Skip to main content

A Head Full of Coronavirus Research, Part 20

By September 28, 2020Commentary

I am temporarily retiring the Drowning series and returning to Head Full, but I don’t remember where I left off, so I am starting with Part 20.

First, let me comment briefly on the situation in the Upper Midwest–my home state of Minnesota and its neighbors.  There was a rise in cases, and a bump in hospitalizations, perhaps deaths.  As in Minnesota, it is hard to get good data from other states on actual dates of tests, positive cases and deaths.  So I am always cautious about discussing trends.  Minnesota, even with its absurd policy of testing everyone, everyday, appears to have a lower rise, although, again, without pure data I hesitate to make a comparison.  But Minnesota clearly had it worse in the late winter, early spring.  There is one very obvious reason why that might be the case.  We have a very busy international and domestic airport.  I personally was in Miami, San Diego, Los Angeles, San Francisco and several other destinations from mid-January til things shut down in mid-March.  So were a ton of other Minnesotans, and many were still traveling overseas.  So we likely had our epidemic seeded multiple times by travelers from abroad and from hot spots in the US.  And if we had more cases then, we are likely to have fewer now.

In other wonderful local news, the Children’s Hospital is looking at significant layoffs and reduction in services, due to big drops in volume and revenue.  So we all understand what that means, let me spell it out–children are avoiding needed health care.   (CH Story)

Here is another story full of good news for children–keeping them out of real school is badly damaging them.  (New Yorker Story)   Remote learning isn’t remotely like learning.  It is a joke.  It hurts minorities and low-income kids the worst.  It deprives children of absolutely necessary social experiences.  The teachers’ unions and politicians who are doing this should be severely punished, especially because it is all politically motivated–having kids home makes it hard for parents to work and damages the economy, which is exactly what the teachers want to do heading into the election.  The most disgraceful part of a very disgraceful epidemic response.

Oh, but wait, if you have courage and actually want to follow the science and do what is best for children, you can insist that they have in-person learning and the risk from CV-19 turns out to be very low.  The story about Florida’s experience validates that, much to the disappointment of the hysteria crowd.  (USA Today Story)   There has to be a way to force schools to open to exclusively in-person education.  The children are actually safer at school than at home.  Florida has a real leader, we have an Incompetent Blowhard.

Now a couple of real studies.  Gabriela Gomes’ did the most innovative early modeling that allowed for variability in susceptibility and infection.  While others harshly criticized the approach, it has been more accurate and others have begun to emulate this model.  In a new paper, Gomes et al debunk a particularly critical paper, showing the superiority of their approach to that that of the critics.  (Medrxiv Paper)   The paper they were analyzing used a homogeneity in susceptibility assumption and had an absurd assumed case fatality rate of over 1%.  It is likely actually less that one-tenth of a percent and still falling.  In any event, the Gomes group showed that their model was a far better fit to the actual experience and was much better at predicting cases and deaths two weeks out.  In addition, the heterogeneity approach shows a much lower supposed effect of lockdowns and other measures.

This paper looked at the experience of some low and middle-income countries, asking if they might be at population immunity.  (Medrxiv Paper)   A number of these countries had sharp peaks in cases and then rapid falls notwithstanding the easing of restrictions.  Using a model, they found that 51% to 80% of the population had been infected in these countries, with only a very small percent of cases detected.  If true, this would suggest very substantial slowing of transmission.  The case fatality rates would be very low, much lower than for flu.

This paper is another attempt to estimate how many cases there really are.  I am at the why bother stage.  We need to stop pretending that this huge proportion of asymptomatic infections means anything at all.  It doesn’t.  (Medrxiv Paper)   The authors use “machine learning”, which is just another way of saying model, to find that there is a large proportion of undetected cases and to estimate what that is.  Seroprevalence data was used in part.  At some points in the epidemic, they found that only 10% of cases were being identified, which is consistent with the general number reported by a number of studies.  That percent has probably gotten higher as testing has gone wild.

And let us end with more good news.  Mental health has deteriorated due to the terrorification program and associated measures.  A group in the UK discussed changes in a part of London.  (Medrxiv Paper)   The time period studied was the early peak of the epidemic in the UK.  Not only did outpatient and inpatient use of mental health services decline, the number of mental health patients who died increased.  At the same time service use was declining, people were expressing greater anxiety, depression and other mental health issues.


Join the discussion 2 Comments

  • SteveD says:

    ‘We need to stop pretending that this huge proportion of asymptomatic infections means anything at all. It doesn’t.’

    It may mean we are closer to population immunity than we think. I am not sure what level of immunity a very light asymptomatic infection may induce.

Leave a comment