Post #1186: Part 2A, the new COVID-19 numbers, the new COVID-19 odds.

Posted on July 26, 2021

This is a continuation of yesterday’s post.

Yesterday, I tried to get a grip on where we are in this U.S. fifth (Delta) COVID-19 wave.  I gave my best guess for where this is heading, in terms of new cases, hospitalizations, and deaths.

In this post, I start to look at “the odds”.  Mostly, what are the odds of getting serious ill if you are fully vaccinated?  But also a look at some other key ways in which the Delta variant is different from prior variants, and really has shifted all the relevant numbers.

Mostly, I want to get across one point:  Just because you haven’t seen something happen, so far, doesn’t mean it’s not going to happen now.  This wave is different, this variant is different.  This next set of posts will try to quantify just how different.  As I’ve said since early June, the numbers are against us on this one (Post #1160).

The new case count went vertical almost two weeks ago (Post #1173).  And that is not a surprise.  It falls right out of the numbers, when you combine this new more-infectious variant with a lack of COVID-19 hygiene.  From that post:

A particular concern, raised by a friend of my wife’s, is the rate at which your vaccine-related immunity fades over time, and the actual level of protection an aging vaccine gives against the Delta variant.  If you were vaccinated six months ago, should you be thinking about getting a booster shot?  Turns out, there’s a lot of apparently contradictory information about that.  That key question will take some time to sort out.  The key vaccine-booster issue is for tomorrow’s post.

In this post, by contrast, I just want to show a few numbers to underline that this is a new and rapidly-evolving situation.  If you follow the herd, or rely on instincts developed over the past year, you may be led astray.

If I had one piece of practical advice to give, it’s to buy yourself a few quality N95 respirators (masks), right now, if you don’t already own some.  And start wearing them in indoor public spaces.

Currently, NIOSH-certified N95s, even the flagship 3M N95s, are readily available at local hardware stores and on-line.  I expect those high-quality masks to evaporate off the shelves again, once this wave really gets rolling.  So take this as a window of opportunity to upgrade your COVID-19 protection.

If you don’t grasp how vastly better N95s are compared to common cloth masks, or to surgical masks, please read Post #938 for the math, or for the discussion read Post #935, If you have ten-cent lungs, by all means wear a ten-cent mask. 

2.1 Delta is faster.  Deal with it.

Where we are:  The last reported number for new cases / 100,000 / day for the U.S. was from last Friday’s data, at 15.5 new cases / 100,000 / day.  We won’t have another fix on the current U.S. trend in new COVID-19 cases until tomorrow, when counts of cases from Monday will be reported.

Where I think we’ll shortly be:  My projection (Post #1185) is that within about three weeks, the U.S. fifth (Delta) wave will exceed our third (winter, worst-so-far) wave.  If we’re lucky, it’ll only do that briefly, like so:


If that strikes you as far-fetched, then you definitely need to keep reading this section.  If, by contrast, you get it, feel free to skip to the next section.

Have a look at what has happened to Great Britain.  The Delta variant started spreading there just a few weeks before it took hold in the U.S.

They’ve nearly equaled their winter peak new case rate. The peak new case rate shown (~ 47,000) works out to be 71 / 100,000 / day.  That’s just a bit below the U.S. winter wave peak of 78 / 100,000 / day.

That increase in cases in Great Britain occurred in the seven weeks from June 1 to July 21.

And they have a higher vaccination rate than ours. 

And they were in the middle of a nation-wide partial lockdown.

I hope that, with that as context, my projection for the U.S. doesn’t look quite so far-fetched.  In fact, I’d call it rather mainstream, possibly conservative, given what has actually just happened in Great Britain.

My point is, with Delta, all the numbers have changed.  You need to change your thinking accordingly. 

Just FYI, apparently Orlando’s current new case rate already exceeds its Winter 2020 peak.  If you understand how things have changed, you realize that’s not a one-time event.  That’s the new normal.  That’s a story you’re going to be reading many times in the next few weeks, just substituting a different place name.

2.2  Delta is more virulent.

A few days ago, in Post #1178, I did some analysis in reaction to some common internet disinformation about the seriousness of Delta infections.  The disinformation is that Delta is less deadly than prior variants.

That’s wrong.  In that post I used the simple hospitalization and mortality case rates (rates per person diagnosed with COVID-19) to show that Delta is at least as deadly as the prior strains.  Starting from scratch, documenting sources and methods, I came up with this graph:

That’s enough to demonstrate that the Delta variant certainly isn’t less deadly than the prior variants.

But as I noted at the time, something isn’t quite right about those curves.  If Delta was about just as deadly (or as likely to cause hospitalization) as prior variants, those curves should have come down sharply at the end of the time period.

Why?  First, almost all the risk of hospitalization and death is among the elderly. 

To see that, look at the COVID-19 hospitalization and mortality case rates by age.  This is data from Virginia, as of yesterday, and summarizes all the data for the entire pandemic.  It’s reasonably representative of the U.S. as a whole. The height of the bar represents the fraction of diagnosed cases who were hospitalized or died.

Source:  Calculated from data available from the Virginia Department of Health.

As you can see, the case rates of mortality and hospitalization are much higher for the elderly.  In this case, the 60+ population contains:

  • 19% of all cases.
  • 60% of all hospitalizations.
  • 89% of all deaths.

Second, the higher rate of vaccination for the elderly has differentially removed them from the pool of persons being infected.  If the virus remained the same, removing those high-risk individuals should have brought mortality and hospitalization down sharply.

In short, as this has increasingly become a young person’s pandemic, hospitalization and mortality should have fallen.  But they have not.

I’ve been trying to find the data sources to allow me to estimate just how much worse the current variant is, but in fact, I don’t have to do the work.  All I have to do is cite properly-done research that shows what the numbers tell me has to be true.

Here’s the popular reporting, and here’s the research itself.  This comparison between Delta and the prior variants account for factors that affect risk of hospitalization, including age, sex, and presence of comorbidties.  It’s really trying to get at the risk of hospitalization for identical persons, one infected with Delta, one with the prior variant.

The popular reporting boils it down to a single number:

Source: Reuters.

But if you read the actual research, you’ll see that there’s considerable uncertainty around that “double” figure.  The “95% confidence interval ran from 1.39-times to 2.47-times as likely to be hospitalized.

Despite the uncertainty, this study reconciles what I’ve been seeing in the numbers.  The new variant is more virulent.  That has offset the reductions we should have seen as the elderly were differentially removed from the pool of newly-infected individuals.

Obviously, I’d like to see other reputable studies that show the same thing.  You don’t want to rely on one piece of research, even if it’s competently done.  But this appears to be the only rigorous analysis of Delta and hospitalization so far.

The upshot is that Delta is not only more infectious, it’s also more virulent.  For a given person, the risk of hospitalization after being infected is significantly higher with Delta than with prior variants.

To be continued

I haven’t really gotten to the key question yet:  How has Delta changed your risks and safety,  assuming you are fully vaccinated.

  • What’s your risk of getting a “breakthrough” infection.
  • What’s the risk of serious outcome, if you do get such an infection?  A
  • How long does immunity last?
  • Should you be thinking about getting a booster shot.

There appears to be some seriously conflicting information on these issues.  In particular, some of the numbers out of Israel on the fading of immunity over time do not seem credible to me, and I suspect that somebody is mis-interpreting the actual data.  It’s going to take me at least another day to sort through that.

Those questions are what I hope to address tomorrow, along with an update on trends.