U.S. COVID-19 Lockdowns: Correlations, Outcomes, Idiocy

Covid Deaths, Not Lockdown Deaths, part 5

U.S. COVID-19 Lockdowns: Correlations, Outcomes, Idiocy

November 21, 2020

How Lockdown Failed Before it Began

I'll start with "Sweden Has a Lower COVID-19 Death Rate Than the US" by Jon Miltimore, September 9, 2020, Catalyst Independent, but first from FEEE (Foundation for Economic Education and Epidemiology - kidding on the last E). 

Miltmore writes: "The year 2020 will go down in history as a historic calamity. But this was not because COVID-19 struck," he declares, noting deadly respiratory viruses have existed as long as humans have." Longer, in fact. And they've always been fairly harmless, or whatever maybe not always, but this one is invisible, which makes it extra harmless, and the disaster arose "because central planners erroneously believed the best way to protect humanity from an invisible respiratory virus was to order healthy people to remain in their homes under almost all conditions, in many cases under threat of fine or imprisonment." I remember when they tested us all and made the healthy ones stay home. Sad day for humanity there. No, of course the point is it's invisible, you can't be sure who IS healthy and not contagious, so they ordered the sick and anyone who might be - which is everyone - to come out only as needed and cautiously. Just why that's an erroneous approach isn't clear, once the idea is unscrambled. 

Anyway, this piece cites something run in the Wall Street Journal as “The Failed Lockdown Experiment,” wherein "Donald L. Luskin, the chief investment officer of TrendMacro, a global investment strategy consulting firm, says data show lockdowns are actually correlated with a greater spread of the virus." Ah! Leave it to a business stimulation expert and data manipulator to find what all global disease experts have missed. He finds a correlation. There are a lot of those. 

Apparently Luskin checked virus spread until lockdown but not after, found that of all the times he checked, it was the worst right at the start of maximum lockdown. Luskin's explanation, re-worded just a bit so everything he has correlating is reversed from how he presented it. If they correlate, that should be fair; a = b is the same as b = a.

Measuring from the start of the year to each state’s point of maximum lockdown—which range from April 5 to April 18—it turns out that a greater spread of the virus correlated with lockdowns. States with larger Covid outbreaks also had longer, stricter lockdowns. The five places with the heaviest caseloads - the District of Columbia, New York, Michigan, New Jersey and Massachusetts — had the harshest lockdowns. 

It could be that strict lockdowns were imposed as a response to already severe outbreaks.

Yeah, it very well f@*#ing could be, huh? And it got way better after the lockdowns in each place. Case confirmation might increase, but transmission will decline if people follow the guidelines. Deaths always do increase afterwards, as that maximum caseload dies off over the following weeks (see here), and then it improves.  I'm not a subscriber, haven't seen any illustration Luskin used, but as I read it:

And so lockdown was said to have failed. 

Of course there were measures underway to slow the spread prior to that, including a suggestion of - basically - voluntary lockdown, during which time the virus spread rapidly anyway. He'll argue people keeping apart and even staying indoors with their families instead of taking the bus to work or going on a travel vacation is why it spread so widely between households and across whole regions. But clearly the orders - to the extent they were heeded - slowed what was going to be an even more amazing spread. This stuff has a reproduction rate over 5 sometimes, if left to run freely. This SARS CoV-2 is some powerful stuff.

Miltmore notes lockdown duration, but not what happens during that lamentable span, which is a vast improvement. Fox News' Dr. Marc Siegel told Tucker Carlson “Lockdowns don’t work if there is already a lot of virus in the area," but he did note "over 20 states that have come out of lockdown have actually seen an decreased number of cases." Again a = b, so we could say 20 states that finally saw a decreased number then came out of lockdown, enjoying that advantage. Does it get better or worse after that? Siegel wouldn't be able to answer that honestly. (Dr. Siegel Tells Tucker: Lockdowns Producing ‘More Deaths From Despair’ Than Coronavirus Daily Caller, May 20) he also claimed "that “deaths from despair” could surpass 75,000 in the U.S. “because of COVID-19 and people not having jobs.” ... "It’s going to be more deaths from despair than from the virus itself," while the headline says lockdowns are "producing more deaths" (present tense). See part 4 for my assessment of that claim.

Poorer Outcomes: ND vs. NY

Finally, dr. Siegel also "mentioned the absence of coronavirus cases in South Dakota “which was never in lockdown.” Joel Smalley, MBA, also found this concept significant, penning the article "Study: Dems COVID19 Lockdown Measures Causing Most Deaths" which finds as of June 27, "the empirical data very strongly suggests that mortality outcomes are improved with fewer interventions." That doesn't seem to make sense, but he points out: 

"...of the 12 states that have experienced no excess death at all during the period in question (Alaska (R), Arkansas (R), Hawaii, Idaho (R), Kentucky, Maine, Montana, North Carolina, North Dakota (R), Oklahoma (R), South Dakota (R), and West Virginia (R)), 5 of them (Arkansas (R), Kentucky (D), North Dakota (R), Oklahoma (R), and South Dakota (R)) had no ostensible lockdown. " 

They tend to be Republican, as he selectively notes ... they're also all of our sparsely populated "podunk states," lacking much for big cities, with less using public transport, and none was hit by an early surprise outbreak that we know of.  Texas is pretty Republican too but faired poorly and had a lockdown - whatever the correlation there. Hawaii is Democrat but island-based, which seems to be a huge plus for controlling covid, especially with help of a lockdown. A lot of factors play, but being this is a highly contagious virus, Smalley found the political ones most worth considering, and just went the hell off:

"It is evident that Democrat states have a much stronger tendency towards intervention and this has led to much poorer outcomes for citizens of those states.... A possible reason could be in the nature of collectivism versus individualism, where Republicans might be more likely to take appropriate responsibility for their own welfare, making decisions and taking actions according to their own perceived risks, whereas members of the collectivist states may be more inclined to rely on the diktats of the state even though they may not be logical or reasonable. This conjecture would need much deeper investigation to be upheld."

He surmises Democrat states are full of irresponsible Socialists who follow the state's "diktats" - his actual word - and just to "protect the common good" rather than their own Ayn Rand interests, they hunker down in terror behind walls and masks, and then naturally, evil things happen and a lot of people die. The same was happening in more populous Republican states like Texas and Florida even as Smalley wrote, but ... Nebraskans for example take responsibility and do nothing about a harmless flu except washing their hands and praying, just like always. Naturally it spreads less because ... prairies and real American values scare the hell out of this commie Chinese virus? There's no point to studying that at all. If you believe it, you just keep doing that and re-arrange the facts so they seem to agree.

However it seemed in late June, we have a better view now in November to judge which approach truly has the worse outcome. A handy data map I saw on Twitter shows per-capita rates of confirmed infections for all48 contiguous states, as of November 13. Here it's annotated to show the five no-lockdown states as I recall it being when I checked way back. ND, SD, Nebraska, Iowa, and Arkansas. These are the kind of sparsely-populated states that could possibly do without stern measures, and perhaps did fine for a while. But Iowa and the Dakotas especially wore out that buffer - maybe by standing together like that - to become the most-infected states, per capita. It turns out they weren't spread-out enough. They didn't each have their own personal workplace, grocery store, or bar, so it finally spread all over. By this, a week ago North Dakota had nearly 8,000 per 100,000 infected, or near 8% - CONFIRMED. That's 2.7 times as infected as New York (which is at 2898.9, or 2.9%, and probably with a much higher confirmation rate). 

North Dakotans had come together to log 2,270 new COVID-19 cases on November 13 before before Governor Burgum announced a statewide "lockdown" as of the 14th: a public mask mandate would be in effect, with businesses limited to 50% or even 25% capacity, school sports and big gatherings banned. Critics of his reaction to another "casedemic" might fail to note it came with a peak of 334  covid-related hospitalizations on the 12th. Hospitals were reportedly at 100% capacity with more than 250 new patients a day since November 1st., and so shorthanded they had covid-positive nurses keep working (Grand Forks Herald). Since the lockdown, cases dropped sharply to about 1,000 a day but quickly began creeping back up to 1,500 now. Hospitalizations declined but remain high, and deaths are averaging 16/day, with up to 30 (reported) in a single day, and it seems to still be rising (Bing tracker - state site). Again, the state's population is not even one million. What they have now is terrible, but it was about to become enormous, even there.

Joel Smalley and Dr. Siegel must be disappointed. Even in North Dakota, Burgum locks the people down and again it correlates with a dramatic rise in cases and deaths. Is there no end to this Communist plot?

I looked at the North Dakota's covid and all-causes deaths and compare them to New York state - minus New York City because that's how the CDC table I use lists it, and for a fairer comparison. There are some tallying issues here; New York's numbers per CDC are higher than I get subtracting NYC's dead from the state's total at Bing, while Bing lists considerably more deaths than the CDC for North Dakota. I haven't taken the time to sort that all out, so the comparison here may be skewed. Also the numbers for ND are so small it's hard to be sure about any pattern that seems to emerge, but ..., As of now, ND seems to have roughly or exactly caught up with 'the rest of New York' for COVID-19 fatalities, per capita - just over one in a thousand killed so far, or by week 42 ending Oct. 17. They've both gotten sharply worse sincer then, but the public data on that isn't clarified yet. 

* 12,134 COVID-19 deaths in NY aside from NYC (by CDC's numbers) = 1,091/m or 0.109%

* 824 in ND (by Bing's fuller-seeming numbers) = 1,081/m or 0.108%. (it's gone up 16 since the graphic: 840 now = 1,102/m. But NY's gone up too.)

Here are the numbers for those plotted covid deaths (red line). That's the same actual number, not population-adjusted. Any week where they have the same death toll is a week where North Dakotans died at 14.6 times the rate of New Yorkers (again excluding the even worse-hit New York City). They're close to equal these days, with North Dakota usually being lower, but sometimes higher. Right now (not sure how to get a comparable daily tally for NY), North Dakota might have hit triple digit deaths first. 

Expanding a bit to consider "non-covid" excess deaths: starting with natural causes rather than all causes, and calculating excess over same-week 2019 levels, New York aside from the city had 14,173 above 2019 levels up to week 42  (after that, the numbers are too incomplete to bother with). Using a baseline of 80 deaths/week below 2019 (a possibly high average level) adds 3,360 to w42 = 17,553 excess deaths, almost all of them probably due to COIVD-19, confirmed or not. Adjusted for population, that would be 1,577 per one million, or 0.158% dead up to week 42 ending October 17. 

Natural Causes excess for North Dakota: total to w42= 470 - more than the CD's total to then of 384 covid deaths, or Bing's 404, but not by much, considering the mini-spikes seen at weeks 10-12, 17/18 and 35 forward. The baseline here is hard to call, and I went with a mild 14/week from week 14, and just half that before. That would give 967 excess deaths to w42, which feels a bit too high. But then again the scale of unconfirmed deaths is unknown and it's had a normal-length spring and summer to add up. That would be 1,269/million, or 0.127% of North Dakotans killed by then. That's still better than New York, but of course it's gotten worse since. 

By my rough baseline-adjusted calculations, New York has had 30.8% of covid fatalities go un-noted (otherwise, it's less than that - or more if I did it low). That would be a bit on the good side of normal, where a U.S. average seems to be roughly 1/3 missed, or about 33%, and 50% or more missed is not uncommon; the U.K. seems close to that. Belarus, by apparent policy of counting low, misses perhaps 90% of its massive covid deaths (to be addressed in a future post). North Dakota looks good in contrast, but it seems they keep a bit looser track than in New York and have so far missed close to half of their deaths (probably less than the 58% I get comparing 967 excess to 404 covid). 

Anyway, we'll see who has the worse outcome as we move deeper into flu season, but I'll say it now: nice outcome there, Joel Smalley, MBA. That was just as stupid as the Trend Macro guy's correlation work.