Post #684: State re-openings, and the key unanswered question

Posted on May 8, 2020

Soure:  The New York Times.

This is my first followup to Post #673, talking about re-opening state economies.  By that, I mean allowing a wider variety of business and social activity.

For this analysis, I went through various news organization summaries of state re-opening strategies.  My original goal was to begin setting up a statistical analysis of the impact of re-opening.  But it’s complicated enough that I have abandoned that for now.

Instead, here are my observations, from going through the re-openings with a fine-toothed comb.

First, it’s far too soon to tell if re-opening is having an impact on coronavirus case counts.  With the exception of Georgia, most re-opening activity did not start until 5/1/2020 or later.  Given the lag time between infection and symptoms (typically five days), and symptoms and reported test (maybe another five days), and the generally slow pace of re-opening, there would be no way to see any impact in the data yet.  Give it another couple of weeks.

Second, while there is some weird variation in what is and isn’t being done, most re-opening strategies are following a relatively similar pattern.  As far as I can tell, resuming (distanced) outdoor activity has the clearest consensus behind it.  That is, opening parks, beaches, and similar.  Where states take that explicit action, that’s almost always the very first thing they do.

Third, many states begin re-opening restaurant eat-in business via outdoor dining.  Three states directly address it, another handful address it indirectly by allow limited outdoor gatherings.  The rest start off by limiting capacity, requiring distancing, and so on.

Four, even if it not banned by the state, people avoided certain activities anyway.  That particularly pertains to church services.  Turns out, a handful of states (including Delaware, on the East Coast) never put any significant restrictions on church services.  And yet, e.g., the Catholic diocese that covers Delaware shut down all services anyway.  Thus showing that when the stakes are high, large responsible organizations can show good sense.

Only one state is allowing limited re-opening of schools:   Montana.  They never did have many cases, and they’re down to zero or one new case per day.  Re-opening schools is now (or will soon be) a local option there.

Everybody hates bars, and I can’t quite figure out why.  Even the states that are opening up some limited forms of sit-down dining are still keeping bars shut.  It was just a strange point of consensus, given the variations that states adopted.

Maybe half the US population is in one of three groups of states that have banded together to coordinate their re-openings.  The Western States Pact (WA, OR, CA), the Midwest Coalition (IL, KY, OH, WI, MI, MN), and the Northeast Coalition (NY, NJ, MA, CT, PA, RI).

But that doesn’t say what needs to be said:  Compare those three lists of states, to the map above.  When you do, you see the US states as largely falling into two groups.  Those who formed regional coalitions, and are keeping most restrictions in place.  And (with the exceptions of the DelMarVa region, LA, and NM) all the other states, who are opening up right away.  Although there is no formally-announced coordination in that latter group, there’s a de facto coordination.  The opening dates are frequently identical, and the phrase “safer-at-home” as the description of the re-opening strategy appears often among those states.  In effect, there are three formal coalitions keeping restriction in place, one informal one removing them immediately, and then a handful of truly stand-alone states or small regional cooperatives (such as DelMarVa).

There is some state-to-state variation in what is restricted and not, and how fast the phase-in occurs. My own observation is that some businesses feature far out-of-proportion to their economic importance.  For whatever reason, a lot of states feel the need to single out bowling alleys for specific treatment, along with tattoo parlors.  I can’t even imagine what the genesis of that is, but I’m not bowler.  And I don’t have tattoos.  I would say that many, possibly most, states have allowed routine dentistry to resume at this point. 

As far, as near as I can tell, literally no state has put forth any plans for determining when to abandon re-opening, if something goes wrong.  So, if failure is an option, the response to that failure either hasn’t been worked out yet, or doesn’t exist.


Where are the ongoing new infections coming from?

While there appears to be a fair degree of consensus among states on how to go about re-opening, I don’t think that’s science-based.  For the most part, I think that’s strictly herd behavior.  For something with these risks, everybody wants to do what everybody else is doing.

The only consensus move that appears supported by the epidemiology is the move toward opening outdoor areas.  I summarized that in prior posts.  In an analysis of 7000 infections in China, where the source was known, only one occurred in an outdoor setting.  In a statistical analysis of Japanese infections, outdoor settings had just 5% of the infection rate of seemingly-equivalent indoor settings.  So the drive to allow (e.g.) outdoor dining first, to re-open restaurants, appears well-founded.  As does the drive to re-open parks, golf courses, beaches, and similar areas for well-spaced outdoor recreation.

On May 6, Governor Cuomo presented the results of an analysis of 1000 New York State coronavirus hospital encounters.  Most of the people who were seen in at these hospitals were, by their own report, sheltering at home and (largely) not going out.  If true, that certainly knocks any science-based re-opening for a loop, because that should be the safest scenario possible.

So, first, that’s exactly what epidemiologists need to be doing right now — identifying the channels of disease transmission.  But second, if that’s as far as the analysis goes, then there is no answer to this question.

Cuomo speculated that this might be the results of older, largely retired or unemployed individuals, sheltering in place, but being exposed by behavior of younger family members.  That is, only the older infected household members are sick enough to require hospitalization.  So this analysis addressed who it was that got sick, but did not really identify the disease vector.

This is consistent with Chinese contact tracing, which showed that the highest-risk area for transmission of disease was within the household.  But there, again, it’s technically correct to say that, but that completely misses the point.

The point is, what’s the vector that brings that disease into the household in the first place?  So far, I have yet to find even one single helpful bit of analysis in that regard.  And that’s what you’d really like to know, as these states re-open a wide variety of retail location.  But at this point, it looks like we may never have any useful answer to that question.