Projections for Peak Dates Corona Virus for Each State

[sheehs1]

I have not seen this posted here. A friend sent it to me. I have not read all thru the website of Axios.

You can click on your state to see the projected peak date and hospital resources.
Works better on my laptop than my iPhone. Touching with my finger on iPhone doesn't bring up the projection but clicking on it on my laptop does. (strange). The Data appears to be from the same IHME Covid-19 forecasting study used in last nights update by the Task Force team.

https://www.axios.com/coronavirus-s...hZmGZnKQWPLc8-6L5x6fwLQ7aAJDEFKCZ10WguhWUb4Yg

Mods if you need to move this to another thread feel free. I wasn't sure where to put it.

 

[KCGeezer]

Interesting. DW told me about this one.


https://covid19.healthdata.org/projections

 

[ExFlyBoy5]

Bookmarking to come back in 3 months to see how wrong these will probably be.

 

[audreyh1]

Already posted on another thread.

 

[sheehs1]

Already posted on another thread.

Which thread? I had not seen it but it is difficult or should I say time consuming to sometimes catch everything in the longer threads.

 

[sheehs1]

Interesting. DW told me about this one.


https://covid19.healthdata.org/projections

Thanks! Yes..once this is over it will be interesting to see how these models hold up. After all, they are all just models at the moment.

 

[audreyh1]

Which thread? I had not seen it but it is difficult or should I say time consuming to sometimes catch everything in the longer threads.

Sorry - it was the healthdata.org chart that was posted and discussed on another thread.

 

[Exit 2024]

Thanks! Yes..once this is over it will be interesting to see how these models hold up. After all, they are all just models at the moment.

you need to do screenshots then, they are adding actual data to those models daily(?) and curves change a lot, especially for states with less then 100 total deaths

 

[audreyh1]

you need to do screenshots then, they are adding actual data to those models daily(?) and curves change a lot, especially for states with less then 100 total deaths

Yes, I did exactly that - I printed today’s to PDF and labeled it Apr 1 because I realized it would change daily.

 

[sheehs1]

Sorry - it was the healthdata.org chart that was posted and discussed on another thread.

No problem. I have been very tuned in to this whole thing, listening to every Task Force briefing, reading and driving myself a bit nuts I guess. I had not seen peak time projections for each state. A news person asked this question of the Task Force the day before I posted this thread. They did not really answer except to say every state would have their own curve but did say they had models for each state.

What i found interesting was the peak is a month apart in some cases. In Virginia, which is where I am, it is mid May. In PA where the other in-laws live it is Mid April. The in-laws are not taking it seriously although after a conversation with my son-in-law, also in Virginia, I think they are now.

In my city, which is small at around 8,000 in the city and 25,000 in surrounding county we have few grocery options. Just had 3 from the local WalMart grocery pick up line test positive. They shut down the grocery curb side pick up -maybe for a day or two to sanitize.

 

[Chuckanut]

Bookmarking to come back in 3 months to see how wrong these will probably be.

Your comment reminded me of this quote from Theodore Roosevelt:

It is not the critic who counts, nor the man who points how the strong man stumbled or where the doer of deeds could have done them better. The credit belongs to the man who is actually in the arena; whose face is marred by dust and sweat and blood; who strives valiantly…

All I would add is in today's world we would include the strong women in the above quote.

 

[sheehs1]

you need to do screenshots then, they are adding actual data to those models daily(?) and curves change a lot, especially for states with less then 100 total deaths

Good idea! Much will change with more tests. Let us all hope the curve is flattened and remains flat thru out.

 

[Midpack]

That is interesting thanks. Our state peaks in the second week according to both sources above, hope it's true.

 

[JoeWras]

Thanks! Yes..once this is over it will be interesting to see how these models hold up. After all, they are all just models at the moment.

Re: https://covid19.healthdata.org/projections

I took a screen shot of the April 1 projection for NC. They have since updated it late yesterday.

The update has changed significantly for the better. The peak is now projected much earlier (April 15 vs. April 27), and is much lower. (30 vs. 51 deaths per day.) Low enough to not overrun the state ICU total.

It is kind of like hurricane forecasting. They always say: "Don't focus on the center line." I guess you can say the same here. The lines are in their "shaded area" so far. Fortunately, below the center line, significantly.

Hopefully people will continue to cooperate and keep the peak down. We have a doctor friend who works in an "urgent care" setting, and she's ready to drop because of the constant intensity. People are very freaked out, and she's seen her share of cases looking to be serious. That said, she still feels like they are managing so far.

I've also noticed our local news is having a hard time coming up with sensational headlines lately. That's a good thing.

 

[USGrant1962]

Re: https://covid19.healthdata.org/projections

I took a screen shot of the April 1 projection for NC. They have since updated it late yesterday.

The update has changed significantly for the better. The peak is now projected much earlier (April 15 vs. April 27), and is much lower. (30 vs. 51 deaths per day.) Low enough to not overrun the state ICU total.

It is kind of like hurricane forecasting. They always say: "Don't focus on the center line." I guess you can say the same here. The lines are in their "shaded area" so far. Fortunately, below the center line, significantly.

Hopefully people will continue to cooperate and keep the peak down. We have a doctor friend who works in an "urgent care" setting, and she's ready to drop because of the constant intensity. People are very freaked out, and she's seen her share of cases looking to be serious. That said, she still feels like they are managing so far.

I've also noticed our local news is having a hard time coming up with sensational headlines lately. That's a good thing.

Yes, big downward revision on that model. Yesterday the total deaths median was, IIRC, ~92K [actually 93.5K]. Today it is down to 82K - good trend.

The death modeling on that one seems reasonable, but the hospital resource use (beds/ICU) forecasts have already been proven to be bunk. Real demand has been about 10% of that model.

And FWIW, the headlines are getting more political, so I guess that is a good thing?

 

[Sunset]

....
In my city, which is small at around 8,000 in the city and 25,000 in surrounding county we have few grocery options. Just had 3 from the local WalMart grocery pick up line test positive. They shut down the grocery curb side pick up -maybe for a day or two to sanitize.

Confirms to me, to continue spraying all my groceries with diluted bleach, rub them, and rinse off, and quarantine the canned/non-refrigerated stuff for 4 days.

Maybe I'm wasting my time, but better safer than sorry.

 

[dixonge]

I think what these charts and adjusted peak dates are indicating is this:

Social distancing *and* lockdowns work.

I like the chart on this page (the big one down at the bottom) since it breaks it down by state and compares projections done on Apr. 1 and Apr. 6 - five days can make a huge difference!

Is Your State Flattening the COVID-19 curve? The Latest Data

Also includes some charts comparing NY and CA - seems as if CA did a better job...

P.S. That link is a pro-marijuana site - not really sure why *they* are covering the issue so well *shrug*

 

[stepford]

While this is all good news, I remain a bit skeptical. Figures like daily case counts and deaths have a lot of noise in the data - day-to-day variations that don't imply any long term trend. A day or two of lower numbers is undoubtedly good to see, but I'd want to see at least a week long trend before I'd really place a great deal of faith in new projections.

 

[cathy63]

I am holding my breath to see how things shake out during the first part of this week. Last weekend I got my hopes up after a couple of days of lower numbers being reported locally, only to see them explode to new highs on Monday and Tuesday. It turned out that a lot of the administrative processes for counting patients and test results and reporting them up to the city/county/state don't happen on weekends and those cases get grouped into the Monday/Tuesday results. There was a quick mention of changing that in one of the press conferences, but I don't know if they can really require private facilities to pay a lot of admin people overtime just to get the data a little faster.

You can see the weekly hump shapes on the new cases chart on the L.A. Times tracking page: https://www.latimes.com/projects/california-coronavirus-cases-tracking-outbreak/

I suspect that California is not the only state with this pattern.

 

[Montecfo]

What i found interesting was the peak is a month apart in some cases. In Virginia, which is where I am, it is mid May. In PA where the other in-laws live it is Mid April.

Sheehs,

I found that strange also (also in Va). But the 2nd chart in the thread above showed Va peaking April 20.

That seems more logical to me given the flattening of new numbers infected.

 

[JoeWras]

The agents of plentiful clicks and bad news have also noticed the IHME models are getting optimistic. So, why not throw some water on it? They have a different view for NC, with numbers much higher, even if social distancing as is stays in place.

https://abc11.com/health/models-predict-when-nc-could-see-a-peak-in-covid-19-cases/6081002/

On the other hand, if the same or similar policies remain in place, the estimates indicate that about one-third of that -- an estimated 250,000 North Carolinians -- may be infected by June 1.
...
Epidemiologist Kimberly Powers explained why these models are vastly different than the widely-cited IHME models from the University of Washington.

For one thing, the IHME model showed that on Sunday there would be 1,300 hospitalizations in North Carolina, when in fact there are only 270 currently.

So, Powers said, the model was predicting a much earlier upswing than we've actually seen.

Additionally, she said the IHME model was based on a very, very strict social distancing policy based on what was happening in Wuhan after the outbreak there and not the actual guidelines being implemented here.

 

[dd564]

In MN, our governor's model is far more dire than any model found online.
Better safe than sorry, but his was predicting 50,000 deaths in MN alone, and currently we are under 1,000 cases yet.

 

[brewer12345]

Lower than forecast deaths is great. That said, peaks do not tell you when we can, say, go out to eat at a restaurant without fear of infection. All this flattening the curve will save lives, but it doesn't bring the effects of the pandemic on the economy to an end sooner: it prolongs it.

 

[stepford]

All this flattening the curve will save lives, but it doesn't bring the effects of the pandemic on the economy to an end sooner: it prolongs it.

Epidemiologically yes, but not necessarily economically. It's not at all clear to me that the economy would recover more rapidly from, say, 1M deaths over a few months than 250K deaths over 18 months. These are obviously just wild numbers being thrown around, but my point is that shock has an effect on duration of economic downturn. A longer epidemic with a lesser shock may give rise to an overall more rapid recovery.

 

[gwraigty]

Today's press conference is still going on, but a reporter asked the health director why the IHME shows an earlier peak (April 12) than they keep forecasting (mid-May). Well, she's going to address that in more detail at another time. Sigh. But now it's been revealed that the Cleveland Clinic modeling is now predicting a best case scenario that we'll hit our peak in Ohio from mid-May to mid-June! I'm at a loss for words to still sound civil about this.