

[swood1000]

Or, the July 14th WSJ article...

“Israelis Fear Schools Reopened Too Soon as Covid-19 Cases Climb
Outbreaks in schools had infected at least 1,335 students and 691 staff by Monday since the reopening in early May”

https://www.wsj.com/articles/israeli...mb-11594760001

Some of the data is open to a wide variety of characterization. The Daily Beast reports that in Israel schools openings were the problem: Israeli Data Show School Openings Were a Disaster That Wiped Out Lockdown Gains

Israel’s unchecked resurgence of COVID-19 was propelled by the abrupt May 17 decision to reopen all schools, medical and public-health officials have told The Daily Beast.

WSJ reports that Israelis Fear Schools Reopened Too Soon as Covid-19 Cases Climb but includes a different angle:

Israel’s top public health official, Siegal Sadetzki, resigned last week in protest of the government’s handling of the coronavirus pandemic, including the school re-openings.

“Much of the source of the second wave stems from the opening of the education system in a way that was not adapted to corona and mass-gatherings,” she said in a Facebook post explaining her resignation.

While some argue the schools were a major contributor to the overall surge, others say it is difficult to separate them from the broader reopening.

Israeli authorities allowed for the quick resumption of large gatherings like weddings and prayer services. There was little enforcement of rules requiring masks in public.

“Due to the fact that the restrictions were released very fast, it’s difficult to disentangle the effects of each separately,” said Eli Waxman, a physicist who heads the panel of experts advising the government on the coronavirus.

This alternative approach portrays it as a problem with the Israel as a whole dropping precautions prematurely, with the schools being just one of the effects.

[Jeffrey]

Some of the data is open to a wide variety of characterization.

True, the reopening efforts are very dynamic and it’s hard to be certain of origins without high quality testing and contact tracing.

[Mtn.Devil.91.92.01.10.15]

Probably a little early for much data. Most of these countries only reopened their schools in May and June. See the following (WSJ is paywalled):

Is It Safe to Reopen Schools During Covid-19 Pandemic? Europe Is About to Find Out (WSJ May 14)
Is It Safe to Reopen Schools? These Countries Say Yes (WSJ May 31)
Reopened schools in Europe and Asia have largely avoided coronavirus outbreaks. They have lessons for the U.S. (WaPo July 11)

Well, and if those countries are also doing the work of social distancing and masks, it's sort of comparing apples and lobsters.

There's no telling what in person classes might do to our current hotbed areas.

[CDu]

Is the boss still working you hard?

Sorry Jeffrey! I promised last night to put together a vis-a-vis comparison of flu and COVID risk by age. Last night got away from me, but here it is:

From the CDC, we have data on flu deaths per year over the past decade: https://www.cdc.gov/flu/about/burden/past-seasons.html

The average annual deaths is around 37,500 (30,500-54,300). I don't have data prior to 2010-11, so I'm not sure exactly how they estimated an annual death rate of 60,000 per year for flu. Only once in the past 10 years has the estimated annual flu death rate topped 60,000 (in 2017-18). But, that's neither here nor there.

From that same website, you can select each individual year to see the breakdown by age. From the previous discussion, the 0-17 was of interest. In that age group (actually two age groups combined in the CDC data), the average annual death toll was 454.

Comparatively, there had been 46 confirmed deaths from COVID in the 0-19 age group as of May 30 (out of ~71K COVID deaths overall; see weblink below). Since then, we've doubled the death toll, so let's say we're at 90 kid deaths right now, in just 4 months worth of COVID season. If we scale up for July and August (two more months), we get to around 135 deaths. So, a little less than 1/3 the number of deaths as average annual flu deaths. Not the 1/7 that the website derived as their projection back in mid-May.

BUT, that still doesn't account for the fact that kids have been interacting with far fewer people during the lockdown than they would if they were in school. I would estimate that the contact rate (number of people a person comes in contact with in a day) is probably at least 10-fold lower than if they were in school as normal. So even if we multiply the increased risk from being in school by just 4, that's a higher death toll than the flu. And kids would still be at risk of flu deaths.

Now, yes, the overall risk of death is still quite low for kids. But I think the estimated 1/7 risk vs flu is a wildly optimistic assessment for the reasons above.

As for the death risk given infection by age, here goes:

I've used 2 sources here. First, the 2017-18 flu season data broken down by age. I used this year because it had the largest number of cases and deaths, so hopefully is the least noisy. Pooling all those years of data seemed too much (i.e., I'm lazy). The second is the CDC's COVID data mentioned above.
https://www.cdc.gov/flu/about/burden/2010-2011.html
https://www.cdc.gov/mmwr/volumes/69/wr/mm6924e2.htm

COVID death risk by age (given infection):
<=9: 0.064% (13/20,458)
10-19: 0.067% (33/49,245)
20-29: 0.15% (273/182,469)
30-39: 0.4% (852/214,849)
40-49: 1.0% (2,083/219,139)
50-59: 2.4% (5,639/235,774)
60-69: 6.7% (11,947/179,007)
70-79: 16.6% (17,510/105,252)
80+: 28.7% (32,766/114,295)

I mentioned before that these %s are probably overestimates of the case fatality rate, as asymptomatics are underreported. Now, deaths are almost certainly also underreported, though not to the same degree. So let's say that the death rate is off by a factor of 5 (so 80% asymptomatic not being captured in the data, which is on the high end of what I've seen published; CDC says about 40%), with older ages (where the cases are more severe and thus more likely to be identified) being closer to accurate (let's say a factor of 3 off) and younger ages (less severe and more likely to not get tested) more like 10 times underreported. Then we get:

<=9: 0.0064% (0.064%/10)
10-19: 0.0067% (0.067%/10)
20-29: 0.015% (0.15%/10)
30-39: 0.08% (0.4%/5)
40-49: 0.2% (1%/5)
50-59: 0.5% (2.4%/5)
60-69: 2.2% (6.7%/3)
70-79: 5.5% (16.6%/3)
80+: 9.6% (28.7%/3)

Compare that with the flu (using the data from 2017-18, and even ignoring that this is just among symptomatic flu cases):
0-17: 0.0056% (625/11,190,943)
18-49: .02% (2,803/14,428,065)
50-64: 0.05% (6,751/ 13,237,932)
65+: 0.86% (50903/5,945,690)

[Sorry - they didn't break out the older age groups as nicely in the flu data]

So, I'd say that, even assuming the more generous estimates of the undiagnosed asymptomatic cases, COVID is more lethal across all age groups, and MUCH more lethal in the older age groups. The difference is much smaller in pediatrics, but the concerning thing is that the COVID fatality rate for those ages 30-59 (so the vast majority of education employees and parents of students) appears to be comparable to the flu fatality rate for 65+ year olds. That is a BIG problem.

As for the kids, the fatality rate suggests that we've likely been somewhat fortunate (in terms of COVID deaths) in that they have been far less exposed than normal as a function of being out of school.

This alternative approach portrays it as a problem with the Israel as a whole dropping precautions prematurely, with the schools being just one of the effects.

And what about the reopening in the US suggests we haven't been premature in dropping precautions?

Europe and Asia have largely contained the virus now, which is why returning to school is more feasible. We have... not contained the virus. Which makes returning to school much more dangerous. The experiences of Europe are thus not really comparable to what we should expect. That's not to say that opening schools is necessarily wrong. As I said before, it's a really complicated issue.

Note: this is not meant to belittle the flu, which for anyone over 50 should be a SERIOUS concern (get that flu shot, folks!). Just meant to illustrate the severity of COVID.

[budwom]

New national career highs for deaths Thursday (943 at least) and cases (77,255). This is getting the attention of many, but not all people, SMH. Without significant change, we'll blow past the 200k mark around Labor Day.

[freshmanjs]

New national career highs for deaths Thursday (943 at least) and cases (77,255). This is getting the attention of many, but not all people, SMH. Without significant change, we'll blow past the 200k mark around Labor Day.

943 is nowhere near the national daily high for deaths

[MChambers]

New national career highs for deaths Thursday (943 at least) and cases (77,255). This is getting the attention of many, but not all people, SMH. Without significant change, we'll blow past the 200k mark around Labor Day.

And earlier this year I thought it wouldn't be until Nov. 1. Man . . .

[CDu]

Thanks, best get to your real job. Hope your boss is pleased with your performance!

Are all of you feeling better? Hopefully, it’s a minor cold/flu.

PS - Yes, we're doing much better. Mrs. CDu, as usual, led the way in recovery, followed closely by CDu Jr, then me. Symptom severity highly correlated with duration of illness too, although "minor cold" is a pretty good assessment. Thanks for asking!

[Jeffrey]

PS - Yes, we're doing much better. Mrs. CDu, as usual, led the way in recovery, followed closely by CDu Jr, then me. Symptom severity highly correlated with duration of illness too, although "minor cold" is a pretty good assessment. Thanks for asking!

Thanks for your update... that’s great news!

[CDu]

943 is nowhere near the national daily high for deaths

Yep, and 77,000 is probably not really our national daily high for cases either. It's just the daily high for confirmed cases. The actual cases in March and April probably dwarf the confirmed cases due to the lack of testing available at that time. I'll repost the graphs that illustrate, conceptually, what we might be seeing (i.e., wildly underidentifying cases early on, still underidentifying but doing a better job of things now):

side-by-side Covid.jpg

So either the gray bar would have to be way overestimating cases earlier or way underestimating cases now for us not to have a lower case load currently than at our peak.

I will say, though, that we appear to be heading in the wrong direction.

[Jeffrey]

Sorry Jeffrey! I promised last night to put together a vis-a-vis comparison of flu and COVID risk by age. Last night got away from me, but here it is:

From the CDC, we have data on flu deaths per year over the past decade: https://www.cdc.gov/flu/about/burden/past-seasons.html

The average annual deaths is around 37,500 (30,500-54,300). I don't have data prior to 2010-11, so I'm not sure exactly how they estimated an annual death rate of 60,000 per year for flu. Only once in the past 10 years has the estimated annual flu death rate topped 60,000 (in 2017-18). But, that's neither here nor there.

From that same website, you can select each individual year to see the breakdown by age. From the previous discussion, the 0-17 was of interest. In that age group (actually two age groups combined in the CDC data), the average annual death toll was 454.

Comparatively, there had been 46 confirmed deaths from COVID in the 0-19 age group as of May 30 (out of ~71K COVID deaths overall; see weblink below). Since then, we've doubled the death toll, so let's say we're at 90 kid deaths right now, in just 4 months worth of COVID season. If we scale up for July and August (two more months), we get to around 135 deaths. So, a little less than 1/3 the number of deaths as average annual flu deaths. Not the 1/7 that the website derived as their projection back in mid-May.

BUT, that still doesn't account for the fact that kids have been interacting with far fewer people during the lockdown than they would if they were in school. I would estimate that the contact rate (number of people a person comes in contact with in a day) is probably at least 10-fold lower than if they were in school as normal. So even if we multiply the increased risk from being in school by just 4, that's a higher death toll than the flu. And kids would still be at risk of flu deaths.

Now, yes, the overall risk of death is still quite low for kids. But I think the estimated 1/7 risk vs flu is a wildly optimistic assessment for the reasons above.

As for the death risk given infection by age, here goes:

I've used 2 sources here. First, the 2017-18 flu season data broken down by age. I used this year because it had the largest number of cases and deaths, so hopefully is the least noisy. Pooling all those years of data seemed too much (i.e., I'm lazy). The second is the CDC's COVID data mentioned above.
https://www.cdc.gov/flu/about/burden/2010-2011.html
https://www.cdc.gov/mmwr/volumes/69/wr/mm6924e2.htm

COVID death risk by age (given infection):
<=9: 0.064% (13/20,458)
10-19: 0.067% (33/49,245)
20-29: 0.15% (273/182,469)
30-39: 0.4% (852/214,849)
40-49: 1.0% (2,083/219,139)
50-59: 2.4% (5,639/235,774)
60-69: 6.7% (11,947/179,007)
70-79: 16.6% (17,510/105,252)
80+: 28.7% (32,766/114,295)

I mentioned before that these %s are probably overestimates of the case fatality rate, as asymptomatics are underreported. Now, deaths are almost certainly also underreported, though not to the same degree. So let's say that the death rate is off by a factor of 5 (so 80% asymptomatic not being captured in the data, which is on the high end of what I've seen published; CDC says about 40%), with older ages (where the cases are more severe and thus more likely to be identified) being closer to accurate (let's say a factor of 3 off) and younger ages (less severe and more likely to not get tested) more like 10 times underreported. Then we get:

<=9: 0.0064% (0.064%/10)
10-19: 0.0067% (0.067%/10)
20-29: 0.015% (0.15%/10)
30-39: 0.08% (0.4%/5)
40-49: 0.2% (1%/5)
50-59: 0.5% (2.4%/5)
60-69: 2.2% (6.7%/3)
70-79: 5.5% (16.6%/3)
80+: 9.6% (28.7%/3)

Compare that with the flu (using the data from 2017-18, and even ignoring that this is just among symptomatic flu cases):
0-17: 0.0056% (625/11,190,943)
18-49: .02% (2,803/14,428,065)
50-64: 0.05% (6,751/ 13,237,932)
65+: 0.86% (50903/5,945,690)

[Sorry - they didn't break out the older age groups as nicely in the flu data]

So, I'd say that, even assuming the more generous estimates of the undiagnosed asymptomatic cases, COVID is more lethal across all age groups, and MUCH more lethal in the older age groups. The difference is much smaller in pediatrics, but the concerning thing is that the COVID fatality rate for those ages 30-59 (so the vast majority of education employees and parents of students) appears to be comparable to the flu fatality rate for 65+ year olds. That is a BIG problem.

As for the kids, the fatality rate suggests that we've likely been somewhat fortunate (in terms of COVID deaths) in that they have been far less exposed than normal as a function of being out of school.



And what about the reopening in the US suggests we haven't been premature in dropping precautions?

Europe and Asia have largely contained the virus now, which is why returning to school is more feasible. We have... not contained the virus. Which makes returning to school much more dangerous. The experiences of Europe are thus not really comparable to what we should expect. That's not to say that opening schools is necessarily wrong. As I said before, it's a really complicated issue.

Note: this is not meant to belittle the flu, which for anyone over 50 should be a SERIOUS concern (get that flu shot, folks!). Just meant to illustrate the severity of COVID.

Thank you, very much, for all of your effort! I really appreciate your detailed analysis, which is extremely valuable to me and my family, as I make critical decisions!

[rsvman]

This has been my approach. When I’m outside I stay away from people and have a mask if needed but don’t wear it unless I’m close to others. This is for walks, running, bike rides, golf, etc. If I have to go inside a place that’s not my house I wear a mask. Is this not cautious enough?

I think it is. It is what I am doing, as well.

[CDu]

Thank you, very much, for all of your effort! I really appreciate your detailed analysis, which is extremely valuable to me and my family, as I make critical decisions!

No worries! I don’t envy your decision. Sometimes having options can be more stressful. But I am sure that whichever way you go it will prove the right choice!

[cspan37421]

CAN WE PLEASE STOP COMPARING COVID TO THE FLU?!?!?

Thanks.

OK, think of it more like the common cold. After all, it's spreading rapidly ... becoming more common! And it's a virus, like the common cold!

It's just way better at killing you, or damaging your major organ systems, but that's just one point of difference, compared to the two similarities above.

[jimsumner]

CAN WE PLEASE STOP COMPARING COVID TO THE FLU?!?!?

Thanks.

Can we compare it to automobile accidents? That one's also out there.

[OldPhiKap]

OK, think of it more like the common cold. After all, it's spreading rapidly ... becoming more common! And it's a virus, like the common cold!

It's just way better at killing you, or damaging your major organ systems, but that's just one point of difference, compared to the two similarities above.

Yup. Nailed it.

[OldPhiKap]

Can we compare it to automobile accidents? That one's also out there.

This, too.

Very few people get struck by lightning, so to hell with getting off the water when the storm rolls in. My metal john boat and a bit of faith in the odds is all I need.