OT: Current Covid trends in Big Ten states
With the decision looming on whether or not to start fall football camps in the Big Ten, I decided to take a look at the data for current trends in Big Ten states. Presumably, shrinking numbers should portend better chances of football moving forward.
Results are below, but first an overview of the data points (all data from covidactnow.org):
Daily new cases per 100k population: ranges are low (1 and below), medium (1-10), high (10-25), and critical (>25).
Infection rate: this measures the degree to which the virus is spreading or shrinking. A value of 1.0 means each infected person infects 1 other person. Anything above 1 means the virus is expanding; less than 1 means it is shrinking. They again group the numbers as low (< 0.9), medium (0.9-1.1), high (1.1-1.4), and critical (>1.4).
Positive test rate: this measures the percentage of tests returning positive results. Note that this does not reflect changes in the volume of testing. Ranges for this are low (< 3%), medium (3-10%), high (10-20%), and critical (> 20%).
Note: Covidactnow also measures ICU capacity and contact tracing, but I don't expect those numbers to impact the decision for the season, so I have not included those. In case you are curious, every Big Ten state has plenty of ICU capacity, and doesn't do a good job of tracing contacts.
Here are the current numbers for Big Ten states:
- Illinois
- Daily new cases: 11.1 (low end of the high range)
- Infection rate: 1.11 (low end of the high range)
- Positive test rate: 3.8% (low end of the medium range)
- Trend: all 3 of these factors seem to be at a peak that is starting to plateau and possibly trend downward.
- Indiana
- Daily new cases: 12.3 (low end of the high range)
- Infection rate: 1.10 (low end of the high range)
- Positive test rate: 8.1% (medium range)
- Trend: similar to Illinois, these factors have been on an upward trajectory that appears to be peaking and starting to come down.
- Iowa
- Daily new cases: 15.5 (high range)
- Infection rate: 0.97 (medium range)
- Positive test rate: 9.6% (upper end of the medium range)
- Trend: daily new cases and infection rate are trending down after reaching a recent peak. Positive test rate appears to be holding steady.
- Maryland
- Daily new cases: 14.4 (high range)
- Infection rate: 1.15 (low end of the high range)
- Positive test rate: 6.4% (medium range)
- Trend: daily new cases are on a sustained upward trajectory since the beginning of July. The infection rate appears to have reached a peak from a steady increase, and appears to be starting a downward trajectory. Positive test rate has been holding steady.
- Michigan
- Daily new cases: 7.3 (medium range)
- Infection rate: 1.06 (medium range)
- Positive test rate: 2.2% (low range)
- Trend: Daily new cases seems to be plateauing after a steady, though slow, rise. Infection rate appears similar to others: reached a peak after a steady rise, and is now on a downward trajectory. Positive test rate has been steady since the beginning of June.
- Minnesota
- Daily new cases: 12.6 (high range)
- Infection rate: 1.07 (medium range)
- Positive test rate: 4.8% (medium range)
- Trend: Minnesota's trends appear very similar to Maryland's. Daily new cases have been increasing and appear to still be on an upward trajectory. Infection rate has peaked and is starting to come down. Positive test rate has increased slightly, but has largely been steady.
- Nebraska
- Daily new cases: 15.2 (high range)
- Infection rate: 1.15 (low end of the high range)
- Positive test rate: 9.2 (upper end of the medium range)
- Trend: Daily new cases have been increasing, though may be starting to plateau. Unlike the other states so far, the infection rate is still on an upward trajectory, as is the positive test rate.
- New Jersey
- Daily new cases: 4.9 (medium range)
- Infection rate: 0.99 (medium range)
- Positive test rate: 1.3% (low range)
- Trend: daily new cases and infection rate have been mostly holding steady for the month of July after coming way down; currently in the midst of a slight rise. Positive test rate has been holding steady.
- Ohio
- Daily new cases: 11.2 (high range)
- Infection rate: 1.01 (medium range)
- Positive test rate: 4.6% (medium range)
- Trend: daily new cases appear to have hit a plateau after a steady rise. Infection rate has been slowly and steadily decreasing. Positive test rate holding steady, maybe starting to decrease.
- Pennsylvania
- Daily new cases: 7.5 (medium range)
- Infection rate: 1.08 (medium range)
- Positive test rate: 6.1% (medium range)
- Trend: daily new cases and positive test rate have been slowly and steadily increasing over the month of July, and do not appear to show signs of plateauing. The infection rate has been remarkably steady, if on a slight downward trend.
- Wisconsin
- Daily new cases: 15.6 (high range)
- Infection rate: 1.05 (medium range)
- Positive test rate: 6.9% (medium range)
- Trend: daily new cases and positive test rate have been on a steady increase since June, though are possibly showing signs of plateauing and starting to adjust downward. Infection rate has been on a slow and steady decline over the month of July.
Summary
Of the 11 states comprising the Big Ten:
- Daily New Cases
- 8 are in the High range, 3 are in the Medium range.
- Of the 8 states in the High range, 4 are not far (within 2 points) from moving into the Medium range; none are close to moving into the Critical range.
- Infection Rate
- 4 are in the High Range, 7 are in Medium.
- Of the 4 in the High Range, 3 are close to moving into the Medium range and are on a downward trajectory. The one still increasing (Nebraska) is closer to the Medium range than Critical.
- Of the 7 in the Medium range, all are on a downward trajectory (with the one possible exception being New Jersey).
- Positive test rate
- 9 are in the Medium range, 2 are in the Low range.
- Of the states in the Medium range:
- 5 are either holding steady or decreasing.
- 4 are either increasing or showing potential signs of increasing.
- None are close to breaking into the critical range, while 3 are close (and on trajectory) to drop into the Low range.
- The 2 states in the Low range have been holding steady for a sustained amount of time, and are not showing signs of moving into the Medium range.
- Overall
- I will admit that this painted a bleaker picture of the current state of things than I expected.
- That said, there does seem to be an overall trend of having reached a recent peak, and things are starting to come down. Nebraska is the only state that stands as an exception to this.
- The Big Ten will have a tough decision to make in the next 5 days. The current trends support moving forward with the season, but you would hate to bring more people into close environments that could cause the current trends to reverse back upwards.
Hopefully this at least provides a contextual and data-driven view of the Big Ten landscape.
And, to think of a possibility that we're still in the first wave, with a second, new wave forecast for sometime this fall. Considering that there are few students on campus now, a fall campus wave seems imaginable.
I thought this was the second wave and a new wave would be coming on the tail end of Fall, late fall early winter when Flu season hits?? Honest question.
What happened was we pushed the first wave back in the states that never really had a first wave. Basically held the front end of the the bell curve down for awhile, but it seems like the virus is gonna do what the virus is gonna do and those states are in their first waves. There is some evidence that it's mutated to spread more easily as well, fun.
Mutated to spread more easily? Have a citation? (Serious question ...)
I've read that there are two variants and that most of the USA got the one from Europe as opposed to the one from China.
I has google: https://www.livescience.com/new-coronavirus-mutation-explained.html
The second "European" strain that you mention is the one that has become dominant in most parts of the world despite obviously not being the first on the scene. This has led some to believe it's more infectious. And when comparing pseudoviruses using the two spike variants, researchers found that the new strain was far more effective at entering cells. It's not conclusive but an interesting result. It might not necessarily mean the actual virus is more infectious.
And it's possible that the reason this mutation has become dominant is simply that the original strain was better contained in China whereas the mutation that originated in Europe was allowed to spread more easily.
It's hotly debated whether this particular mutation actually is more infectious but viruses do tend to mutate in a way that they're more infectious and less lethal since those mutations will have more success spreading. More than plausible.
We’re seeing it in the south. My Hypothesis for this is because it’s hot and folks are seeking the indoors wherever possible, thus breathing lots of recirculating air after being in congregation with other people
this is why I believe the second wave In the north, should there be one, would be mid fall when everyone spends more time indoors needing heat. You’ll see the opposite trend in the south. Less congregating indoors.
the trends generally correlate, though not perfectly, to seasonal changes where people spend more or less time outdoors.
Amen! Thanks for an intelligent comment. Folks do not spend as much time outdoors in the Deep South as they do in the spring due to hotter temps and higher humidity. Northern states do not have the persistent heat and humidity and the early to mid fall climate is delightful so any real second wave would come in November.
I'ts not all about the weather. Folks need to take their own precautions as well.
I know what you meant to say, so I'm being a bit of a dick here, but what you typed reminded me of one of my favorite Stooges lines. You wrote: "Folks do not spend as much time outdoors in the Deep South as they do in the spring..."
Curly, while pretending to shave a customer, said "Tell me, is it warmer in the city than it is in the summer, or vice versa?"
COVID sucks, but the Stooges last forever.
No kidding. I have to shower twice a day from sweating so much. A stark contrast to never having to wash during the rest of the year.
Look out how MLB is doing. More games postponed today. Football doesn't stand much of a chance once all the teams start full contact.
Yeah, if the MLB is struggling this bad, I can't imagine college football is going to happen (or at least survive more than a few weeks). These are amateur athletes, on rosters of 100+ guys, and they aren't pros who can be confined to hotels. They're going to be out with friends, meeting ladies, etc. The season might start, but I don't think it'll last long unless peoples position is just 'eff it. We need football'
Thanks for the data!
Since March, I've been following the data rather than what our national leaders are telling us...and it is bleaker than what we are being told.
I'm interested, though, that your conclusion was that the trends support moving forward with the season...two bullet points after having indicated that the data was bleaker than you expected.
My concern is that when school starts - and many people head back to work - that the already tenuous situation is going to get much worse...with so many people being in closer contact with one another.
My sophomore U of M son is moving back up in four weeks. My biggest concern is what happens when we move him back home just before Thanksgiving. I'm operating under the assumption that he's going to catch the virus. I don't see how a large percentage of these kids don't get it...being in such close promiximity to one another in the dorms, in their apartments, at restaurants, etc.
I'm interested, though, that your conclusion was that the trends support moving forward with the season...two bullet points after having indicated that the data was bleaker than you expected.
Maybe that's just my desire for a season coming through. But my point was that, even though the current state is bleaker than I expected, the fact of an overall downward trend makes the idea of a fall season more tenable. If current trends hold, then within another week we would see more improvement.
I disagree with your assessment that our numbers, in Michigan at least, are looking all that great. We're clocking about 5x the number of cases each day than we were at the bottom of the curve. Bars are being closed statewide again, which suggests there's an impending problem up north.
Also, the recent data showing significant lung damage, even in people who had mild symptoms (or were asymptomatic entirely) is very concerning, especially for people whose careers depend on their lung capacity.
Also, the recent data showing significant lung damage, even in people who had mild symptoms (or were asymptomatic entirely)
Can you point me to the source(s) you read about this? I am genuinely asking the question, no snark. I think many people (myself included) are curious about the possible long-term ramifications of Covid.
I agree, Hatter.
Our cases have been ticking up quite a bit over the last month.
And the issue with lung damage is what is being overlooked by what I refer to as the "Covid-19 Optimists." Yes, the death rate is not that high (although, I think even one unnecessary death is one death too many), but the long-term health implications of getting the virus are still very much an unknown.
This is what is making me very nervous about sending my son back to U of M. Is he likely to get the virus? I think the answer is yes. Is he likely to die from it? No...but I'm concerned that in ten or twenty years there is going to be a large percentage of people having serious lung issues.
I really think a decision should have been made sooner, but whatever they chose i hope it works out.
thanks for this info and analysis Moleskyn.
Tangentially related, but I was playing with some data last night to try to estimate death rates by state, and in a nutshell, at the US level about 1.7% of positive tests lead to death. Had to make some assumptions about how long it takes between the two. I settled on 2 weeks vs 3 to be conservative given first few days may be as asymptomatic and a few days to get a test result back. By no means is this a mortality rate for the disease given not all those who contract get tested. It also doesn’t distinguish between ages, comorbids, etc. but still a pretty high rate.
Good data. I don't agree with highlighting the case rate. It's a data point but not there is too much focus on it. Positive test rate is the best indicator of virus spread in my opinion. Fatality rate is also very telling and generally flat or positive in the midwest.
I get your point about the case rate, but I think it's beneficial alongside the others to get an idea of the overall scale.
I actually disagree with your point about the positive test rate being the best indicator. I think the positive test rate is hard to contextualize without knowing the number of tests being done. If fewer tests are being done, then that likely means they are only testing people who are symptomatic or who are more likely to have the virus...leading to a higher positive rate. More testing means you are more likely to be testing non-infected people, which would lower the positive rate.
I am no expert in this, but I think you can infer the amount of testing being done by looking at the daily new cases alongside the positive test rate. For example, if the positive test rate is holding steady, but new cases are increasing...then that must mean more testing is being done, right?
But in my opinion, the daily new cases and positive test rate are only telling the story of what is already known. I think the infection rate is a little more forward looking - if the average infected person is infecting fewer people, then you will see the daily cases and positive rates go down.
Good stuff. I work with state health departments around the country and they are still way behind the curve in terms of technology and analytics. I think this generates a lot of poor policy decisions based on biases. That's why you see remarkably different approaches depending on politics of leadership and populations. This leads to a lot of frustrated citizens who don't trust the leaders to begin with and their trust further erodes with seemingly random decisions
I'm don't think the B1G can proceed with a football season when entire teams are in quarantine. (I'm wearing my Captain Obvious hat as I type this)
I'm not making a statement with this, I'm simply trying to understand. Why are they not reporting things like hospitalization rates? I spoke with a nurse last week who works at the UM ICU and he said they had (at the time) 11 patients. It just seems like critical cases is a far more important metric than infection rate.
This is a good point. And this is where I'm trying to think through the lens of the people making the decision to go forward with the season or not.
If I am looking to make that kind of decision, I care more about how prevalent the virus is, and how much it is spreading. It doesn't matter if hospitals have the capacity to take on more patients; I don't want the bad PR of athletes getting sick and needing hospitalization.
From my perspective as an ordinary citizen, the hospital/ICU capacity is extremely important. People are always going to get sick, and there will always be a need for ICU care. So as long as we are not overloading the healthcare system, I am not too concerned. And that's where, looking at the ICU capacity for the Big Ten states, it is interesting that all of them have plenty of capacity at the moment. Nobody is even in yellow territory; they are all green for ICU capacity.
Depends on what your view is and the risks you are trying to manage. If you are simply trying to avoid lawsuits, that's one view. If you truly care about health and safety of players and staff, that's another angle. There is the financial aspect as well. And of course fan base and donor input. So many variables to manage.
The simple answer is that hospitalization rates are a lagging indicator - once you have an issue with hospitals, you're already behind the eight ball.
Two recent examples are Texas and Florida. In early June, cases started rising for both states. People trying to wish the problem away pointed out that the hospitalization rate and number of deaths were not rising with the cases.
However, cases continued to rise all through June and into July, and as expected, the hospitalization rates and deaths followed. Both states, especially in the hot spot areas, are now seeing hospitals nearing or at capacity, and daily deaths continue to climb.
So if all anyone looked at was hospitalization rates, no one would know there was a problem until it was too late.
Cases in Florida are decreasing, as are hospitalizations.
https://www.worldometers.info/coronavirus/usa/florida/
But deaths are rising.
Florida cases were increasing "into July", as I said. And they are now decreasing, as you pointed out. I'm not sure why you are bringing this up, though? I brought up a specific example of a rapid rise in cases over a specific time frame, to use as a method to compare the case trends to hospitalization & death trends over that same time period. Are you challenging the point that hospitalization trends lag case trends?
From my post: "However, cases continued to rise all through June and into July, and as expected, the hospitalization rates and deaths followed."
And your own charts prove that point. The highest case date looks to be about 7-12, and the highest hospitalization dates are 7-21 and 7-22. By the time the hospitalizations peaked, the cases were already decreasing - indicating that Florida has already made some headway, and that hospitalization trends lag case trends.
And your charts also prove that the cases drop quicker than hospitalizations when the epidemic is waning - i.e. they lag cases both on the way up and the way down. Reinforcing the point that tracking cases is the best way to identify the most recent trends.
Given your post history, it sure seems like you're trying to say I was lying or making something up, but in reality you have proven that was not the case. Thanks for posting the charts!
EDIT: and to my point about hot spots having hospital capacity issues - this is from 3 days ago. https://www.sun-sentinel.com/coronavirus/fl-ne-broward-hospitals-getting-overflow-20200728-akz7k5wmubb2billpnofsqtqdy-story.html
I didn't accuse you of lying. You are most welcome for the charts. Here is another from IHME on ICU capacity and use for Florida.
Fair enough - I apologize for assuming, that was wrong and petty of me.
Well thank you. I understand what you were saying and you have reasoned view above. My intent was to add context of where Florida is today, and I should have been more clear on that. I frequently am an ass on this site, but when you are a mule it is difficult not to bray.
Why are they not reporting things like hospitalization rates?
FWIW it's a given that hospitalizations and deaths are lagging indicators; a bump in the infection rate will inevitably result in a bump in both hospitalizations and deaths 2-3 weeks down the line.
For example, Maryland's hospitalizations hit a low of 390 something in June. As cases have spiked here, the hospitalizations have gone back up to a recent high of 590ish as of this morning.
This is true, but important context to add is that Maryland is nowhere near capacity in their hospitals, and at this moment does not in any way seem in danger of overloading hospitals. IHME data
Yeah. Our outbreak high of hospitalizations was like 1200-1300 back in April IIRC. At that point, there were concerns that we would have to utilize the temporary ICU beds set up in the Baltimore Convention Center.
It speaks to the changing demographic of patients we are seeing in Maryland; the majority are 39 and under, and thus, less likely to require ICU stays/hospitalization.
Exactly. I'm so tired of the hype about increasing cases, hospitalizations, etc when they are still small relative to overall capacity or populations. That's why this issue is so polarizing. People use the pandemic as a hammer to push their agenda.
"People use the pandemic as a hammer to push their agenda."
Which is ... what, exactly? Just saying LIBRULL won't do the job. Details, please ...
To the larger point, it doesn't seem that the healthcare systems in many areas are close to being overwhelmed. As well, getting a small to moderate dose of the virus won't be a big deal for most people.
But think of the childish obstinance and stupidity that have been on display in the Sun Belt. We should be thankful that at least some people got on board with reasonable public health ideas. Had that not happened, some systems could have been overwhelmed.
August 1st, 2020 at 11:07 AM ^
The “long term damage” stuff makes really all of the other rates irrelevant in terms of when and how we can move forward. That emerging narrative will really box us in in terms of what we can and cannot proceed with. As long as “the long term effects are still unknown” talking point gains a lot of traction I really don’t think anybody will be able to do anything in the fall/winter.
I skipped to the "overall" section...
Can you say where you're getting trend data from?
I have been tracking Ohio closely and the 7-day moving average of the rate of change in new cases has been around or above 5% since mid-June. Sure that rate was 15% on June 22 and yesterday it was 4%, but simply having a positive rate of change is bad. So not only are the cases not going down, but the number is generally going up by 5% per day. Case in point, just yesterday, Ohio set a new record for new cases.
To simplify, here is a chart of new cases in Ohio since June 6. I don't see a plateau.
I based my trends on the data at Covidactnow.
Here are the trends for Ohio for the 3 factors I mentioned:
So based on these trend lines, I said
daily new cases appear to have hit a plateau after a steady rise. Infection rate has been slowly and steadily decreasing. Positive test rate holding steady, maybe starting to decrease.
The goal should be a plateau at a low level rate of new cases, not at the historically highest level of daily new cases
Correct. But a plateau after a long and steady rise means at least the rate of increase has reduced. And if the rate of increase continues to go down, then eventually the daily new cases will go down as well.
In other words, while it is plateauing higher than we would like to see, the trend (and other data points) points to that starting to track downwards.
I don’t know how they count cases in MI but in NC if you go in and are positive you asked to come back in a few days. When you return if it’s still positive you are asked to come back in a week, if positive, the result comes out as 3 positive cases and is not separated into one person with the virus.
We have also had numerous people that filled out the paperwork and decided the wait was too long and a few days later received confirmation that they are positive and never took the test.
Certainly makes one wonder how this is being managed.
Yes that sounds right. I know of one person who filled out the paperwork and got sent home and told to come back another day because there were too many people to test that day. He never went back but got a letter stating his test was positive.
But if you bring this sort of thing up you're a science denier.
Wow, everyone has this same "person they know"
I've personally heard this story about 5 times but no one personally knows them when you pin them down.
I will let you know, our friends daughter in-law runs/manages testing sites around Charlotte and the person that received the results is my wife’s friend who works in Raleigh. We have heard of many others not only in NC but in SC. Believe it if you want to and if you don’t that’s fine also.
Yeah, I'm putting that into my Covid Apocrypha binder next to the friend of my wife's friend who died in a car wreck and cause of death was listed as Covid 19.
I don't post anything anonymously on the internet with the expectation of being seen as a credible source. So I'm not sure what you're trying to accomplish with this comment.