136 New Cases in MS count over 1,000
- Thread starter Dogs
- Start date
Next few weeks are not going to be very encouraging. Stay strong friends. We will get through this.
218 hospitalized today. Up from 211 yesterday but down from 235 on Sunday. Been between 185 and 235 for 6 days in a row.
10 straight days of a flat curve on date of onset of symptoms. These are being modified constantly, but still very encouraging.
View attachment 15731
Basically, what this graph tells me is that we're getting tests back at more variable rates than when people start to see symptoms. 136 certainly isn't great, but only adding 7 to the hospital in that amount is a bit encouraging, especially with it starting to hit a number of nursing homes. I expect the hospitalization number to jump from that news yesterday. Might take it a little whole.
We're definitely not in exponential growth here right now. Hopefully the trend continues.
10 straight days of a flat curve on date of onset of symptoms. These are being modified constantly, but still very encouraging.
View attachment 15731
Basically, what this graph tells me is that we're getting tests back at more variable rates than when people start to see symptoms. 136 certainly isn't great, but only adding 7 to the hospital in that amount is a bit encouraging, especially with it starting to hit a number of nursing homes. I expect the hospitalization number to jump from that news yesterday. Might take it a little whole.
We're definitely not in exponential growth here right now. Hopefully the trend continues.
Last edited:
I am genuinely curious and not trying to start any political ********. Can someone explain to me why looking at "date of onset of symptoms" is a better metric than "confirmed cases"? Date of onset of symptoms is 1) subjective and 2) you are infected days before you show symptoms. And the chart is inherently incomplete (which it's clear about) since there are cases that started in the last few days that haven't been reported. Seems like looking at how many cases are confirmed each day (regardless of onset of symptoms) would give you a better look at rate of infection plus access to tests.
I can't find a chart from MSDH that shows how many cases are reported each day.
Again, not trying to criticize anyone. Just seems to my stats illiterate brain that this is a clumsy way to look at the big picture.
I can't find a chart from MSDH that shows how many cases are reported each day.
Again, not trying to criticize anyone. Just seems to my stats illiterate brain that this is a clumsy way to look at the big picture.
min hopeful that the rural nature of the state will help flatten the curve. But I’m sure there’ll be a spike at some point.
Yes, the hospitalization rate is encouraging without any other better metrics to work from. If MS can keep that rate from spiking too much more, it will weather the storm.
I think this is a a way to say they are way behind on testing, so rather than have big spikes when a bunch of tests get processed, they are providing results in the form of the day the test was administered without coming out and saying it.
It won't be positive if/when the data starts to tell a different story... Just basically reading and interpreting. Although I do lean optimistic.
We had no confirmed cases on March 10th and it took until April 1st to get over 1,000. This next week will determine how well we've done. If it takes three days to add our next 1,000 then we've hit an exponential curve up and will be hammered. If it takes If it takes 6-7 days we've got a much flatter parabolic curve and still have time to at least make sure our hospitals can handle all patients hopefully.
Cases come back at different rates and often in batches. People get tested at different points in their illness. With the delay some tests are seeing, people can be almost totally recovered before they actually know for sure they had it and their positive gets added to the daily report. Obviously the positive test is reported the day it comes back from the lab(otherwise you'd be updating in a backlog how the graph I link everyday does -- but we don't do that).
Essentially, you've got about a two week window either way on time of infection vs confirmed positive test, compared to a 2-3 day window based on time of onset of symptoms, making it the more uniform method.
The hospitalization number is the most important one though. I expect to see it go up notably with it looking like it's in a couple of nursing homes right now. Although that doesn't make the situation overall dramatically worse...
Essentially, you've got about a two week window either way on time of infection vs confirmed positive test, compared to a 2-3 day window based on time of onset of symptoms, making it the more uniform method.
The hospitalization number is the most important one though. I expect to see it go up notably with it looking like it's in a couple of nursing homes right now. Although that doesn't make the situation overall dramatically worse...
Wonder if the MSDH has given any guidelines/recommendations for hospitalization or is it strictly a case by case basis ?
No "growth rate" in reported numbers means anything. People are infected all over. Testing is not pervasive. Reporting is not standard and persistent. You will see local mins/maxs as the graph climbs and none of it should add any fear to your plate. Just do your part to stay well, stay put and help not carry the virus around.
I don't think we can tell anything. We are just sporadically adding to known cases in a non-consistent manner. There is no data that could come out in the next few days that would cause me to believe that we are/are not making marked progress. Too many variables. I do believe that the measures in place will have an impact, I just don't think we will have the right data to know it for quite some time. I do think that in all the data available, hospital occupancy is the most tangible way to know how we are doing. I still don't see:
I'm probably missing a number of variables that would be needed to have any chance at doing something meaningful statistically. Until testing becomes as pervasive as the flu test (i.e. if I walk in anywhere I can be tested and get results that are reported), numbers will mean very little.
- Total number of tests requested
- Total number of tests administered
- Total number of people refused testing w/o symptoms
- Total number of people refused testing w/ symptoms
- Average delay from test to results
- Total number of results pending
- Uniform reporting requirements for states
I'm probably missing a number of variables that would be needed to have any chance at doing something meaningful statistically. Until testing becomes as pervasive as the flu test (i.e. if I walk in anywhere I can be tested and get results that are reported), numbers will mean very little.
I would love for them to begin back filling data to put it in on the date of test administered. That will tell a better story than reporting on the day of results.
You kinda lost me here. If it were currently growing exponentially here, the number of people hospitalized from the disease(you can bet every single one of them with any of the symptoms in the hospital are tested and added to these numbers) would also be growing exponentially. That's ultimately the only truly important number here IMO...
I'm certainly not saying that can't and won't happen. Just that it hasn't really seemed to be the case so far. I agree with Sesh though that in one more week, we will really know where we are. It's possible these blips of optimism are for naught. We will see.
To restate what codedawg and Engie said in a different way, looking at cases by date of onset of symptoms removes some of the bias that comes from testing ramping up and results coming back faster.
If we tracked number of cases by date of test taken or date of test result, it would be hard to tell anything from it because we've ramped up testing, so that would drive an increase, and we've also ramped up testing faster than our ability to process tests, so there will be another increase as our ability to process tests ramps up. REporting it based on date of symptom onset removes all of the latter increase and mitigates some of the former.
The downside, as you said, is that there is a lot of variability around when symptoms start, but that's going to exist and show up in the numbers regardless b/c we are primarily only testing symptomatic people.
This CDC graph is interesting. Shows a general trend downward for tests administered from a peak on 3/17. Area in gray is not good data yet due to delays in reporting but trendline is for sure downward.
View attachment 15732
View attachment 15732
Interesting but strange...Per Covidtracking, we're processing between 95 and 120k tests per day now. I know we're producing far more tests than that now, so my supposition is that we don't have the ability to run more tests than that as of yet.
New tests coming now will basically be a 15 minute in-office flu test type of thing to the best of my understanding....
Somewhat surprised there are just 22 deaths with none in Hinds or Madison Counties. Since the denominator is still too high due to low test counts that mortality rate is probably much lower than 2%.
All I did was a simple graph of the confirmed cases each day and let excel calculate an exponential trend line. Rsquare was over 95%. But so was a polynomial trend line and it trended up a lot slower.
The exponential trend showed 1,085 today and we are actually at 1,073. It gets over 2,000 on Saturday.
Yeah it’s probably the most simple math you could apply, but I think it (along with a lot of stuff from people way smarter than me) shows that we are at a really critical point. Have we done enough or not? We are about to find out really, really soon.
The exponential trend showed 1,085 today and we are actually at 1,073. It gets over 2,000 on Saturday.
Yeah it’s probably the most simple math you could apply, but I think it (along with a lot of stuff from people way smarter than me) shows that we are at a really critical point. Have we done enough or not? We are about to find out really, really soon.
I stated similar elsewhere. The only number that is a reliable indicator right now it hospitalizations. At least, I think that those are consistently reported and countable.
And just like that got word from a friend that he and his wife are recovering from COVID. He was not allowed to be tested because he did not have a fever but did have other symptoms. Wife did get tested because she had enough symptoms. This is one reason why reported numbers are not reliable. My friend will never show up in the data.
Latest
-
01
Maroon and White Football
"Nothing changes": Mississippi State's LJ Prudhomme and Marcus Williams discuss expectations with bruised team
-
02
Maroon and White Football
Shuffled Mississippi State offense puts up yards, but no points in first shutout since 2020
-
03
Maroon and White
Everything Mississippi State HC Jeff Lebby said after LSU loss
-
04
Maroon and White
Rapid Recap: No. 22 LSU 31-0 No. 16 Mississippi State
-
05
Maroon and White
Inside the numbers: Mississippi State-LSU first half stats