Read on Twitter
Memphis/Shelby County
Mid-Wk Covid Report (Holiday Edition)
tl;dr
* cases are flat, but only bc testing is flat
* that means we're missing a lot of cases
* 124 dead over 3wks (previous highest 3wks, 94)
* but we can save 1750 lives in the next 60 days
* do your part
1/
Mid-Wk Covid Report (Holiday Edition)
tl;dr
* cases are flat, but only bc testing is flat
* that means we're missing a lot of cases
* 124 dead over 3wks (previous highest 3wks, 94)
* but we can save 1750 lives in the next 60 days
* do your part
1/
Deaths.
826 ppl in our community have died from covid this yr.
That's 2.5 more covid deaths than homicide deaths in 2020.
Be skeptical of those who want to hype homicides to hire more police & equip them with military equipment...but then oppose any effort to fight covid.
2/
826 ppl in our community have died from covid this yr.
That's 2.5 more covid deaths than homicide deaths in 2020.
Be skeptical of those who want to hype homicides to hire more police & equip them with military equipment...but then oppose any effort to fight covid.
2/
Right now, we're averaging 10 deaths each day.
To put that in perspective, a person in Shelby County is dying from covid every 2hr 24min.
Let that sink in. As you watch that Christmas movie, a neighbor will die from covid.
3/
To put that in perspective, a person in Shelby County is dying from covid every 2hr 24min.
Let that sink in. As you watch that Christmas movie, a neighbor will die from covid.
3/
More Memphians have died from covid in the past 75 days than from homicides the entire year (357 days).
Every day brings a new record.
The news hypes the record homicide numbers, but I can't find a headline about the daily record covid deaths.
4/
Every day brings a new record.
The news hypes the record homicide numbers, but I can't find a headline about the daily record covid deaths.
4/
Don't get me wrong, it's a real problem that we've had over 300 homicides this year.
I'm just calling attention to the way homicides are sensationalized & used politically, especially compared to the many more covid deaths.
5/
I'm just calling attention to the way homicides are sensationalized & used politically, especially compared to the many more covid deaths.
5/
Before I get off on a tangent, here are total cases, graphed on a linear scale.
Notice how the shape of the curve changes, in Aug, then again in Oct.
And notice how steep it is now in Dec.
6/
Notice how the shape of the curve changes, in Aug, then again in Oct.
And notice how steep it is now in Dec.
6/
And here are total cases, graphed on a log scale. (Remember, this is the "flatten the curve" scale.)
Look at how we successfully flattened the curve from Aug through Sept.
But then look at how we un-flattened it, starting in Oct. And look how much more un-flat it is now.
7/
Look at how we successfully flattened the curve from Aug through Sept.
But then look at how we un-flattened it, starting in Oct. And look how much more un-flat it is now.
7/
Here is our case rate.
Remember, Harvard's Global Health Institute says that stay-at-home orders are necessary at 25 daily cases per 100k ppl.
We've been over that for 45 days now.
We're at 79 per 100k.
That's actually down from 90 per 100k on Sunday.
8/
Remember, Harvard's Global Health Institute says that stay-at-home orders are necessary at 25 daily cases per 100k ppl.
We've been over that for 45 days now.
We're at 79 per 100k.
That's actually down from 90 per 100k on Sunday.
8/
The drop in cases here is artificial.
We saw a dip the week of Thanksgiving too, as testing slowed during the holidays...and then cases exploded afterwards as ppl sought out testing again.
We'll see another spike over the next several wks.
9/
We saw a dip the week of Thanksgiving too, as testing slowed during the holidays...and then cases exploded afterwards as ppl sought out testing again.
We'll see another spike over the next several wks.
9/
Here you can see how testing really ramped up over the past 2wks...and then dropped over the past few days.
When testing drops, we find fewer cases.
And when we test less, ppl don't know their status.
And when ppl don't know their status, they spread covid.
10/
When testing drops, we find fewer cases.
And when we test less, ppl don't know their status.
And when ppl don't know their status, they spread covid.
10/
Here, you can see that cases are growing faster than testing is.
But, also, look at how closely the case growth follows the test growth since late Oct.
The more we test, the more cases we find. (And vice versa.) That's how you know we're not testing enough.
11/
But, also, look at how closely the case growth follows the test growth since late Oct.
The more we test, the more cases we find. (And vice versa.) That's how you know we're not testing enough.
11/
The positivity rate has been above 10% for over a month now. And we know that, at 10%, you are missing a significant number of cases.
This is a real problem bc missed cases means ppl are walking around, infectious, spreading the virus.
12/
This is a real problem bc missed cases means ppl are walking around, infectious, spreading the virus.
12/
The most accurate covid forecast model was built by @youyanggu. He recently launched a "now-cast" model that estimates that there are 3.2 times more cases in TN than we know.
So the 741 avg daily cases could very well actually be more than 2,000 cases per day.
13/
So the 741 avg daily cases could very well actually be more than 2,000 cases per day.
13/
Cases are currently growing at a 1.4% daily rate. But that's only bc testing is down.
At 1.5% growth over the next wk, we'll avg 1,000 cases per day by New Year's Eve.
14/
At 1.5% growth over the next wk, we'll avg 1,000 cases per day by New Year's Eve.
14/
The first 10k cases took over 100 days.
From 10k to 20k, 28 days
30k, 52 days
40k, 52 days
50k, 25 days
We hit 60k on Sat
It took 14 days
70k will come in 11 days
80k, 9 days later
90k, 8 days after that
100k, 7 days after that, on Jan24
1 month from tomorrow
Exponential.
15/
From 10k to 20k, 28 days
30k, 52 days
40k, 52 days
50k, 25 days
We hit 60k on Sat
It took 14 days
70k will come in 11 days
80k, 9 days later
90k, 8 days after that
100k, 7 days after that, on Jan24
1 month from tomorrow
Exponential.
15/
In yesterday's briefing, they said we have a transmission rate of 1.03 (Covid Act Now has us at 1.11).
Dr. Haushalter said that would lead to 16,936 new cases in the next 30 days.
If nothing changes, though, I think it will be more than double that.
16/
Dr. Haushalter said that would lead to 16,936 new cases in the next 30 days.
If nothing changes, though, I think it will be more than double that.
16/
If daily growth continues at 1.5% for the next month, we're looking at 36,824 new cases.
If we cut the growth rate to 1.25%, then we're still looking at 29,493 new cases over the next month.
In other words, if nothing changes, it will be way worse than they say.
17/
If we cut the growth rate to 1.25%, then we're still looking at 29,493 new cases over the next month.
In other words, if nothing changes, it will be way worse than they say.
17/
The health dept invited Dr. Manoj Jain to give projected deaths. He gave worst-case, same-case, & best-case scenarios.
It was really helpful.
He said the difference between the best & worst case scenarios is 547 deaths.
In other words, we could save 500+ ppl over 2mo.
18/
It was really helpful.
He said the difference between the best & worst case scenarios is 547 deaths.
In other words, we could save 500+ ppl over 2mo.
18/
I used Dr. Jain's approach (1.35% case fatality rate) & applied my projected cases. And this scenario turns out worse than his worst-case scenario.
Mostly bc his model has a peak around Feb1, then a sharp decline, which would be great, but I don't see the evidence for that.
19/
Mostly bc his model has a peak around Feb1, then a sharp decline, which would be great, but I don't see the evidence for that.
19/
But my main disagreement is Dr. Jain's assumption that the fatality rate will stay at 1.35%.
He's the expert, not me, but I think a better assumption is that the fatality rate increases exponentially once hospitals are overwhelmed.
20/
He's the expert, not me, but I think a better assumption is that the fatality rate increases exponentially once hospitals are overwhelmed.
20/
If anyone disagrees with my assumption, I'd love to hear why.
It seems far-fetched to assume we can keep the fatality rate at the same levels once hospitals are overwhelmed.
It seems clear to me that ppl who would live today will die a month from now, if nothing changes.
21/
It seems far-fetched to assume we can keep the fatality rate at the same levels once hospitals are overwhelmed.
It seems clear to me that ppl who would live today will die a month from now, if nothing changes.
21/
But if my assumption is correct, we're looking at 1,200 more deaths than Dr. Jain's worst-case scenario.
So we're actually talking about saving more than 1,750 lives.
This is where we need pro-life Republicans to step up. Don't protest the health dept, protest the deaths.
22/
So we're actually talking about saving more than 1,750 lives.
This is where we need pro-life Republicans to step up. Don't protest the health dept, protest the deaths.
22/
One last thing.
Given Gov Lee's refusal to issue a statewide mask mandate, I feel like this is an important thread.
Don't misunderstand. Masks work. We need masks. But they aren't a magic bullet.
We need masks, but we need more than just masks.
https://twitter.com/youyanggu/status/1339306972189843456
23/
Given Gov Lee's refusal to issue a statewide mask mandate, I feel like this is an important thread.
Don't misunderstand. Masks work. We need masks. But they aren't a magic bullet.
We need masks, but we need more than just masks.
https://twitter.com/youyanggu/status/1339306972189843456
23/
We have had a mask mandate here in Shelby County for months, but we still have one of the worst pandemics on the planet.
Sure, we're doing better than most of the state, but that's only bc our state is literally the worst in the world.
So, yes, masks.
But what else?
24/
Sure, we're doing better than most of the state, but that's only bc our state is literally the worst in the world.
So, yes, masks.
But what else?
24/
The only solution I see is to pay local govt, small biz, and individuals so that they can make safe decisions.
We can see the finish line.
Now is the time to hunker down for the Winter so that we can all make it to Spring.
We can save 1,750 lives in the next 60 days.
25/25
We can see the finish line.
Now is the time to hunker down for the Winter so that we can all make it to Spring.
We can save 1,750 lives in the next 60 days.
25/25
