Where coronavirus hospitalizations are falling
axios.com
axios.com
It's not perfect, but in terms of the readily available data, it's not terrible.
> As states open up and elective surgeries, car accidents etc increase the denominator will inevitably increase
OTOH, there's a credible argument that that is a kind of positive progress itself. And, no, it's not inevitable that the denominator will go up if COVID already has hospitals at capacity.
> What is wrong with the much simpler metric of “number of beds used by covid patients”?
That's a good measure for any state, but bad for comparing conditions among states. Share of available hospital beds occupied by COVID-19 patients might be better, but “available” is not always clear since physical capacity and staffing and other support may not yield the same maximum. Because of required staffing ratios, etc.
One thing I find a little suspect about this article, although I do respect Axios, is that hospital usage in Texas has not peaked yet and is currently rising (https://covid19.healthdata.org/united-states-of-america/texa...).
https://apps.texastribune.org/features/2020/texas-coronaviru...
Excess deaths aren't a reliable indicator of COVID-19 impact because of post hoc ergo propter hoc problems.
Am I missing something? It doesn't look like this has any data on hospitalizations. The graphs are all based on "projected" data.
EDIT: i must have missed it, from a comment below, "It says 'Among the 40 states that have consistently provided data'"
If I sound nervous, it's because I am.
"A number of friends and visitors to the site have noticed a discrepancy in how we count “new” cases and how the state of Georgia is counting new cases.
For example, on April 21st, Covid Mapping Project reports 934 “new” cases. Covid Tracking, a companion site run by a team from The Atlantic, doesn’t report differentials but they do report a day by day tracking of cases. On April 20, they report Georgia had 18,947 positive cases and on April 21st they reported 19,881 cases. A differential of 934. Same as us.
However, if you got the GA DPH website (where most of us source our data), you’ll see 10 “new” cases. That was later revised to 84. And then 138. It's a moving number that will likely change every day for at least a week. Quite different than the 934 we’re reporting."
Here's the dashboard I use: https://dashboards.securedatakit.com/public/dashboard/e430d0...
The U.S. has 86,541 coronavirus deaths. South Korea has 260. Even if we adjust for population, we expect the US to have 1700, were we following South Korea’s example.
How is it that people have just swung fully into embracing our having 50 times as many deaths as we could have prevented? How is it that people believe the rush to “reopen,” without implementing measures at the level of countries like South Korea and Taiwan, won’t result in an even worse outcome?
So one way of looking at it -- no matter what we do, some large percentage (say 70%) of the population will get infected, and a percentage of them will die. By continuing the lock downs, those infections get pushed down the road a bit but they won't be prevented. The only thing that lock downs are good for is to keep the health system from being overwhelmed (where a larger percentage of those infected will die due to lack of resources).
Now my question, is the above a reasonable set of assumptions? I don't see it spelled out like that anywhere, but reading between the lines that seems like what the pro reopening people are saying.
Now, what I'd personally like to see, is instead of a full re-opening, is an order that says any job that can be done remotely, the employer must allow the employees to do remote. Because there are a number of cases I've seen reported here where people were forced into the office just because management wanted to see butts in seats.
My gut is that short of more direct Federal aid to individuals (in the form of checks), we'll see people forced back to work as businesses re-open. Consumer spending will still be shit, since no one but fools will be going to football games, theaters, anywhere with crowds. And people will be clinging to their money. So the velocity of money will dwindle.
And unless the government indemnifies businesses against lawsuits arising from COVID infections, businesses will be slow to re-open as well. Wrongful death lawsuits for failing to provide a safe work environment will be very popular.
Yes, except it's only part of the argument. The other big one is, the economic shutdown is doing significant harm itself: not just lack of funds for rent/mortgage, but also food shortages, delayed surgeries, and so on.
The argument is that the economic shutdown combined with the virus deaths only being delayed and not stopped is going to cause significantly more harm than just stopping the lockdown.
(Hour-later edit, fix the wording at the end)
I'll point out at the current rate of infection in California it would take over 100 years to get to herd immunity levels. We will have a vaccine next summer. And before that we will have enough contact tracing an surveillance in place. As well an a vast scale the public for the most part is now aware of the dangers and cautious.
People are blaming/crediting governments with total control of the outcome, when it's far more likely that yet-unknown environmental, genetic, etc. factors explain much of the variation between countries. That doesn't mean we should excuse government incompetence, but we shouldn't assume the same measures in a different country would have yielded the same death rate.
ETA: If you think something is wrong here, please comment in addition to downvoting. Do you think Japan's response was actually better than Italy's, or Spain's, or any of the countries with much tighter restrictions and many more deaths? If so, why (beyond the circular argument that they have fewer deaths)?
For example, is Reff < 1 in Korea only because of contact tracing? They're doing a commendable job with that, and it's certainly helping, but by how much? Is their voluntary physical distancing actually responsible for most of their success? Or some yet-unknown factor?
I'm not sure how anyone could answer that now with confidence. That uncertainty shouldn't stop us from implementing the best ideas we have, but I believe it should be acknowledged.
As far as I see it this is exactly this threads OPs point. i.e The same measures won't reduce the death rate to the same amount.
But this does not mean they have not reduced the death rate and aren't good advice everywhere.
Singapore has 21 deaths as of May 14. Adjusting this number to the US population gives 1,117 deaths. The US has 89,908 deaths and rising. I would gladly trade with Singapore.
In particular, how do you explain Japan? They barely locked down, and they're missing most of their cases so the contact tracing is at best reducing Reff by ~20% (per below). I personally suspect that Japan has benefited from some combination of masks, a culture of hygiene and distance, and the weather; but I'm far from confident, and disappointed to see others behaving with such apparent confidence in the face of all we don't know.
Hopefully scientists will figure out that chrysanthemum pollen or TB vaccinations reduce the severity, and then we can deploy that tool here. Until that happens: we have to act like this disease is going to explode. That isn't confidence, it's just common sense.
The cost of the coronavirus in mortality is massive, but the cost of the reaction is too, even just in lives as above. Some measures seem clearly cost-effective to me (masks, not standing too close, keeping kids in stable groups, etc.), but others (closing schools completely) do not. It's not hard for me at all to imagine a response that causes more societal harm than the virus itself.
Imagine that the virus did what it did in NYC or Lombardy, but did it in every city in the US. Given our understanding of the virus, this is very possible. Now forget the statistics: just think viscerally about what that would mean for schools, office jobs, retail businesses, hospitals. You'd get all those deaths plus the economic damage of a lockdown. This isn't an either/or.
Basically: if we're not going to control this thoughtfully, we'd better pray that some as-yet-unknown effect stops the spread most places, or that people naturally socially-distance enough to stop it. I hope one of those happens, maybe it will happen, but I have zero reason to believe it will right now.
And I certainly agree the virus alone has economic impact, even if the government (recklessly) took no steps to control it. So do the math with the incremental cost in GDP for whatever intervention you favor, but remember that every extra percent of lost GDP is ~30k dead 75-year-olds in lost health care spending under the model above, even before considering any other economic impact. There is no safe path here, just different and uncertain tradeoffs.
In addition, they have a highly pro-social society that took it upon itself to do things like engage in universal mask wearing and social distancing measures.
That being said, we’re still very early in the course of this disease, so it’s hard to know how the responses will measure up in terms of durability. Since many epidemiologists are predicting a dark course this coming winter, I suppose we will see, but my bet would be on South Korea, given that their response is more firmly rooted in their experience with SARS.
The US has neither the aggressive government response, cluster-based methods of Japan, nor an especially pro-social society. You can’t really compare “half-hearted” measures across countries absent this context, but if even if you wanted to, it’s quite clear that our “half-heartedness” is far more deadly.
I certainly agree that stuff like Japan's existing mask culture (and general culture of hygiene/distance) is helping, and that those can and should be emulated. That's a separate question from their current government response, though.
And I worry about winter too. Everything about seasonality is speculative now (and it's shocking how little we know about why a cold and flu season exists), but I can't help suspect that many countries that think they've eliminated the coronavirus have an unpleasant wave ahead.
And thanks for the reasonable counterpoint--whoever downvoted you should explain themselves too.
> In 1935, with funding from the Rockefeller Foundation, Japan opened its first public health centre (PHC) in Tokyo. The country then put in place a programme that led to the building of another 187 nationwide. They survived the war and the occupation. But the thing of note is that before and after the war, their priority, says Taniguchi, was always to “stay watchful all the time” for the emergence of TB cases. If one was found, they were tasked with rushing to the patient's residence, checking for clusters and sterilising the house
Seventy-five years on, 469 PHCs are in operation across the country, with each manned, on average, by 64 medical professionals, including one to two licensed doctors. They still locate clusters, track links and conduct tests. It is this “accumulated wealth of expertise, rarely found elsewhere” that has made the difference. Japan has not had to rely on mass testing strategies because it’s health care history had already left a cluster-crushing strategy embedded in its system
https://moneyweek.com/economy/global-economy/601264/cluster-...
Maybe you're saying that R < 1 almost everywhere in Japan, and they're disproportionately identifying the few places where it's > 1 ("circumstances where things can grow out of control") and intervening? I could believe that; but that still fails to explain why R > 1 there only in a few hotspots, while NYC ended up with ~20% infected even after lockdowns.
From what I understand, this is the thinking guiding the Japanese government’s response and their zoned lockdowns. Whether or not this is true, or will continue to be true, is unclear, but as you’ve pointed out, they’re roughly on track with South Korea thus far.
To your second point: no one really knows. Looking at the example of other countries, universal, correct use of masks does seem to be an important mitigating factor. This was not practiced in NYC until well into the outbreak and social distancing measures were slow to be adopted as well (either by people taking it upon themselves or by government decree).
I certainly don't mean to suggest we shouldn't be trying our best ideas to control the virus, even if we're not sure whether they'll work or why. I just don't like the idea of low death counts as conclusive evidence that whatever a government is doing is good, especially to the extent it lets authoritarians that have seen low death counts (for reasons related or unrelated to their response) like China promote their models to the rest of the world.
Unfortunately, we can't bring 80k down to 1700 by just saying the number ought to be lower or being really sad about it. If there's some way to immediately stop coronavirus deaths, we don't know what it is and we don't reasonably expect to find it soon.
In every region of the world, countries with lots of medical and biotech resources are generally doing worse. Korea, China, and Japan are all doing worse than Thailand and Vietnam, Eastern Europe is doing better than Western Europe, and Canada and the US are the worst hit countries in the Americas. The list of countries which have seemingly controlled the coronavirus despite being unable to control tuberculosis and malaria is very long.
South Korea was well prepared by SARS, and did extremely well on (2)+(3). Other Western nations had good policy responses, but didn't have the resources in place by February/March.
However, many countries that were unprepared have since made significant efforts to increase resources and deploy successful policy. A look at the epidemic curves in May shows that results are diverging even between Western countries that had similar initial growth. Some, despite starting with a similar initial case spread, are seeing significantly greater declines than others. Some are deploying testing resources at a higher rate.
Ignoring current performance and lumping all Western nations together completely ignores all of this and makes it seem like humans are helpless to control epidemics, which is an absurd claim.
There seem to be two camps on this topic. One says that this is great, and we should continue every effort possible to stop the spread, presumably for an unlimited number of months or even years. That means staying locked down with minimal economic activity and somehow dealing with the consequences as they come. This is the "wait for the vaccine" camp, which I find hard to join because even if a safe vaccine can be created, it could be a year or two before it is sufficiently tested and deployed. California seems to be leading this camp in the US.
The second camp says that excess healthcare capacity means we should begin reopening the economy until capacity becomes strained. This is the "let the virus spread" camp, and implicit in the argument is that we should allow more people to die from the virus. This camp is winning. Most states and Western nations have started to ease restrictions knowing full well that it means the virus will spread. The hope is that less stringent restrictions can still allow the rate of transmission to be controlled enough to avoid catastrophe.
(There is also a third, fringe camp in the US of right-wingers, and while it is easy to dismiss them, they are actually winning in states like Georgia and Texas)
My understanding at the start of lockdowns was that these were intended to "bend the curve" so we could get control of the situation and then allow the virus to spread at a rate that would not overwhelm healthcare capacity. But more and more people I know seem to be in the "wait for a vaccine" camp. Personally, my imagination struggles to comprehend what another 12+ months of lockdowns would mean for civilization. The loss of life from disease, though obviously abhorrent, is actually a concept my brain can manage (which is not to imply it has more merit, just an honest observation).
And if California is truly waiting for testing capacity, they may be waiting a long time because their rate of testing is extremely low compared to places that have opened up. New York has twice the per capita testing as CA and they have already began opening up parts of the state (despite being hit many times harder by the virus). Even NYC is about to hit the criteria for some reopening. Meanwhile, LA announced they would continue their lockdown until August. That sure sounds like they are on a "wait for a vaccine" plan.
The European nations that have eased restrictions have higher per capita testing than most US states. But they do not all have robust test and trace programs. Denmark opened up weeks ago, including many schools, but only last week announced a test and trace program.
a) You give the contact tracing system a complete and accurate history of your contacts.
b) You actually self-isolate when told you've had contact with an infectious person.
The civil liberties crowd will fight tooth and nail against a such that it is not possible to compel b. In the absence of enforcement, the conspiracy theorists will ignore it and the merely lazy will cheat.
Not even a competent attempt at contact tracing is going to work on the American people.
Of course, they had politicians who reacted quickly, firmly, universally, and most importantly early. In the USA, Corona was allowed to grow much much more and far fewer resources were brought to bear in an effective way against it. So its going to be much more difficult to get back under control, but not impossible, if we can turn it around.
Whats really happening besides that though is that the world is watching the self proclaimed world leader be the least competent and most impotent in their response, and throw its hands up and give up on good strategies like contact tracing. Its embarrasing.
While it's easy to take political sides with this, that's in the past. We could have had better CDC leadership than Redfield. We could have had better leadership on the NSC than we had after the advisory committee was disbanded. We could have had better presidential leadership. We could have used testing kits from Germany until US kits were effective.
There's a lot of shoulda/coulda/woulda. But here we are.
In much of the country that isn't NYC, including many of the places aggressively reopening, the first wave hasn't even peaked yet, so it's not so much that we missed the boat on the first wave and are now into preparing for potential later waves, we’re still actively screwing up the first wave.
I wouldn't say "no one":
> Los Angeles Mayor Eric Garcetti said Wednesday that the city will "never be completely open until we have a cure,"
This is also a mentality I've seen numerous times on the web, including this site.
This is absurd, and indicates a politician pandering for the next election.
For ebola, with 25% to 90% mortality, it can be justified.
For corona, with under 0.1% mortality for working-age people, it's madness.
"The average EVD case fatality rate is around 50%. Case fatality rates have varied from 25% to 90% in past outbreaks."
-- Many of those making these comments are probably WFH with some level of normalcy so they don't really see how and why a lot of people are anxious to open things back up.
-- In the same general vein, they see a push to reopen as part of an agenda that puts "the economy" over people for the benefit of Big Business/The Rich.
-- They think there must be a near-term technological/policy fix if only the bad politicians would stop messing things up.
California’s plan has multidimensional standards, it's not accurate to say it's “gated by test and trace”. But yes, the ability to effective monitor leading indicators of caseload and isolated the infected and exposed is fairly significant, and test is key to both and the trace part of test and trace an important factor in the second.
Speak for yourself. The US doesn't have a vaccine, enough testing or tracing.
Wishing doesn't make it so. End the lockdown now.
Frankly speaking, the idea that opening up will somehow fix everything isn't true. People are still going to be afraid to go to public places, and we have multiple examples of cities or areas that decided to open up too early getting slammed by a second wave of infections (See: Hokkaido, Texas).
So yes, in this case it's a 'moving goalposts' situation because our knowledge and way of dealing with the coronavirus evolves over time. That's the entire point of slowly opening up with a viable test infrastructure. If you don't adjust as the situation evolves, then you get hit with a second wave making the lockdown pointless and prolonging the economic effects. Especially if people start suing local businesses when they get sick for not following the guidelines.
Where are you seeing that Texas is getting "slammed by a second wave"? I'm looking at Worldometers, and their daily new cases looks relatively flat. Apparently there was a big jump yesterday, but the state is claiming that correlates with a much higher rate of testing. (This is part of why we should look at the 7 day average, not just the day by day numbers.)
If their rate does happen to be increasing, what makes that a second wave, vs them just not hitting the peak of their first wave yet?
As for the second wave, the fact that we've hit a large jump in cases is specifically what I'm talking about. It does not bode well for the next few weeks, especially because there has been little to no effort from the state to actually implement measures to ensure that we keep the curve flat. Local health officials have also been raising the alarm on both our lack of testing infrastructure and the idea that a second wave is already occurring.
https://apps.texastribune.org/features/2020/texas-coronaviru...
They're not hiding bodies, so the excess fatality rate would reveal any lies they're telling.
I haven’t seen any evidence showing Texas getting slammed by a second wave. All data sources I have seen, including the linked article, show a stable rate or relatively small uptick.
>My understanding at the start of lockdowns was that these were intended to "bend the curve" so we could get control of the situation and then allow the virus to spread at a rate that would not overwhelm healthcare capacity.
Yes. The standard argument went along the lines of we have to limit movement and interactions for a while so that we can "flatten the curve" and not overwhelm the hospital system. Now that that has happened in most cases, though, the goalpost has moved towards the availability of a hopefully effective vaccine in the next year or two [ADDED: and/or maybe effective testing/tracing whose effectiveness in general is TBD]. Which, even if we assume some behaviors change, doesn't seem like a practical long-term strategy.
Only if you don't look at countries where it seems to be working pretty well, mostly in Asia.
My comment wasn't "all countries in Asia", it was about countries in Asia using contact tracing well.
It was about health care capacity. It was also about testing capacity. It was also about producing/procuring PPE and logistical planning for all the things to come. It was also to learn more about this thing, since there are so many unknowns.
The biggest dividing line here seems to be masks. Countries whose leaders initially told them not to wear masks are struggling to achieve universal masking compliance. Most seem to have given up on being aggressive and are now having these debates of complacency.
Bay area: https://www.mercurynews.com/2020/04/22/coronavirus-bay-area-...
San Diego: https://www.10news.com/news/coronavirus/what-to-know-about-t...
Los Angeles: https://losangeles.cbslocal.com/2020/05/14/coronavirus-covid...
From the first article I linked: "Are you leaving the house? Facial coverings are required to be worn by anyone older than 12 while waiting in line or inside any essential business, using public transit, riding in a taxi/Lyft/Uber or seeking healthcare, as well as anytime you come within 6 feet of someone from outside your household."
https://www.usatoday.com/story/opinion/2020/04/04/czech-gove...
Sure, the US isn't as advanced or well-run as... the Czech republic, but we should at least try and aim for what those kinds of countries can accomplish.
https://en.wikipedia.org/wiki/Defense_in_depth_(computing)
They're a bit uncomfortable, but so are ventilators...
Given the terrible situation in Brazil (and Mexico), is it not possible that the whole of the Americas could be cordoned off, while Europe and East Asia continue with efforts to suppress the virus?
I suspect a reasonable middle ground strategy is to aim for just some immunity (say 15% or 25%) while not overwhelming healthcare, after which tracing and isolation has a chance of working while also allowing a somewhat functioning society (Note: I mean for dense cities. This could still mean 5% in a country, as in Spain for example). The hard part is ensuring that as few as possible of those that are infected are in risk groups. This is something that proved difficult in many places.
There is a set of restrictions that allows the aggressive testing/tracing/isolation work for a naive population, and a different set of restrictions that is sufficient for to contain the spread in a population with 20% immunity. My feeling is that the former set of restrictions isn’t viable in democratic countries, or sustainable for the time we are talking about (1-2 years) without leading to something that costs more than is won.
I mean, I know Iowa is a really dense place where everyone uses public transportation, but maybe they could at least make an effort...
This isn't some small, statistically insignificant difference. It's a massive difference in the number of cases and deaths.
They have 300 deaths and 10k confirmed cases, suggesting they have confirmed 10% to a third of all cases.
It’s also impressive that they managed to keep the outbreak so small without extreme mitigations like other places with similar outbreaks.
I don’t think that most western countries can emulate it though. Next time, perhaps (this is South Korea’s second SARS outbreak).
They certainly can't if they don't even... try.