Im-Politic: The Case for Shutdowns Remains Far from Open and Shut

Tags

Arizona, California, CCP Virus, coronavirus, COVID 19, Doug Ducey, Gavin Newsome, hospitalizations, Im-Politic, infections, reopening, shutdown, Wuhan virus

OK, enough with the History Wars/Cancel Culture stuff (for now)! Let’s turn to something relatively uncontroversial (!) – like the resurgence of CCP Virus cases in the United States. And in particular, let’s focus on the common argument that these increases, along with higher hospitalization rates, show conclusively that the states that closed late or reopened early or never closed or never closed much (you get the picture), were tragically and even recklessly mistaken, and that the states that closed early or stayed closed for the longest and have reopened only very gradually were the truly responsible actors.

There’s one set of actions that certainly seems to clinch the case for the “closers” – the decisions of many of the late closing/early opening states to pause or slow their reopenings, or even roll back many. According to this report, the number of states and cities taking such steps now totals 15.

But this loudly nagging question remains: What’s the actual evidence that supposedly irresponsible reopening measures have caused either the growing case or even hospitalization numbers? Pretty flimsy, as it turns out.

Let’s begin with Arizona – widely described as a quintessential reopening disaster. It’s true that confirmed virus cases have shot up since very late May, with new daily cases per hundred thousand residents soaring from 248.5 on May 27 to 1,063.9 on June 28. That’s about a four-fold increase. (All such infection figures and change come from the Washington Post‘s continually updated virus tracker feature.) 

But figuring out the impact of shutdown-type policies involves not only identifying how many new cases and other indicators have worsened recently. It much more importantly involves comparing rates of worsening during the shutdown periods and afterwards.

In Arizona’s case, Republican Governor Doug Ducey began imposing limits in mid-March, and canceled school for the remainder of the current academic year and issued a stay-at-home order on the 30^th. (My Arizona dates come from this timeline.) On that day, new case numbers stood at 16.7 per hundred thousand. On May 11, he decided to reopen restaurants. On that day, new cases hit 163.8 per hundred thousand – meaning that they.rose a little less than ten-fold during the shutdown.

As described above, the latest Arizona numbers show 1,063.9 new daily cases. So during reopening phase, the state’s new daily case numbers rose about 6.5-fold. In other words, cases rose much more slowly during the reopening phase (which Ducey paused as of yesterday) as it did during the shutdown phase. So that’s supposed to be evidence that its shutdown policies have been reckless?

Even taking into account the laws of large and small numbers (which teach that the lower the absolute number baseline you begin with, the easier it is to produce big percentage gains), I’m glad I don’t have to use them to strengthen the case for stricter reopenings – if only because the virus seems to spread so rapidly. As a result, independent of improving testing rates (which themselves should be revealing more cases), the rate of infections should be speeding up the more infections are recorded, not slowing down.

Even harder to explain from the standpoint of shutdown enthusiasts: On March 30, when the state’s major shutdown phase began, Arizona CCP Virus-related daily hospitalizations stood at 51 (total). By May 11, when major reopening began, it had fallen to 41. Since hospitalization rates are often called the most important measure of progress or backsliding against the virus, that’s a big sign that shutdowns work, right?

Not exactly. For yesterday’s new hospitalization number (the latest available) was only two. In all fairness, between the start of major reopening and June 8, this number doubled – from 41 to 82. But since then, the daily hospitalization rate has sunk like a stone. Even granted the reality of lags between new case identifications and hospitalizations, these figures make clear that the state of Arizona’s economy and the regulations governing activity have had no discernible impact on its new case numbers or its hospitalization numbers.

Of course, Arizona’s only one state. So let’s look at another – and the biggest in terms of population. That’s of course, California. Even better, California was a state praised for shutting down early and aggressively, and has reopened cautiously – till very recently, when it mandated some rollbacks. And its experience strengthens the case for sweeping shutdowns not one iota.

California’s new case numbers have been rising, too – though without the recent spike seen in Arizona. Even so, they’re now at an all-time daily high as well: with 553.1 new cases per hundred thousand residents. This total is only a little over half of Arizona’s latest, which would seem to reenforce the case for stricter shutdown policies. But again, that’s not the new case statistic that deserves the most attention. It’s imperative to look at California’s new daily cases and how they’ve changed during both the state’s shutdown phase and its reopening phase.

The Golden State’s Democratic Governor, Gavin Newsom, issued a state-wide stay-at-home order on that went into effect March 19 — actually, not so very different from when Arizona’s shutdowns began in earnest, though they were narrower. That day, the state’s reported cases per 100,000 residents stood at 2.6.  

On May 8, California began a gradual reopening. That day, its reported cases per 100,000 residents was 159.7, meaning that during the shutdown, they skyrocketed by 61 times. Because the latest such figure is 553.1 per thousand, the rate of increase during the reopening was 2.46 times. Again, not results that speak well for the spread-inhibiting record of shutdowns.

The California hospitalization story is more complicated than Arizona’s – but anything but a slam dunk for shutdown supporters. The state tracks hospitalization differently from that of its neighbor, recording daily changes in net hospitalizations (i.e., admissions minus discharges). What the numbers show, according to this chart, is that is that shortly after the stay-at-home order went into effect (March 20), they fell pretty steadily through very early June,stayed roughly level until the middle of the month, and soared thereafter.

What’s especially interesting about these figures is that they seem completely unrelated to the shutdown phase and the reopening. After all, net hospitalizations kept falling for nearly a full month after the gradual reopening began (though at a somewhat slower rate). And the big increase in net hospitalization that began in mid-June didn’t start until some five weeks after the reopening. As a result, I don’t see much of a shutdown/reopening connection there, either.

It’s certainly possible that reopening decisions helped increase infection rates by encouraging Americans in those states to believe that all was clear, and that life could go back to normal, without any mask-wearing or social distancing and the like. But which state governors have made those claims? It’s more likely that many residents of those states decided to throw caution to the wind regardless of what officials said – which doesn’t speak well for them, although it also doesn’t say much about the enforceability of shutdowns. What’s the alternative, however? Handing state and local officials China-like powers? Raise your hand if you’d be happy with that outcome.

Moreover, these shutdown/reopening figures tell us nothing about another, and too often neglected, crucial dimension of the virus policy debate – what are the public health costs of prolonged continuation of shutdowns? Heart Disease Patients’ and Victims of Depression’s Lives Matter, too, after all.

So far, it seems clear that, like crises and other difficult situations in general, the keys to dealing with the CCP Virus pandemic won’t be hyping clearcut formulas,  false either-or-choices, and declarations of certainty as ringing as they are unjustified and blame-casting. There’s still no substitute for good judgment, common sense (and yes, that includes mask-wearing when in crowded indoor spaces, steering clear of them and outdoor crowds as much as possible, and washing your hands fanatically), and an ability to learn.  What a shame that the national stockpiles of those qualities seem so meager.  

Im-Politic: How Much Did the Lockdowns Really Help?

Tags

African Americans, CCP Virus, coronavirus, COVID 19, economy, Georgia, hospitalizations, Im-Politic, lockdowns, public health, reopening, shutdown, Virginia, Washington Post, Wuhan virus

Is it time to start putting the CCP Virus economy-reopening debate shoe on the other foot, at least when it comes to one key measure of progress or backsliding against the pandemic? More specifically, is it time to put less emphasis on finding out whether states that have reopened relatively quickly have seen their virus situations worsening, and more on whether states that closed early and/or have stayed largely closed have achieved progress that’s been any better?

This question occurred to me this morning upon reading in my Washington Post that when it comes to new infections and fatalities, Virginia has just seen record highs recently whether we’re talking about single day totals or the more informative seven-day averages. That’s striking because Virginia has been one of those states that shutdown substantially quite early, and has reopened very slowly.

So I began wondering how Virginia’s record compares with a state that reopened very early – Georgia. And the numbers clearly show that their performances over the most relevant timeframes have been…pretty comparable. Which represents new evidence that the economically devastating lockdowns have been under-performers for containing the virus’ spread.

Virginia and Georgia are particularly interesting to compare because of their similarities. The latter’s total population is estimated this year at 8.63 million while the latter’s is a not greatly bigger 10.74 million.

Both states also have relatively big populations of African-Americans – who have been among the virus’ biggest victims. Blacks represent 31.03 percent of all Georgians, and 18.81 percent of all Virginians.

That Washington Post Virginia article did mention one area of continuing improvement for the state: new hospitalizations. They’re especially important both because fears of hospitals getting overwhelmed by the pandemic were prime justifications for the original shutdown orders, and because they’re the best measures of whether the virus is being contained or not. After all, numbers or new cases seem to depend heavily on increases in testing (which naturally reveal more and more infections). And controversies over identifying genuine CCP Virus-induced deaths remain heated – in large part because methodologies vary so greatly state-by-state.

By contrast, there have been no debates over how many patients with virus symptoms have been admitted to healthcare facilities. The only uncertainties are those stemming from how promptly these facilities report their admissions to state health departments.

That kind of uncertainty is still clouding Virginia’s data. As of today, (see this link and scroll down till you see the option for hospitalization data) the state has only reported new hospitalizations through May 20, and these data are divided between confirmed cases and probably cases. (The former are the great majority, though.)

Even so, because of Virginia’s lockdown policy – which began in earnest at the end of March, began easing in phases for the state’s least populous areas in mid-May, but which largely continue for its most populous areas (those closest to the District of Columbia)– it should be among the gold standard states for virus progress if turning off most economic activity is considered crucial. (Here’s an unusually informative lockdown timeline for Virginia, Maryland, and the District.)

Its interactive hospitalization chart is a little hard to read, but it seems to show that on March 31, the seven-day moving average of new admissions stood at just under 59, and through early May (when the lockdown began to be lifted). moved up steadily to a little over 81. So they rose by just under 39 percent. By May 20, this average had decreased all the way to just under 45. In other words, daily hospitalizations dropped by a little less than 45 percent. And for the entire period, the seven-day moving average for new hospitalizations dipped by 2.34 percent.

Georgia’s lockdown began only a bit later than Maryland’s (on April 2) but serious easing began much earlier (on April 24). Indeed, Governor Brian Kemp was widely pilloried for the decision.

During its three weeks of lockdown, Georgia’s seven-day average daily hospitalization numbers went from about 80 to about 130. (The non-interactive chart below is even harder to read precisely than Virginia’s interactive graphic, but check it out for yourself below.)

This roughly 62.50 percent rise in daily hospitalizations was much higher than Virginia’s during its lockdown period Did this discrepancy mean that Georgia ended its lockdown too soon? Or was its somewhat heavier African-American population density the major difference? Search me.

Georgia’s reopening has been more aggressive than Virginia’s, and that could well explain why its seven-day average hospitalization figure remained just about flat from the start of this phase through May 22.

But I’m not entirely persuaded that the lack of improvement during this period means that Georgia’s relatively fast reopening has flopped. Because for the first three weeks of this reopening, the state’s seven-day average new hospitalization figure fell by about half – faster than Virginia’s during its slower reopening. And as the Post has reported, despite Virginia’s caution, daily (although not yet seven-day averages) have been rising recently, too.

The fairest conclusion to me seems that the hospitalization data give an edge to Virginia’s more cautious lockdown-reopening strategy, but that the edge is on the modest side. And most important, it’s far from clear that this margin justifies both the economic and healthcare costs of relatively longer and/or more thorough lockdowns.