Coronavirus pandemic

[Eth]

According to testing data compiled by Johns Hopkins, there has been a 2.4X increase in positive tests from May 10th to July 4th (21,602 vs. 52,391).  But this 2.4X increase in positive tests can be attributed to a 2.4X increase in daily testing during the same time period (267,961 vs 644,415).  Both May 10th and July 4th data had the same percentage positivity rate of 8.1%.  More testing leads to more positive tests (shocking results!). 

Very impressive cherry-picking.

Let's try again with a start date of June 10th instead of May 10th. Now the positive tests go from 20,761 to 52,391 (+152%), while the total daily tests go from 429,207 to 644,415 (+50%). More testing leads to more positive tests...but not that many.

[SEWIGuy]

According to testing data compiled by Johns Hopkins, there has been a 2.4X increase in positive tests from May 10th to July 4th (21,602 vs. 52,391).  But this 2.4X increase in positive tests can be attributed to a 2.4X increase in daily testing during the same time period (267,961 vs 644,415).  Both May 10th and July 4th data had the same percentage positivity rate of 8.1%.  More testing leads to more positive tests (shocking results!). 

Very impressive cherry-picking.

Let's try again with a start date of June 10th instead of May 10th. Now the positive tests go from 20,761 to 52,391 (+152%), while the total daily tests go from 429,207 to 644,415 (+50%). More testing leads to more positive tests...but not that many.

Not only that, but if you are testing more, you want the rate to go DOWN.  Not stay the same.  It means that a lot more people have the disease and can spread it to others.

[tradephoric]

According to testing data compiled by Johns Hopkins, there has been a 2.4X increase in positive tests from May 10th to July 4th (21,602 vs. 52,391).  But this 2.4X increase in positive tests can be attributed to a 2.4X increase in daily testing during the same time period (267,961 vs 644,415).  Both May 10th and July 4th data had the same percentage positivity rate of 8.1%.  More testing leads to more positive tests (shocking results!). 

Very impressive cherry-picking.

Let's try again with a start date of June 10th instead of May 10th. Now the positive tests go from 20,761 to 52,391 (+152%), while the total daily tests go from 429,207 to 644,415 (+50%). More testing leads to more positive tests...but not that many.

OK, but then why not pick April 10th?  Now the positive tests go from 34,229 to 52,391 (+53%), while the total daily tests go from 157,405 to 644,415 (+309%).  Comparing April 10th to today would be a better date to use to prove that increased testing results in increased positive tests.  But the reason i chose to compare June 10th to July 4th was because both dates had the same percentage positive tests.  To put things in perspective Florida (which has a slightly higher population than New York) is reporting 10k daily cases, which is the same number of daily cases New York was reporting during their peak in April.  At their worse New York was experiencing 1,000 daily deaths whereas right now Florida is averaging 45 daily deaths.  But since daily cases in Florida are as high as what New York was seeing when 1,000 people a day were dying, do we really expect daily deaths in Florida to increase from 45 to 1,000 over the next few weeks?   Of course not.  Until Florida starts seeing anywhere close to 1,000 deaths a day, things are NOWHERE near as bad as they were in April.

[hotdogPi]

According to testing data compiled by Johns Hopkins, there has been a 2.4X increase in positive tests from May 10th to July 4th (21,602 vs. 52,391).  But this 2.4X increase in positive tests can be attributed to a 2.4X increase in daily testing during the same time period (267,961 vs 644,415).  Both May 10th and July 4th data had the same percentage positivity rate of 8.1%.  More testing leads to more positive tests (shocking results!). 

Very impressive cherry-picking.

Let's try again with a start date of June 10th instead of May 10th. Now the positive tests go from 20,761 to 52,391 (+152%), while the total daily tests go from 429,207 to 644,415 (+50%). More testing leads to more positive tests...but not that many.

OK, but then why not pick April 10th?  Now the positive tests go from 34,229 to 52,391 (+53%), while the total daily tests go from 157,405 to 644,415 (+309%).  Comparing April 10th to today would be a better date to use to prove that increased testing results in increased positive tests.  But the reason i chose to compare June 10th to July 4th was because both dates had the same percentage positive tests.  To put things in perspective Florida (which has a slightly higher population than New York) is reporting 10k daily cases, which is the same number of daily cases New York was reporting during their peak in April.  At their worse New York was experiencing 1,000 daily deaths whereas right now Florida is averaging 45 daily deaths.  But since daily cases in Florida are as high as what New York was seeing when 1,000 people a day were dying, do we really expect daily deaths in Florida to increase from 45 to 1,000 over the next few weeks?   Of course not.  Until Florida starts seeing anywhere close to 1,000 deaths a day, things are NOWHERE near as bad as they were in April.

April NYC is worse than current Florida. However, current Florida is worse than most of the country in April, now that it's in or near the lead in infection growth.

[Eth]

But the reason i chose to compare [May] 10th to July 4th was because both dates had the same percentage positive tests.

(fixed quote – June 10 was the starting date I chose, May 10 was yours)

Exactly my point. You knew the outcome you wanted to show, so you went back to find the exact starting date that would produce that outcome.

As you and I have both shown, you can use the same data to show whatever you want if you just pick the right parts of it to use.

[tradephoric]

But the reason i chose to compare [May] 10th to July 4th was because both dates had the same percentage positive tests.

(fixed quote – June 10 was the starting date I chose, May 10 was yours)

Exactly my point. You knew the outcome you wanted to show, so you went back to find the exact starting date that would produce that outcome.

As you and I have both shown, you can use the same data to show whatever you want if you just pick the right parts of it to use.

Thank you for fixing that quote!  My larger point has always been that just because we are seeing record highs in daily cases doesn't mean things are worse now than they were back in April.  The first graphic shown in nearly every news report is the daily new cases and how we are seeing record highs.  But it's irresponsible for the media to continually show that graphic without putting it into context for the American people.  Many people will believe the virus is more widespread today than it was back in April simply because there are more daily cases.  Without consistently reporting the effects increased testing has on daily case counts, the American people may think the situation is much worse than it really is.  Why not focus on the fact that daily deaths have declined by over 75% and that deaths continue to decline nationally even as daily cases have been on the rise for about the past month.

[MikeTheActuary]

Thank you for fixing that quote!  My larger point has always been that just because we are seeing record highs in daily cases doesn't mean things are worse now than they were back in April.  The first graphic shown in nearly every news report is the daily new cases and how we are seeing record highs.  But it's irresponsible for the media to continually show that graphic without putting it into context for the American people.

Sadly, most Americans (including too many news reporters) are functionally illiterate when it comes to statistics.  Thus the news reports get dumbed down.

The headlines should be:

• The places that used to be the worst in the US have (finally) gotten sufficiently better to begin working through a sustainable "temporary normal" until a vaccine becomes available.
  
• Unfortunately, places that were (compared to the worst places, at least) doing reasonably well are now seeing matters become much worse, with evidence of exponential growth patterns setting in.

It is probably fair to criticize headlines of "record cases" (nationally, or for individual states, or...) being distorted due to increased testing.   But it would also be fair to say that official case counts are metrics that are interesting but not particularly important.

Positivity rates, hospitalization rates, numbers of deaths are better metrics for looking at how the pandemic is going in the country or in a particular area.  They aren't perfect metrics -- positivity rates can be influenced by local testing criteria / local availability of test kits; some of the hospitalization metrics are confused by how different areas/different hospital systems are set up to account for surge capacity; and numbers of deaths might be being distorted by the changes in demographics of new cases and the evolution of treatment regimens -- and some of them are difficult for the public to understand.

But when you look at the available data, it becomes clear that within the past month/month-and-a-half, COVID has gotten out of control in some parts of the US.   That's what the real story ought to be.

[MikeTheActuary]

When I went out today (the 4th), I saw a lot of people ignoring social distancing in general, and ignoring masks at yard parties, so I wouldn't be surprised if CT loses ground soon.

You don't need masks at a yard party so long as you don't hang out closer than 6 feet from others.

I should have phrased my statement better.

I went out yesterday (the 4th).   I observed:

• Less adherence to social-distancing guidelines in general than I've observed over the past several weeks in CT.
  
• At yard parties, there was general disregard for both social distancing guidelines AND mask protocols.

It is true that official mask guidelines indicate that masks are not needed if social distancing is observed (especially if outside, and especially if close-proximity contact with others is brief/incidental).  But the parties I saw seemed pretty "pre-COVID normal", leading me to wonder how much longer CT will hold the title of "best in the country" when it comes to having COVID under control.

I'll also note that while official guidelines do say masks are not particularly useful when social distancing is observed...it should also be remembered that the six-foot/two-meter rule was rather arbitrarily set, based on a study of a cough.  When folks look at more robust studies of how long aerosolized exhaled particulate matter can stay in the air, and that the mode of transmission seems to hinge on prolonged exposure to viral particles, there's an argument to be made that a better distance rule-of-thumb would be 3-5 meters, but-for the problem of not wanting updated guidelines to be ignored due to a change in guidelines being perceived as damaging health officials' credibility.

(C.f. some of the arguments raised by mask-opponents.)

[wxfree]

I've seen people mock the six-foot rule guideline, as if they think we're being told that there's a shield that extends six feet, and that distance magically makes the difference between certain infection and certain non-infection.  One of the problems we have is how people have a hard time with the idea of how risk, and protection, is cumulative.  The difference between kissing someone and standing 6 feet away is probably more than the difference between being 6 feet away and being 12 feet away.  Guidelines have to be easy to understand and follow.  We can't say "spend no more than 5 seconds within a foot, no more than 20 seconds within two feet... and no more than 15 minutes within 6 feet" and expect anyone to remember it.  But if they want to dismiss the threat and mock those who take it seriously, they'll find some way to do it.  I especially enjoy reading about how a cloth mask won't filter anything, how viruses and droplets fly right through, except that it will block things that are almost infinitely small, such as oxygen and carbon dioxide molecules, and make you suffocate.

[Sctvhound]

The coast of Georgia and the northern half of Florida's Atlantic coast are getting more cases per capita than inland. I believe it was true for South Carolina a week ago, but I can only see current numbers.

Still the same. My county (Charleston) has more cases than any other county in the state, even the more populated Greenville County. Over 3,800 actives.

https://carolinasportsthoughts.wordpress.com/2020/07/05/active-coronavirus-cases-in-south-carolina-7-5/

[bandit957]

So is anyone else already making big plans for when the pandemic ends?

[vdeane]

There's also the fact that asking people to stand 9-15 feet apart would start to get impractical very fast.  Things like grocery shopping are already impractical if one is to actually strictly observe the 6' rule.  Expanding it would make such things impossible.  There's always a compromise between what is needed to reduce the spread of the virus and what is practical to actually do.  If there wasn't, we'd just throw everyone into solitary confinement for a few weeks and the disease would be eradicated.

[Max Rockatansky]

So is anyone else already making big plans for when the pandemic ends?

No, especially that can't be reasonably calculated.  Realistically this will probably be a thing with COVID probably at minimum for another 12-18 months.  The way I see it things like masks and restrictions going all over the place will be around for awhile.  Personally I guess that I can't accept that as "normal"  and do what I can to keep up with the things I feel are necessary to do. 

[Ben114]

So is anyone else already making big plans for when the pandemic ends?

I'm hoping to go visit colleges in spring / summer 2021.

(Luckily for me, I want to stay within the Northeast, and it's unlikely of any bad hotspot now that we have it under control.)

[hbelkins]

According to testing data compiled by Johns Hopkins, there has been a 2.4X increase in positive tests from May 10th to July 4th (21,602 vs. 52,391).  But this 2.4X increase in positive tests can be attributed to a 2.4X increase in daily testing during the same time period (267,961 vs 644,415).  Both May 10th and July 4th data had the same percentage positivity rate of 8.1%.  More testing leads to more positive tests (shocking results!). 

Very impressive cherry-picking.

Let's try again with a start date of June 10th instead of May 10th. Now the positive tests go from 20,761 to 52,391 (+152%), while the total daily tests go from 429,207 to 644,415 (+50%). More testing leads to more positive tests...but not that many.

Not only that, but if you are testing more, you want the rate to go DOWN.  Not stay the same.  It means that a lot more people have the disease and can spread it to others.

Not necessarily. You're probably finding a lot more asymptomatic people, and I'm pretty sure I read over the weekend yet another study that pooh-poohs the concept of asymptomatic transmission. What's the difference between being asymptomatic and not knowing you have it vs. being asymptomatic and testing positive? Nothing, really. Higher numbers of asymptomatic tests doesn't really mean anything. The important numbers are hospitalizations and deaths. I think the statistics are still that 40 percent of positive tests are asymptomatic, 20 percent exhibit only very mild symptoms, and only about 10 percent requiring hospitalization, especially among the elderly and those with significant comorbidities. In Kentucky, fully 60 percent of the deaths have come from those who were in nursing homes. And isn't the national fatality rate 4 out of every 1,000? That's a 99.6 percent survival rate.

[wxfree]

Not everyone who survives will recover.  But maybe the tragedy here isn't how many people die or suffer organ damage or extreme pain, maybe the tragedy here is how many people simply don't care because to do so would be inconvenient.

[renegade]

So is anyone else already making big plans for when the pandemic ends?

Yeah, me:  a chili run on my way down to the Bourbon Trail.   

[MikeTheActuary]

So is anyone else already making big plans for when the pandemic ends?

Nope.  I've learned better. 

I'm probably to blame for the pandemic.  I was SO proud of myself, getting nice and organized, and getting travel plans in place for most of the year very early on.  Given how I shuttle among CT, QC, and TN, I've found that if I lock in "defensible" travel dates, I can usually get work and my father's doctors to accommodate my already-made travel plans....and for 2020 I had those plans made in such a way that I wouldn't wear myself out with back-to-back  (or back-to-back-to-back) trips.

And then the pandemic went and made all those plans moot.  I don't know when the US/Canadian border will reopen, so it's pointless for me to plan to go to QC, and I don't know when my father's assisted living will be open for visitors / when Connecticut will stop requiring people arriving from Tennessee to quarantine upon arrival.  So, in a couple of weeks, I will be able to say "I haven't spent this much time in one place in my memory/probably in my life."

Pre-pandemic, I had 4-5 trips queued up:

• Fly to Denver, and drive a loop CO»WY»SD»ND»MT»ID»OR»ID»UT»CO, to set foot in the 6 states in the lower 48 that I still lack.
  
• Operate the one of the CQWW radio contests from Sept-ÃŽles, QC
  
• Drive up to the Lac-Saint-Jean region of Québec
  
• Drive QC138 out to Kegasha
  
• Cruise, probably to the southern Caribbean

Those are still "on the list", although I'm becoming more inclined to drop the cruise and just fly to someplace and simply "be unplugged" for a week or two, and I'm tempted to add a cross-country rail trip after having watched a few YouTube videos.

[Duke87]

when Connecticut will stop requiring people arriving from Tennessee to quarantine upon arrival. 

So, Connecticut does not actually require this, although they would prefer people believe and act as if they do. Examine things closely and you will note that there is no mechanism for enforcement in place.

On the other hand, New York is able to fine people found to be in violation - so it is an actual requirement there. They've also started making people arriving at airports fill out forms, which turns up the heat as well since now you have to either go on official record as being required to quarantine or commit perjury. Though I do wonder what happens if you simply refuse to complete the form.


Regardless, the whole thing is a bunch of security theater anyway. Last night the mayor of Hoboken was on TV making noise about how "even if you have been tested and tested negative you still have to quarantine for 14 days". Why? You afraid people are going to spread a disease it was just confirmed they don't have?
Indeed, this creates a perverse incentive for people to not go and get tested after returning from travel, in order to avoid having any questions asked.

[kalvado]

Regardless, the whole thing is a bunch of security theater anyway. Last night the mayor of Hoboken was on TV making noise about how "even if you have been tested and tested negative you still have to quarantine for 14 days". Why? You afraid people are going to spread a disease it was just confirmed they don't have?
Indeed, this creates a perverse incentive for people to not go and get tested after returning from travel, in order to avoid having any questions asked.

Because negative test doesn't mean " confirmed they don't have virus". Two tests with a week interval seem to be a somewhat reasonable "no" at best.
That's why I don't understand if tests are actually that useful - the only case they are actually important for a patient is when a person has actual symptoms, and then a proper treatment may be chosen. But since there is no good treatment anyway...
Otherwise, it is mostly about public health - quarantine of those who may be spreaders - travellers, close contacts, confirmed positive cases. it is about meeting any ONE criteria..

[SEWIGuy]

According to testing data compiled by Johns Hopkins, there has been a 2.4X increase in positive tests from May 10th to July 4th (21,602 vs. 52,391).  But this 2.4X increase in positive tests can be attributed to a 2.4X increase in daily testing during the same time period (267,961 vs 644,415).  Both May 10th and July 4th data had the same percentage positivity rate of 8.1%.  More testing leads to more positive tests (shocking results!). 

Very impressive cherry-picking.

Let's try again with a start date of June 10th instead of May 10th. Now the positive tests go from 20,761 to 52,391 (+152%), while the total daily tests go from 429,207 to 644,415 (+50%). More testing leads to more positive tests...but not that many.

Not only that, but if you are testing more, you want the rate to go DOWN.  Not stay the same.  It means that a lot more people have the disease and can spread it to others.

Not necessarily. You're probably finding a lot more asymptomatic people, and I'm pretty sure I read over the weekend yet another study that pooh-poohs the concept of asymptomatic transmission. What's the difference between being asymptomatic and not knowing you have it vs. being asymptomatic and testing positive? Nothing, really. Higher numbers of asymptomatic tests doesn't really mean anything. The important numbers are hospitalizations and deaths. I think the statistics are still that 40 percent of positive tests are asymptomatic, 20 percent exhibit only very mild symptoms, and only about 10 percent requiring hospitalization, especially among the elderly and those with significant comorbidities. In Kentucky, fully 60 percent of the deaths have come from those who were in nursing homes. And isn't the national fatality rate 4 out of every 1,000? That's a 99.6 percent survival rate.

Even if the 40% figure is right, and nobody really knows if it is, that is still A LOT of poeple who have the disease with the abilty to transmit it presymptomatic or when they actually develop symptoms.

And I agree that hospitalizations and deaths are by far the more important stats.  But they seem to lag where large number of positives crop up.  Well at least the hospitalizations are - hopefully we won't see death rates crop up in Texas and Florida.

[Eth]

And isn't the national fatality rate 4 out of every 1,000? That's a 99.6 percent survival rate.

I suppose that depends on how many cases have gone undetected. Currently in Georgia (as of midday yesterday) there have been 95,516 detected cases and 2,860 deaths, which would be a fatality rate right at 3%. (That rate is falling; it was as high as 4.3% as recently as mid-June.) The actual number of total infections is generally estimated to be somewhere in the range of 4 to 9 times that number; if it's at the high end of that range, that would be fairly close to that 4/1000 figure.

And I agree that hospitalizations and deaths are by far the more important stats.  But they seem to lag where large number of positives crop up.  Well at least the hospitalizations are - hopefully we won't see death rates crop up in Texas and Florida.

The dynamic here is certainly interesting in that death rates have continued to decline, dropping from about 40/day in mid-June to around 12/day now, while in that same time period hospitalizations have roughly doubled from 75-80/day to 150/day. I can't help but feel like we're just waiting for the other shoe to drop.

[ftballfan]

Michigan reported their first day with no new deaths yesterday since mid-March!

[kalvado]

I suppose that depends on how many cases have gone undetected. Currently in Georgia (as of midday yesterday) there have been 95,516 detected cases and 2,860 deaths, which would be a fatality rate right at 3%. (That rate is falling; it was as high as 4.3% as recently as mid-June.) The actual number of total infections is generally estimated to be somewhere in the range of 4 to 9 times that number; if it's at the high end of that range, that would be fairly close to that 4/1000 figure.

That's exactly the math - but 10-12x that officially diagnosed is a more common range.
So anyone who gets it has a 90% chance of not realizing they just won the lottery, 6% recover from a bad-bad cold, 4% get it for real and if it is for real, it is 1 out of 10 chance of death.

[tradephoric]

Over the past 7 weeks Arizona's test positivity rate has risen from 6.5% on May 16th to 25.9% on July 4th yet deaths over the same period has remained relatively flat.  Arizona has a similar population to Massachusetts and their test positivity rate peaked in April at 29.4%.  When compared to Massachusetts outbreak, i'd expect Arizona to be averaging well over 100 deaths a day at this point but last check they were only average 31 deaths a day.  I get that deaths lag new cases, but to this degree?  Was there a massive outbreak in Massachusetts nursing homes that we just aren't seeing in Arizona?  There has to be a reason why not as many people are dying in Arizona's outbreak.