COVID-19 Megathread 5: The Trumps catch COVID-19

[💥💥 brandon bro (he/him/his)]

Seems highly dishonest to me to compare the observed outcomes with the upper limit of the projections instead of the mean of the projections.

[💥💥 brandon bro (he/him/his)]

DeSantis pulled this out during his presser today

And? Seems pretty straightforward to me.

I was just posting the chart stop being so God damn snarky

except it is a blatantly dishonest chart

Seems highly dishonest to me to compare the observed outcomes with the upper limit of the projections instead of the mean of the projections.

[💥💥 brandon bro (he/him/his)]

DeSantis pulled this out during his presser today

snip

And? Seems pretty straightforward to me.

I was just posting the chart stop being so God damn snarky

except it is a blatantly dishonest chart

Seems highly dishonest to me to compare the observed outcomes with the upper limit of the projections instead of the mean of the projections.

Of course people like you would think its dishonest

[💥💥 brandon bro (he/him/his)]

It's worth noting that in that CDC link the death counts are incomplete.

The provisional counts for coronavirus disease (COVID-19) deaths are based on a current flow of mortality data in the National Vital Statistics System. National provisional counts include deaths occurring within the 50 states and the District of Columbia that have been received and coded as of the date specified. It is important to note that it can take several weeks for death records to be submitted to National Center for Health Statistics (NCHS), processed, coded, and tabulated. Therefore, the data shown on this page may be incomplete, and will likely not include all deaths that occurred during a given time period, especially for the more recent time periods. Death counts for earlier weeks are continually revised and may increase or decrease as new and updated death certificate data are received from the states by NCHS. COVID-19 death counts shown here may differ from other published sources, as data currently are lagged by an average of 1–2 weeks.

So you can't say with certainty that there's more of one type than any other. This being a representative sample (17K deaths out of 40K+) is reliant on there not being any discrepancies or differences in reporting by state/locality, place of death, etc.

Worth noting as well that several states are experiencing outbreaks in nursing homes and elderly care facilities. If you look at these numbers, the number of total deaths 75+ is 4x the number of nursing home deaths. We have no way of knowing how many 75+ year olds are in nursing homes, but when you have outbreaks in facilities dominated by old people, you're going to have a lot of mortality.

But also the dig on the media (?) is very bizarre. When did anybody ever assert this thing was nearly as deadly to young people as the elderly? I heard plenty of reporting that the exact opposite was true.

[💥💥 brandon bro (he/him/his)]

It's worth noting that in that CDC link the death counts are incomplete.

The provisional counts for coronavirus disease (COVID-19) deaths are based on a current flow of mortality data in the National Vital Statistics System. National provisional counts include deaths occurring within the 50 states and the District of Columbia that have been received and coded as of the date specified. It is important to note that it can take several weeks for death records to be submitted to National Center for Health Statistics (NCHS), processed, coded, and tabulated. Therefore, the data shown on this page may be incomplete, and will likely not include all deaths that occurred during a given time period, especially for the more recent time periods. Death counts for earlier weeks are continually revised and may increase or decrease as new and updated death certificate data are received from the states by NCHS. COVID-19 death counts shown here may differ from other published sources, as data currently are lagged by an average of 1–2 weeks.

So you can't say with certainty that there's more of one type than any other. This being a representative sample (17K deaths out of 40K+) is reliant on there not being any discrepancies or differences in reporting by state/locality, place of death, etc.

Worth noting as well that several states are experiencing outbreaks in nursing homes and elderly care facilities. If you look at these numbers, the number of total deaths 75+ is 4x the number of nursing home deaths. We have no way of knowing how many 75+ year olds are in nursing homes, but when you have outbreaks in facilities dominated by old people, you're going to have a lot of mortality.

But also the dig on the media (?) is very bizarre. When did anybody ever assert this thing was nearly as deadly to young people as the elderly? I heard plenty of reporting that the exact opposite was true.

I've been mainly reading media reports discussing infants, children, and teenagers who have died from coronavirus, including people who had no underlying medical conditions. Those reports were basically aiming to address the claims of individuals who said that younger people were "invincible" in light of this pandemic.

Right, it's objectively true that there's non-zero risk to people of every age. I have no idea where this claim that it was "sensationalized" for young people comes from, and why a comparison with the most vulnerable age group population is warranted at all.

This also ignores the fact that young people, if infected, can spread the virus to old people. I saw much more coverage about young healthy people as potential disease vectors than I did about young people actually dying.

[💥💥 brandon bro (he/him/his)]

https://www.nbcnews.com/news/world/u-n-warns-hunger-pandemic-amid-threats-coronavirus-economic-downturn-n1189326

Can we stop pretending that we will be all locked down for years and that it will be worth it past June?

I know most people on this forum come from households earning millions a year but...

Almost all of this article is about problems that will be much more prominent in the global south than the United States...

The areas in the US experiencing the most poverty and what is historically some of the worst health care in the country are experiencing some of the heaviest case loads and fatalities (reservations, urban centers, factories and processing plants staffed by immigrants and blue collar workers). In addition to having more comorbidities, people in these communities typically work manual or service jobs deemed essential. In other words, being "incentivized" to continue to participate in the labor force is making these people more sick than the white collar workers who can stay at home. Notably, the (astroturfed) "liberate" movements and TV pundits talking about reopening the economy don't have any representation from these communities.

Containing the virus versus keeping people fed is of course a false choice in the United States. This is why the social safety net exists. Just give people money. We already do this in several forms; just temporarily scale it up for a few months.

[💥💥 brandon bro (he/him/his)]

Giving out money would be a better answer in a country that was capable of getting a check to everyone within a couple of weeks, or that had a functioning system of unemployment insurance.

This fragile and humiliating state of affairs is the result of decades of political decisions, but no amount of political will can reverse that savage reality in time to get help to everyone who would need it during a lockdown of six months or more.

In short, everyone is afraid, no one knows exactly what to do, and the circumstances that we face are so unprecedented that projections about either the economy or the virus remain extremely uncertain (discounting hyperbolic claims such as Beet's assertion in the first COVID thread that 5% of the world's human population would die).

Sure, I absolutely agree with this. There is no denying that the majority of people in this country are going to come out of this time period much worse than they were going into it and government assistance can't fully remedy that (even if it could supplement household income, there is a lot of damage due to a tearing social fabric, loss of institutions, deaths of family and friends, etc. which government checks can't fix).

My point is, all of these arguments about when to "reopen" the economy are ultimately political arguments about what policy makers and enactors should and shouldn't do. But it's a false dichotomy to say that governments (federal, state and local) are deciding between lockdowns and people's income. If you accept that, e.g., sitting in your car and honking at the Michigan state capitol building is sufficient political action for ending lockdowns, you should also accept that, e.g., putting political pressure on your state legislators to fix your unemployment insurance system is also an option for protecting people without income.

The reason this isn't discussed as a viable option is that most people see it as laughable that many governments would actually provide any sustained assistance here. Despite being a broadly popular nobody seriously thinks people like Mitch McConnell or Robin Vos will make meaningful or sustained efforts here. The nation has been so shocked by decades of ideological devotion to austerity and ghoulish disregard for the poor that it can't even dream to demand the most common-sense assistance right now.

[💥💥 brandon bro (he/him/his)]

@ people posting serological studies to "prove" the infection rate is much higher than claimed

depending on the test used, there's a high false positive that could inflate (and explain) the number of infecteds reported.

In the Stanford-run Santa Clara study, the test false positive rate was between 0 and 10%. You can explain all of the infecteds reported if you assume the test has a 1.5% false positive rate (i.e., if the test has "only" a 1.5% false positive rate, then all of the reported positives in the study could be false positives). (People, including Andrew Gelman who I linked, have highlighted some other statistical shortcomings in this study).

That doesn't mean the findings using serological studies are false, but it means you should view them with skepticism and uncertainty. You should think critically about any story analyzing surveys of test results (especially the serological tests) and authors should be acknowledging the assumptions that go into their numbers.

[💥💥 brandon bro (he/him/his)]

What do you guys think? Could this turn into something or is it a nothing burger?

https://twitter.com/SethAbramson/status/1253785676274970625?s=20

This is garbage. If Trump seriously feared any leverage the Chinese had on him you wouldn't have heard anything about "Wuhan flu".

[💥💥 brandon bro (he/him/his)]

You can have serious disagreements about the timeline for entering different phases of reopening but people who downplay the the reaction to 2K daily deaths for nearly a month and >50K deaths (an  underreported figure that's continuing to grow) due to a disease nobody had heard of six months ago are downright sociopathic.

You would think that these same people who were calling this an overreaction in March when we still had <1K deaths would have learned their lesson by now, but of course they haven't because they're driven by ideological dogmatism and election-horserace watching rather than any sort of understanding of public health let alone compassion.

[💥💥 brandon bro (he/him/his)]

You can have serious disagreements about the timeline for entering different stages of reopening, but the people who downplay that all the epidemiological models used to justify stay-at-home orders and lockdowns have blown-up spectacularly and there is no massaive mortality wave or resource shortage (outside of New York's uniquely ill-equipped hospital system) despite chicken littles promising us we were all doomed even if we did socially distance are downright sadistic.

You would think that the same people saying that every ICU in the nation would be overwhelmed when we only had <1k deaths would have admitted the public response was an overreaction, but of course they haven't because they're driven by ideological dogmatism and election-horserace watching rather than any sort of understanding of the economic/social consequences of lockdowns, isolation, and mass joblessness let alone compassion for the hundreds of millions of lives that have been disrupted.

Very cute.

Just gonna leave these here to demonstrate how utterly wrong you have proven yourself to be repeatedly in this thread

You, on April 5, using a misunderstanding of fractions to minimize the increase in confirmed cases. (Oops, turns out there are now 8 times as many confirmed cases as there were on March 27 )

So the rate of new diagnoses has slowed everyday since 03/27 despite the steady improvements in testing availability?

You, on April 5, saying that panic buying was somehow inflicting more damage than the virus itself:  (oops, now there are seven times as many fatalities as there were the day you posted this )

Panic buying is already a *major, major problem (and so far is actually a bigger problem than anything we've seen on the healthcare side).

You, on April 9, claiming the US is one of the "success stories" of the pandemic whereas today the United States has six times the worldwide deaths per capita and being sixteenth highest fatality rate per capita on Earth.

Is there anyone here who still maintains that the US is on an Italy-type disease trajectory?  Or can we all reasonably agree that the United States looks like it will be one of the "success stories" of the global pandemic?  If so, can we talk about what factors have influenced the U.S. emerging relatively unscathed from this?  Better testing, more docs/ventilators, lower population density, younger demographics, warm weather, etc.

You, on April 21, citing an article to make a definitive claim about age-specific mortality without realizing that the sampling here is incomplete and totally reliant on state reports which are inconsistent in timing and methodology.

Updated data from CDC show that more Americans over the age of 85 have died of COVID-19 than those under age 65.  The risks to young people are terribly overstated by the sensationalist media.  

Face it kid, you are not nearly as smart as you think you are. You've been trying to present studies, findings, and inference which are apparently totally outside of your training or field of expertise to try to make conclusions which are not only methodologically unsound but also hugely embarrassing in retrospect. That is to say nothing about your transparent bad faith and immunity to any sort of refutation of your talking points.

[💥💥 brandon bro (he/him/his)]

1.  The availability of testing is the primary determinant of new case counts, which makes its value as an indicator of the pandemic's progression inherently problematic.  This is mainstream opinion.  U.S. testing capacity has continued to expand and the test-positive rate has declined, suggesting that the level of U.S. testing has sufficiently expanded to keep-up with growth in new cases.

Lmao that's not why your post was so egregiously bad. You tried, in the first week of April, to say that the declining rate of growth in the absolute number of positive cases indicated that we were past peak. Not only was it hilariously untrue that we peaked at the end of March but you literally ignored the fact that there was an increase in the raw number of cases over the prior week and instead opted to herald the fact that we had a decreasing percentage of cases even though this is explained by the fact that the ratio features an increasing denominator.

That level of innumeracy alone is reason to doubt most of what you say.
 

2.  The "worldwide deaths per capita" includes countries where there's practically zero testing, and therefore zero cases and zero deaths attributable to COVID-19.  ~30% of U.S. deaths have been in New York, the U.S. death rate excluding New York is only 120/1M.  Let's compare that to the 67,000/1M in the U.S. who are now U3 unemployed (so it doesn't even include people who have taken reduced hours, stopped looking for work, etc.)  The economic consequences (mostly brought on by lockdowns and other restrictions) of this virus have by far eclipsed the death toll.        

Brilliant analysis. If you exclude New York city then the number of cases per capita in the US goes down. Maybe I was underrating your quantitative instincts before.

Of course, to your point, there is an alternative way to supplement or provide income to these people: give them government assistance! Weird how you never invoke this as an option! It's almost as if you are concern trolling about the economic impact.

Also absolutely ridiculous and unfounded claim that most of the economic damage is due to the lockdowns and not to, uh, a highly contagious and deadly virus which has a high number of asymptomatic carriers. The fact that we saw >5 million people sign up for unemployment insurance before any of these lockdowns were even ordered is obvious evidence for this. The fact that restaurants are still operating and not coming anywhere near standard business levels is evidence for this. The fact that stay at home orders are still overwhelmingly popular by over 2-1 margins is also evidence for this.

I'm sure the response to this is going to be some nonsense claim that "the media" oversensationalized this and convinced people to not order take out at restaurants or some similar nonsense.

3.  The U.S. is not on an Italy-type trajectory.  Outside of a few isolated cases in New York City, hospitals have not had to triage patients.  There was no shortage of ventilators.  Hospital and ICUs across the U.S. are more empty now than they were at the beginning of March.  This, quite frankly, doesn't look like a pandemic in most of the U.S.  Demographic, institutional and cultural differences between the U.S. and Italy go far in explaining why the pandemic's progression has been so different in the two countries.  I don't see how this is a controversial observation.

Uhm... if you look at per-capita case loads by state and compare them to case loads per province in Italy you get rather similar numbers. Just like in the US/NYC, one region (Lombardy) accounting for more cases than the next three hardest-hit regions and has 10x the cases as the sixth largest region (out of 20!) These Italian cases of triaging and shortages were also happening almost exclusively in Lombardy. If "does this feel like a pandemic in the whole country" is your criteria for differentiating the two then it's failing to meet your standard.

Source; data is only two days old.

And that wasn't even the worst part the your post. I was mainly referring to the "success story" part; strange how this catastrophically bad impact on the economy, which was totally avoidable according to you, doesn't stop the US from being a "success story".

4.  All U.S. death statistics rely on state reporting, which necessarily has different standards/lags.  There is no Federal/CDC death count that is independent of what states are reporting.  This is the data we have, and I'm only letting it speak.  All mortality data suggests the elderly are exceptionally more at-risk than <50s, and the chicken littles' obsession with outlier cases and anecdotes is not an accurate representation of what is happening.      

You can let it speak but I still see nothing from you that acknowledges or even indicates that you understand that this data is incomplete (at the time you posted it it had less than half of the total number of fatalities accounted for) and that any inference from it is, to borrow a concept you appear to care about, quite imprecise.

The age-dependent mortality has, of course, been a central feature of reporting since the very beginning. This was a central part of reporting when the virus was still confined to Wuhan! Of course people with pre-conditions are more likely to be elderly; nobody has ever said anything resembling the contrary. To this day I do not understand the point you were trying to make. Of course the weirdest thing about this argument is that it, apparently, presumes that elderly people can only catch the disease from other elderly people... as if there was not a benefit to limiting the number of asymptomatic younger carriers!

[💥💥 brandon bro (he/him/his)]

1.  Arch's presentation of the numbers was not entirely clear, and after I raised my questions about how to interpret the % change figures he was including in his posts he started listing the daily change and cumulative increase figures separately.  I won't apologize for a brief misunderstanding of the numbers he was presenting, especially when I've contributed no shortage of statistics, data and perspective of my own to the thread.

Maybe this means you should... think more before you post? The raw case numbers Arch was reporting were increasing that week; very conveniently, you ignored those for the figures that reaffirmed your pre-existing beliefs despite much more clear and obvious data to the contrary.

It's not hard to see why this episode illustrates that your posts deserve to be scrutinized more than apparently you are doing yourself.

And the costs of >$2Trn is government assistance is of no-concern?  And even so, no amount of government stimulus (assuming its easy to get, which has not always been the case in these times) can supplement the security afforded by a stable job.  Government bailouts and emergency measures are only temporary stopgaps, and cannot be sustainable long-term strategies the economy.  Fiscal considerations aside, the human costs of the current economic downturn dwarf the suffering brought about as a direct result of deaths/infections/PPE shortages or whatever.  

There is no serious person who thinks that lockdowns are going to persist beyond the summer, even if that long. Temporary stopgaps to deal with a temporary crisis.

And there's no guarantee (or even evidence to suggest) that spending will resume to January-February/Q4 2019 levels. Consumers are (rightly) spooked and that's percolated upwards to halt economic activity at higher levels. If you're concerned that government spending can't fully revive a person's income, then reopening the economy shouldn't make you feel much better, especially considering the potential lost activity due to workers and consumers becoming reinfected.

4.  There's no problem using incomplete death counts to talk about age-specific mortality unless there's reason to think death reports from only certain age groups lag more than others.  At any given point in time, an incomplete assessment shouldn't affect the age distribution of deaths because the age of a decadent is irrelevant to when it gets reported.  If you have a reason not to suspect this, please share.  

Erm... you acknowledged in your last post that Italy has a very different demographic breakdown than the United States. That is true! It is also true that states (and even counties!) have differing demographics, including wealth and comorbidities. Unless you have evidence that you're getting an unbiased sample of states then your claims are pretty uncertain.

Also, the age of the decedent does potentially influence the reporting probability if older people are dying in nursing facilities and younger patients aren't.

This is why you shouldn't be talking about statistics with authority! You very obviously don't have the training or experience in this type of work to understand the potential reporting biases (et cetera) that you are failing to account for.

I don't think there's any reason to actually doubt the numbers but my point is you're so far out of your depth in trying to interpret studies like this that you don't even understand why you could potentially be wrong. You'd think after being dunked on time and time again in this thread you would learn some humility and lay low, but apparently not so. A person who gets refuted this often and returns to make the same points repeatedly is clearly dealing in bad faith.

I really don't understand what you're trying to get out of this exercise.  If you want a blue avatar to kick around, DTT is a much easier target.

Maybe I'm trying to drill into your head that you are not that smart and you should listen to people who actually know what they are talking about?

e: my other larger point from my initial post is that you make posts which assess the current point in time as not being bad, but those posts age poorly because... the situation doesn't get any better! Maybe your posts about the severity of the pandemic that you made today (when there are 55K confirmed deaths) will age just as poorly as the posts you made a month ago (when there were <5K deaths).

[💥💥 brandon bro (he/him/his)]

https://twitter.com/_cingraham/status/1254792999080349697

Of course, not all the excess deaths are necessarily due to COVID-19.  It's likely that there are some due to other medical conditions where people sought help too late due to fear of going to the hospital.  In the other direction, there's probably also a decline in deaths due to auto accidents.  But regardless of the exact numbers, it seems clear that "official" COVID-19 deaths are an underestimate (which is not really a surprise).

I would just like to add that this probably isn't as scary as it looks because almost all of the excess mortality from COVID-19 is occurring in older age cohorts, and a lot of these people would have probably died in the next 24 months anyway.  The pandemic has just hastened their demise.  If anything, we should expect to see reduced all cause mortality for several months after the pandemic ends (as a lot of folks who would have died in that future time period ending up dying early from COVID).  

This is, of course, entirely speculative, in addition to being a rather cruel way to minimize the impacts of the virus.

e: NickG actually laid out some empirical evidence below suggesting this is true in nursing facilities to support the threadbare assertion provided above

[💥💥 brandon bro (he/him/his)]

https://twitter.com/_cingraham/status/1254792999080349697

Of course, not all the excess deaths are necessarily due to COVID-19.  It's likely that there are some due to other medical conditions where people sought help too late due to fear of going to the hospital.  In the other direction, there's probably also a decline in deaths due to auto accidents.  But regardless of the exact numbers, it seems clear that "official" COVID-19 deaths are an underestimate (which is not really a surprise).

I would just like to add that this probably isn't as scary as it looks because almost all of the excess mortality from COVID-19 is occurring in older age cohorts, and a lot of these people would have probably died in the next 24 months anyway.  The pandemic has just hastened their demise.  If anything, we should expect to see reduced all cause mortality for several months after the pandemic ends (as a lot of folks who would have died in that future time period ending up dying early from COVID).   

This is, of course, entirely speculative, in addition to being a rather cruel way to minimize the impacts of the virus.

1) It's not speculative.  Almost 60% of U.S. coronavirus deaths have occurred in persons age 75 or older.  Life expectancy for a 75-year old U.S. male is only 11.8 years, and only 5.6 years for an 85-year old.  People who are dying of this disease are mostly already at end-of-life, or at least very susceptible to developing other severe conditions/complications under non-pandemic conditions.

2) This is how excess mortality works in flu pandemics (and while COVID-19 is *not the flu, pathologically and epidemiological it behaves like a flu virus - one aspect of that being mortality is mostly confined to older, more comorbid populations).  Excess mortality from the 1957-58 flu pandemic was followed by a period of reduced all-cause mortality.  That's how epidemiological theory tells us excess mortality from an unmitigated flu pandemic works - a short period of high excess mortality, followed by a longer period of less noticeable below-average mortality.

I was referring to the clearly bogus statistic that people in the US would likely be dying within 24 months. That's the bolded claim and it's contradicted by the data you provided. Nobody argued that a 75 year old has the same life expectancy as a 25 year old.

This a classic motte-and-bailey tactic.

Also, while on this topic, it's incredibly dangerous to equate dying by COVID with dying by other natural causes. The difference betwen dying of, e.g., COVID at age 80 and dying of heart failure at age 85 is that the person who contracts COVID is highly contagious. Heart disease and the other typical diseases these people suffer have R0 of zero. When people contract COVID (regardless of age) the primary effect is that they could die but the secondary effect is that they can transmit the disease to others. Given the number of health professionals who have gotten sick due to being bombarded with the virus (without proper PPE) there's no reason to equate a COVID death with most non-COVID deaths because COVID deaths are likely begetting other hospitalizations, lost productivity, and deaths.

[💥💥 brandon bro (he/him/his)]

I don’t really see how this can be interpreted as evidence the lockdowns are working.  It’s seems just as likely to me that this is an inevitable consequence of approaching herd immunity, with about 25% of NYC’s population already infected.

If lockdowns worked, this decline would have happened much earlier.

Pandemics die down. People build up immunity, the seasons change, and things change.

Do you literally not understand, or simply refuse to, that the whole point of the lockdown isn't to magically eliminate the virus but to keep the numbers steady so the healthcare system isn't overwhemed?

Keep the numbers steady until when?  You mean the initial goal of the lockdown was to keep infections and death constant until a vaccine was developed?  So everyone was always planning on enforcing lockdowns for 18 months?

The fact is that this was NOT the goal of the lockdown.  If our goal was to spread infections around more evenly so that particular local healthcare systems didn’t get overwhelmed, we would have gone about this much differently.  We would have implemented stronger restrictions where we saw empirical signs of rapid growth, and weaker restrictions where we didn’t.  And we wouldn’t have discouraged New Yorkers from leaving town when the virus exploded there.  If we just wanted to reduce the burden on an overloaded health care system, we would have encouraged New Yorkers to leave to lighten the load on NY’s system and spread the virus to places where it could be better managed.

The goal of the lockdown was to reduce total infections.  That’s what all the models told us would happen.  But the models were wrong in many ways.  The models way underestimated the contagion of the virus and overestimated its lethality.  The lockdowns have almost completely failed to turn the spread negative, and by bluntly forcing a one-size-fits-all approach, they have increased the deaths that will result from the virus by failing to better protect the most vulnerable.

There are more options available than indefinite lockdown until there is a vaccine and no preventative measures.

We started lockdowns in March when we had a health care system that was essentially a blank slate for dealing with the virus. Hospital capacity was one of several lagging features; we also had few tests, a cruder understanding of how the virus spread, a supply chain that was woefully ill-equipped for mass-delivering PPE and testing materials, and no infrastructure for either identifying or isolating sick people or for tracking their contacts. Lockdowns were a good tool at the time for buying time until some of the above could be implemented.

It's true the hospital system in many places has not been overwhelmed yet and appears unlikely to be overwhelmed at any point in the future. This is good news! It's also true that we have a much better understanding of how the virus spread (and have taken preventative measures to reduce transmission, e.g., encouraging wearing masks) and have a supply chain which, while not perfect, is much better than it was six weeks ago. This is also good news; it's part of the reason why some states are credibly considering lifting their lockdowns.

But there's still reason to be very cautious and abide to lockdowns or less-severe restrictions. We are behind on testing capabilities. Some areas are still lacking in PPE and this will become worse as service sector work opens back up. The vast majority of states and cities are nowhere close to any sort of contact tracing or protocols for isolating sick people.

The longer an area stays under rules which demonstrably reduce transmission, the more time they are buying to put these measures in place. How much longer is it worth putting these measures off? It's not clear, and we're going to get some useful information as states reopen. But we aren't waiting solely for the vaccine; we're also waiting on some other factors which will be very helpful for treatment and harm reduction when the rest of the country is able to reopen and are conceivably attainable on a much faster timescale.

[💥💥 brandon bro (he/him/his)]

So is today a legit decline or the Sunday-Monday slump?

For fatalities it's probably both. If you look at last week, you had one high-fatality day when the rest of the week dragged the average down to the roughly the week before. A rolling average suggests that fatalities is very slowly declining. But we have three or four weeks at this point with a S/M slump and there's no mechanistic reason to think that would be changing so numbers today are likely underreporting the same way there were two weeks ago.

It's harder to tell with cases. There's some evidence of proportion of positive tests decreasing with more testing but we had several days last week with >30K positive tests. It strains credulity to think that we're really seeing fewer than 20K cases.

There's also the fact that NY appears to be past-peak but other states with large metros (PA, IL especially) are facing increases, but are increasing to lower peaks than NYC, meaning the overall average is declining. The last couple of weeks NYS was ~40% of daily fatalities but today and yesterday it's down to below 30%.

[💥💥 brandon bro (he/him/his)]

These results are all preliminary tests on small samples (<100 patients) but several studies are finding early signs of longer-lasting lung damage to severe COVID patients that's consistent with but likely more damaging than other SARS and MERS infections.

link (Science News)

Some patients who survive COVID-19 may suffer lasting lung damage

[...]

The tissue lesions can be a sign of chronic lung disease. Similar damage has been documented in survivors of SARS and MERS, respiratory diseases caused by coronaviruses similar to the SARS-CoV-2 virus behind COVID-19. Long-term studies of SARS patients have shown that roughly a third of people who recovered from severe bouts were left with permanent lung damage. In the case of MERS, one study found about a third of people who recovered from a serious infection still had signs of lung damage about seven months later.

But while initial lung images indicate that SARS and MARS typically set into just one lung, COVID-19 appears to be more likely to afflict both lungs right away. In 75 of the 90 patients admitted to Huazhong University Hospital with COVID-19 pneumonia from January 16 to February 17, damage was seen across both lungs, Wang and colleagues report. CT scans taken before hospital discharge revealed that 42 out of 70 patients displayed the type of lesions around the alveoli that are more likely to develop into scars.

[...]

Some lung damage seen in the Wuhan study cases are likely to gradually heal or disappear, Wang and colleagues suspect. However, in some patients, lung abnormalities will harden into layers of scar tissue known as pulmonary fibrosis. That scarring stiffens the lungs, making it hard to get enough oxygen. People with pulmonary fibrosis typically suffer shortness of breath, limiting their ability to be physically active.

[...]

Article suggests a lot of this is due to the cytokine storm. There's some other interesting stuff in here like about how ventilators use is associated with longer-term damage (causation seems hard to determine here).

Here's one of the studies quoted in the article. 60% of patients featured still featured lung scarring upon discharge and all but four patients had "mild to substantial" lung abnormalities upon discharge. It looks like the scarring improved for some patients over the course of infection, though, so there's still some potential for recovery after discharge. Mean age in the study was 45.

[💥💥 brandon bro (he/him/his)]

I'm absolutely livid about the meat processing plants bring forced back open. What protections will there be for the workers? Smithfield in Sioux Falls was already a major hotspot. Most people aren't willing to work meat processing plants, which is why many employee illegal immigrants as well. How are these places going to be staffed? There was no forethought or any real consideration put into this decision. Ugh.

Meat packing plants are actually the valid argument to that people would work those dirty jobs for a higher wage because that is what people did from the 1950s to the 1980s. Increased supply of labor both illegal and legal crashed the average wages of meatpacking jobs, Im not arguing against immigration for this reason as this also had the effect of keeping meat cheap for millions of Americans. This is a freeer market which is a good thing but  its absurd to deny higher wages wouldn't encourage other Americans to work these jobs.

I'm not sure this (bolded claim) is true in 2020 the same way it was true in 1960. The work in a meat packing plant hasn't gotten that much less physical demanding (whereas most other manual labor has, meaning packing jobs are relatively less attractive) and there are several cases where after ICE raids plant operators tried to fill in their open positions with domestic labor with poor retention. E.g., in Postville Iowa they trucked in unemployed people from Texas all the way to Ohio after a 2008 raid and most of them quit within four months; this was at the beginning of the Recession when there weren't that many other places for these people to find work so it's not likely they quit because they found better pay elsewhere.

[💥💥 brandon bro (he/him/his)]

I'm absolutely livid about the meat processing plants bring forced back open. What protections will there be for the workers? Smithfield in Sioux Falls was already a major hotspot. Most people aren't willing to work meat processing plants, which is why many employee illegal immigrants as well. How are these places going to be staffed? There was no forethought or any real consideration put into this decision. Ugh.

Meat packing plants are actually the valid argument to that people would work those dirty jobs for a higher wage because that is what people did from the 1950s to the 1980s. Increased supply of labor both illegal and legal crashed the average wages of meatpacking jobs, Im not arguing against immigration for this reason as this also had the effect of keeping meat cheap for millions of Americans. This is a freeer market which is a good thing but  its absurd to deny higher wages wouldn't encourage other Americans to work these jobs.

I'm not sure this (bolded claim) is true in 2020 the same way it was true in 1960. The work in a meat packing plant hasn't gotten that much less physical demanding (whereas most other manual labor has, meaning packing jobs are relatively less attractive) and there are several cases where after ICE raids plant operators tried to fill in their open positions with domestic labor with poor retention. E.g., in Postville Iowa they trucked in unemployed people from Texas all the way to Ohio after a 2008 raid and most of them quit within four months; this was in the middle of the Recession when there weren't that many other places for these people to find work so it's not likely they quit because they found better pay elsewhere.

I will concede that it will be true in a current scenario that a firm literally can't afford to pay more at an individual plant because that would mean a rise in the sale cost (these plants in general have very low margins IIRC) so therefore if they raise wages and prices every other plant that still hires illegals/immigrants will still keep nearly the same price for their wholesale products . Anyway I find it absurd to demand higher wages for these immigrants such as more system checks and higher minimum wage instead of just at that point discourage immigration and let Americans work it. However keeping cheap labor is good for tens of millions of Americans overall as their meat is now much cheaper. If we go to the point of encouraging higher wages at these plants while still having a labor supply overflow it merely just causes higher meat prices and greater unemployment.

My point was a little bit larger than applying to a single firm elevating wages and facing disadvantage in a competitive market, but this really isn't my specialty so I'm not going to push the point too far.

We might not be looking at or responding to the same things but most of the outrage I am seeing about this (which I agree with) is at the lack of mandatory workplace protections and the liability waivers associated with the act. Obviously having a functioning food supply chain is essential work and invoking the DPA is warranted here but it's insane and inhumane to expect workers to work under these conditions, not have any requirements being put in place to keep them safe, and having the employers shielded from liability which also reduces the likelihood of meaningful safety improvements (which are obviously needed given the outbreaks we're seeing) implemented. Seems to me like getting some sort of mandatory OSHA safety rules is higher priority than higher wages. This is especially scary in areas without enforcement of distancing which let infected plant employees infect others in the community.

[💥💥 brandon bro (he/him/his)]

I'm absolutely livid about the meat processing plants bring forced back open. What protections will there be for the workers? Smithfield in Sioux Falls was already a major hotspot. Most people aren't willing to work meat processing plants, which is why many employee illegal immigrants as well. How are these places going to be staffed? There was no forethought or any real consideration put into this decision. Ugh.

Thats not whats going to happen.

Most plants are shut for 2 weeks or shut indefinitely. Trumps order weakens liability for the big packing companies once they do re-open, but in no way means they're going to open anytime soon.

This is all about making sure Tyson, etc won't be sued whenever they do get a skeleton crew back to work. No one will force workers back against their will. The industry has massive labor issues as it is, that would be suicidal.

No one is being coerced with physical violence but in several states governers are threatening the next most-coercive thing.

In Iowa, Governor Kim Reynolds (R) said failing to return to work would be considered a “voluntary quit,” which would terminate an employee’s benefits.

"If you're an employer and you offer to bring your employee back to work and they decide not to, that's a voluntary quit," Reynolds said Friday. "Therefore, they would not be eligible for the unemployment money."

The governor also said employers should file a report with Iowa Workforce Development if they encounter workers who refuse to come back to their jobs.

One of the qualifications for unemployment benefits is that workers must be “willing and able to work all the days and hours required for the type of work you are seeking,” according to the Texas Workforce Commission.

Those who choose not to return become ineligible for unemployment benefits, said Cisco Gamez, a Texas Workforce Commission spokesman. If workers have concerns about whether their employer is following health guidelines, Gamez said they should contact the U.S. Occupational Safety and Health Administration.

Thankfully, Texas is walking that back and developing some sort of workaround for this, although details remain to be seen.

[Update: On April 28, a day after this story originally published, Gamez said the agency is developing parameters for what might allow Texans to continue qualifying for unemployment insurance if they refuse to return to work at a business reopened by Gov. Greg Abbott’s loosened executive order because they fear contracting or spreading the coronavirus.]

[💥💥 brandon bro (he/him/his)]

I've posted it before but it's worth reminding that several serological tests have large false positive rates (some exceeding and several close to 10%). Everybody who reads and cites serological studies should check and refer to the test accuracy and be very cautious when reporting and interpreting results.

[💥💥 brandon bro (he/him/his)]

New CDC update on COVID-19 released yesterday.  The hospitalization rate is highest among adults 65 or older, but is comparable to what is seen during a high severity flu season.  For children (aged 0-17), hospitalization rates are much lower than a typical flu season.  Nationally, hospitalizations for respiratory illness are only above baseline in the Northeast.  

Hospitalizations are a key indicator, because they lead deaths/ICU admissions while not being influenced by the availability of testing (i.e., severe cases progress to the hospital regardless of when/if testing is performed).  

This all great to see, but it's not comparable to a typical flu season.  This is what we are seeing after 6 weeks of closed schools and large gathering bans and a variable but shorter period of non-essential business closures in most states.  We do not do any of this in a typical flu season.

Yeah I'm genuinely surprised that the hospitalization rate is similar to at most a severe flu outbreak. It's a good sign but you are right, direct comparisons between years should mention the public health policy implemented in nearly all of the US. In fact, this is mentioned elsewhere in the report

Nationwide during week 17, 1.8% of patient visits reported through ILINet were due to ILI. This percentage is below the national baseline of 2.4% and represents the fifth week of a decline after three weeks of increase beginning in early March. The percentage of visits for ILI decreased in all age groups. Nationally, laboratory-confirmed influenza activity as reported by clinical laboratories has decreased to levels usually seen in summer months which, along with changes in healthcare seeking behavior and the impact of social distancing, is likely contributing to the decrease in ILI activity.

Five weeks of decline plus a two week incubation period puts us back at week 10, March 8 - March 14, right before people started social distancing. It will be really curious to see what these numbers look like in two weeks; they surely won't be as bad as March because a lot of distancing is still happening and more people will be outside and in the heat/sunlight, but there is still potential for more reporting activity.

But, it's still good that the hospitalization rates are not exceedingly worse than a severe flu outbreak. This means the flatten-the-curve strategy was successful. So far in most places we haven't seen the strain on the hospital system that the worst-case models were predicting.

It's also worth highlighting this from the report

Based on death certificate data, the percentage of deaths attributed to pneumonia, influenza or COVID-19 (PIC) decreased from 23.6% during week 16 to 14.6% during week 17 which is still significantly above baseline. This is the second week of decline in this indicator, but the percentage remains high compared with any influenza season. The percentage may change as additional death certificates for deaths during recent weeks are processed.

with this accompanying graphic


(source is CDC from the quoted report)

Seems a little strange (and dishonest) to me to say that hospitalization numbers are meaningful because they are related to death rates but then not report a big spike in death rates. But, it's good that the death rates are declining. I'm nervous about some second/third-tier states in population (Illinois, Pennsylvania, Maryland, Virginia) still are not quite showing rates of plateauing which means the cumulative number of deaths is still going to be growing for a while.

edit: the quote on hospitalizations:

The overall cumulative COVID-19 associated hospitalization rate is 40.4 per 100,000, with the highest rates in people 65 years and older (131.6 per 100,000) and 50-64 years (63.7 per 100,000).

    Hospitalization rates for COVID-19 in adults (18-64 years) are higher than hospitalization rates for influenza at comparable time points* during the past 5 influenza seasons.
    For people 65 years and older, current COVID-19 hospitalization rates are similar to those observed during comparable time points* during recent high severity influenza seasons.
    For children (0-17 years), COVID-19 hospitalization rates are much lower than influenza hospitalization rates during recent influenza seasons.

*Number of weeks since 10% of specimens tested positive for SARS-CoV-2 and influenza, respectively.

I really wish they posted the raw data here. It's also worth noting that they say that despite social distancing the hospitalizations for 18 - 64 year olds is higher than a typical year. Also seems kinda weird to not point this out if you're trying to summarize the results of this report.

[💥💥 brandon bro (he/him/his)]

https://www.breitbart.com/politics/2020/05/03/nolte-they-told-us-lockdowns-were-about-flattening-the-curve-they-lied/

What had been a national effort to save lives, a noble effort of national unity, has been twisted by Democrat governors into something unnecessarily punitive that reeks of a partisan power play to destroy President Trump’s re-election chances. Give me a better explanation.  Because this is the bottom line…

Unless you’re willing to wait for a cure (that is at least a year away or might never come at all), when there is no risk the health care system’s going to crash, what is the point of waiting one more day to open up the country when waiting doesn’t make us any safer?

Normal, non-partisan people are asking this question as well.

For every "normal, non-partisan" person who ask this, there are two or three also "normal, non-partisan" people who disagree and reject the premise. This is consistently supported by polling data.

The Right has so internalized the idea that it is the "silent majority" that it cannot fathom the idea that there would be an issue where it wouldn't only be in the minority, but in an overwhelming minority.

[💥💥 brandon bro (he/him/his)]

https://www.breitbart.com/politics/2020/05/03/nolte-they-told-us-lockdowns-were-about-flattening-the-curve-they-lied/

What had been a national effort to save lives, a noble effort of national unity, has been twisted by Democrat governors into something unnecessarily punitive that reeks of a partisan power play to destroy President Trump’s re-election chances. Give me a better explanation.  Because this is the bottom line…

Unless you’re willing to wait for a cure (that is at least a year away or might never come at all), when there is no risk the health care system’s going to crash, what is the point of waiting one more day to open up the country when waiting doesn’t make us any safer?

Normal, non-partisan people are asking this question as well.

For every "normal, non-partisan" person who ask this, there are two or three also "normal, non-partisan" people who disagree and reject the premise. This is consistently supported by polling data.

The Right has so internalized the idea that it is the "silent majority" that it cannot fathom the idea that there would be an issue where it wouldn't only be in the minority, but in an overwhelming minority.

I haven't had the opportunity to ask you, but what is your opinion of what Governor Polis has done here in Colorado, thus far? I know from your posts that you are firmly of the "pro-measures" side: that is, the side which believes that the stay-at-home orders are necessary (which they are and have been), and that they take priority over those who want to throw society open (the "anti-measures" side).

Thanks for asking. Rhetorically he's doing well, but for the reopening... it's an interesting federalism exercise but my expectation is that he reopened statewide too early and trusted some localities to be responsible where perhaps he shouldn't have (specifically thinking about Weld and El Paso).

I'm also very, very doubtful that many small businesses will truthfully comply with recommended protocols for reopening, but I also don't know what a good enforcement mechanism there would look like. I think this month we're going to start seeing more serious pressure mounting to reopen more nationwide, and a lot of places (including CO) seem pretty unprepared, but we're going to be at a point where it's not going to be realistic to ask people to follow strict stay-at-home for much longer. So, I'm disappointed by the move but I understand why they did it and most other places are going to start facing the same decisions soon. I'm prepared to be wrong, though, if we don't see a significant second surge.

Most of the (blue) highly populated areas took the initiative to extend their stay at home orders locally; I know several are set to expire this week so we'll see if they choose to extend. I'm very nervous about the Ft. Morgan and Weld outbreaks associated with prison and packing plants, and for the heightened community spread so far. I'm also glad the state is reopened after ski season is done but in the next month we're going to see increased travel in rural communities for mountain recreation and that has the potential to be very bad. But then again without some sort of strict blockading on highways I don't really know what options the state has.

Your thoughts?