And most recently (June 23) it was up to 88%. And the trend line is moving upwards. Given that Arizona’s cases started ramping up in early June, and hit a record of over 3500 daily cases on June 23… I will guess that Arizona will reach 100% ICU capacity by July 10 or so. And if the number of infections continues to rise… Well it’s not going to be pretty after that.
One important thing to note is that in some of the bigger states, like Florida and Texas, the average age of infection has been dropping steadily. What that means is the death rate should be going down, given how mortality has varied by age.
Agreed. COVID is far, far less lethal to younger people compared to the older people. See this data from the CDC. As of the end of last week, people aged 0-44 represented 2.5% of US COVID deaths (under 35 represented 0.8%). People 65+ represented 81% of US COVID deaths (85+ represented 33%).
Also consider the CDC’s cases by age (about halfway down). People under 44 represented 47% of confirmed cases. People 65+ represented 20% of cases.
Putting this together:
Aged 0-44: 47% of infections and 2.5% of deaths
Aged 65+: 20% of infections and 81% of deaths.
Note: I’m not making value statements here, just providing statistical context. It seems like the answer to the thread’s question is simply that COVID doesn’t kill young people at anywhere near the rate it kills older people. And the new wave of infections is taking place in a much younger population (Texas average age under 35, Arizona average age ~35, Florida average age 33).
I wondered about this myself, so I was glad to see a topic already opened. I’m frustrated with the reporting on this subject – the story of the week has been “surge in cases” but that’s a very wobbly statistic to get worked up about, as has been adequately explained here. It seems like what we should be focusing on, is number of hospitalizations and number of deaths – they aren’t affected by the amount of testing we’re doing.
I think this is the biggest factor. The places that are surging now benefit from the lessons learned in other areas, and one of the big ones was in how to protect old age homes from this pandemic.
It’s a lot easier to prevent a breakout at an isolated facility, where you have much better control over who comes and goes, and have professional staff on hand to do a lot of the work needed. The residents don’t need to go to work, or shopping, and so aren’t likely to be exposed if the staff does their job well.
So the early large numbers of deaths we saw in NY likely won’t happen in Texas.
The problem is, these were the deaths that are easy to avoid. The deaths we’re seeing now will be harder to stop. As new infections continue to grow, the number of deaths will also start to grow again. Take a look at that 91-DIVOC site. Most of the decline in deaths is from NY and NJ. There’s a lot of other states that have been essentially flat on numbers of deaths per day and some that are seeing increases. There’s a few with modest declines, but that can change quickly.
We saw the same thing with the number of cases - the overall US numbers were declining, largely because of NY and NJ declining. But eventually the other states began to catch up, and their increases have now wiped out all the gains made by NY and NJ. More than wiped out, in fact, because the overall US numbers are now surging again, and are higher now than they were at the peak of NY and NJ’s worst weeks.
I think the daily death numbers we’re seeing now are a temporary lull, and they’ll start rising again within a few weeks.
So the cover-up is happening in plain sight - we know they’re doing it. So what?
We still don’t have the numbers. So the numbers themselves are being covered up, even if the fact that they’re doing it hasn’t. So it sure seems to me like the governors of FL and NE are getting away with covering up the numbers.
(DeSantis, the FL gov, also blocked any further release of ICU bed occupancy because the numbers kept going up.)
OTOH, it takes more than a few states screwing with the numbers to make much of a dent in the overall trend. At its peak in mid-April, the coronavirus was killing over 15,000 people a week; now it’s in the low 4000s. (Still too many, but damn sure better than it was.)
Florida’s official numbers are hanging around 40 Covid-19 deaths/day. Suppose they’re really double that. That adds another 280/week to the national numbers. Pushes it up to the mid-4000s. Nebraska is a much more low-pop state than FL, with a more geographically dispersed population. They may be fudging their numbers too, but is it going to add even 10 deaths a day to the national numbers if they weren’t?
I think you’ve got to have Texas and a couple of mid-sized states also fudging the numbers big time to get the death rate over 5000/week. Maybe it’s happening and nobody knows about it, but secrets like that are hard to keep.
There is also a minimum two week delay between a rise in cases and rise in deaths. People don’t get Covid and consequently proceed to topple over and fall dead. They slowly get sick, get sent to hospital, get worse, get put on ventilation or otherwise are treated, struggle to stay alive and, eventually, die.
Quarantine allowed the numbers to start going down and then we opened things up. The deaths will continue to go down for two weeks and then switch direction to match the infection rate.
Given that it is largely rising among the young, there could be an even longer delay than 2 weeks since the disease needs to work through the young idiots - who will suffer no ill consequences - until it can make its way to the elderly, through them, and start killing again.
Reviving this thread because the predictable appears to be happening, despite some wishful reading of the trend lines in earlier posts.
Overall deaths in the US are still dropping but when broken down to the state level, deaths are now rising in Arizona, Texas, Florida, and several other states. So, the lag between the (now rapidly rising, not flat) infections and deaths is real and playing out right now. We can hope that between demography and better information on treatment, it doesn’t spike as heavily as a few months ago, but the increase is real. And some hospitals in some of those states aren’t exactly ‘overwhelmed’ yet but hospital capacity is being seriously tested in several areas with potentially worse to come.
So, unfortunately, the answer to the OP has been shown fairly conclusively to be “wait a few weeks” + “national numbers in the US are less useful than regional”.
A good resource seems to be covid19.healthdata.org, which estimates the real infection rate, as opposed to simply listing the infections uncovered by testing, which is going to be badly skewed by the testing rate. For the US, it estimates that at the (so far) height of the pandemic, late in March, nearly 250,000 people got infected per day, leading to the death rates then observed, while now, around 80,000 daily infections occur, of which ~50,000 are detected. Hence, we shouldn’t expect to see similar death rates for some time yet.
In fact, the site includes predictions for different scenarios—prediction based on current trends, with a significant further easing of mandates, and with universal masks. In the worst case, if these predictions are accurate, we should expect 1,000 deaths/day by late October.
I haven’t really had a look at the methodology by which these predictions are obtained, however. If anybody has some insight there, I’d be grateful.
That resource has been consistenty wrong, underplaying predicted death rates by around 50%.
Whilst it had altered its predictions, and no model has been especially accurate, this one has been wrong in ways that even a non-expert could have forseen. It originally predicted in June around 96k deaths by August, and at the time there were already over 80k deaths with an average of over 1000 deaths a day. Later this was modified to around 120k, when it was already obvious that this figure would be reached by mid June.
It has been a very poor predictor.
For example - right now you state that it predicts around 1000 deaths per day in October - but today there were close to 900 deaths and the huge spike in positive tests is only just starting to reach the 3 week phase.
We can expect a significant rise in deaths that may level off a little below 2000 deaths per day for the next month or so - and that would be based upon the existing people who are currently infected.
This does not include future infections - which is to say that what happens after that depends upon the number of new infections over the next month.
Even with very limited expertise i.e None whatsoever, I predicted 130k by July 1st and was hopeful that it would stay below 150k by August 1st.
That was based upon other states picking up on lessons from NY and NJ and improved clinical practice and a leveling out of the rate of infection - I fear that hope is gone due to the opening up and protests and sheer idiocy of gatherings.
We saw in NY the funeral of a leading Rabbi during the height of their pandemic, the Jewish community were urged by Cuomo to restrict the social gatherings in the wake of the funeral social events. This was treated by that community as a form of anti-semitic attack and resulted in large crowds mainly drawn from the Hasidic community.
The result was large gatherings that were inspired by social solidarity and completely ignoring the epidemiological advice - the result was predictable.
What has happened throughout the recent upsurge is a defiant attitude based upon identity politics, selfishness, ignorance and sheer stupidity but the surge of infections absolutely dwarfs what took place on the East Coast.
The idea has been that this surge of infections is somehow different, that it is largely younger people, that testing has detected more cases in populations that are likely to affect more resistant to Covid.
Anyone who believes this is being incredibly naive, if US stays below 180k deaths by August it will only be a matter of luck.
My own guess based upon nothing more than observation is that the number of US deaths will be between 170-180k by August 1st and by October 1st it will be between 200k to 220k.
Unfortunately my guess seems to be as good as anyone elses, which just shows that policy is being made by experts who have as little clue about the situation as any Joe off the street - that alone is not much comfort, I would prefer my predictions to be totally wrong and theirs to be much better.
Early in the pandemic, that was one of the more accurate models, actually, not consistently wrong.
[Looking at the graphs on 91-divoc.]
Arizona without doubt. (Although it also should be noted that the cumulative mortality rate continues to drop.) They started their recent surge earlier and have had enough time for it to show up in death rates.
Texas - their weekly average daily death rate has continued in the range of 0.7 to 1.3/million for over two month and is now 1.1. Not yet any meaningful rise.
Florida - 2.1 is up from their low 2 weeks ago of 1.4 but still premature to call it a dramatic rise.
All U.S. South (regional) - peaked at 2.8 nearly 2 months ago and now at 1.6 (had been 1.5 briefly but pretty much the low).
I expect death rates to increase over the next weeks, would be shocked if it did not to at least some degree, especially for the Southern region, but a statement that the question has been “conclusively” answered is unfounded. For most of the region a definitive conclusion still needs to wait a few weeks.
Wouldn’t an excess amount of deaths only happen on overcapacity, and not approaching capacity?
Early in the pandemic, they were using what essentially looked like a bell curve to model it, and sure, that looked okay when things were increasing. But it’s not a symmetric distribution, it’s right-tailed, and they definitely overestimated the speed of decline.
Or they could just be telling the truth? When they said they got the outbreak under control, it seems like they actually got the outbreak under control. When new outbreaks started popping up, they dutifully reported all of the new outbreaks. When they got the new outbreaks under control, new cases stopped popping up.
Right, and all the other models were plugging random variables into their models and wildly over predicting, for the most part.
I agree with much of this. Studies I’ve read are beginning to converge on an estimated IFR of around .75 to 1.5%, depending on circumstances. When the community spread is under control, it’s probably closer to under 1%; when ICUs are full, it’s probably above 1%, perhaps considerably higher in the short term. Doctors are beginning to learn how to manage patients to avoid the worst outcomes, which is encouraging, if we can just keep the overall numbers down.
I don’t get what the IFR has to do with what DSeid said.
I suppose pigs could, indeed fly, but their numbers don’t add up. Nor does the fact that they have periodically taken what would be extreme measures if they indeed had the outbreak under control. Nor the fact that they ruthlessly (not that they have any other way) stifle their own highly controlled media whenever someone tries to discuss new or ongoing outbreaks, or really anything that contradicts their narrative. The biggest one, however, is that their numbers don’t add up, however…and by very large margins, I’ve heard up to 5-10 times being touted.
It could be as you say, but the thing is, if this is true it would be the one time they didn’t lie about something like this. I admit, it would be ironic if the one time they are telling the truth there is so much doubt, not only by paranoid CTers such as myself but also by many of their own people. The fact that they are doing the same sorts of things (i.e., lying, stifling their media, etc) wrt the ongoing floods, swine flu outbreak, etc, tends to make me think that this is not the one time they are telling the truth. YMMV of course.