I spent a lot of time on my “when will it end” section, immediately below, so I ignored almost all other news. I’ll catch up tomorrow.
Mitigation Measures/Vaccines/Opinion/Analysis
Everybody wants to know when the pandemic will end — when we can stop masking in grocery stores, when we can go to the Whatsit Expo without Lizard Brain freaking out, when we can get on airplanes again without assessing the risk of every person we encounter. When it was like Before Times.
I don’t think it will be for three to five years. Sorry.
I think we are at a point where we — even in BC, with 90% of our eligible population vaccinated — cannot reach herd immunity. (See The math on herd immunity below.)
Is all hopeless forever? No, for several reasons.
First, take comfort in measles. Measles has an R0 of 12 to 18, and yet it has been made, for the most part, a non-issue. This is in large part because the measles vaccination is extremely good: 93% sterilizing immunity (meaning you neither catch it nor transmit it) with one dose, 97% with two doses. At R0 of 12 and VE of .97, “only” 94.5% of the population needs to be vaccinated. Yes, 94.5% is a very high number, especially for the USA, but it shows that it is possible.
And better vaccines are coming. According to the New York Times vaccine tracker, there are THIRTY-TWO vaccines in Phase 3 trials (which test for effectiveness). (I am particularly excited about the GBP510 vaccine, which this article talks about a bit.)
There is also some speculation that a nasal vaccine would do a better job of reducing transmission. Intramuscular jabs like the COVID-19 ones drain into the lymph system and make gamma immunoglobulin antibodies, which hang out in the blood. The antibodies which you really need for an attack that comes in via the respiratory system are alpha immunoglobulins, which hang out in the mucous membranes. Vaccines administered by squirts up the nose seem to be more effective at preventing infection. (And while I haven’t seen anybody talk about nasal vaccines preventing Long COVID, it appears that SARS-CoV-2 virions get into the central nervous system via the olfactory bulb, i.e. through the nose.)
Second, take heart from Florida. The governor of Florida is actively hostile to anti-COVID measures, and yet their case rate has fallen quite a bit. Partly this might be due to the weather, but it might also be due to so many people having infection-acquired immunity. (Seroprevalence surveys suggest that 35% of Floridans had infection-acquired antibodies by 21 Sep 2021.)
If it’s the weather, then maybe we can alter our indoor climate with better HVAC systems. We should know more as time goes on.
Third, the more people get vaccinated, the fewer non-pharmaceutical measures we need to take to get Reffective down below 1. Maybe we will still need to mask up in grocery stores for three more years, but we won’t need to show proof of vaccine. Once we get the case rate down, perhaps test/track/trace can contain it.
Fourth, we may get to herd immunity at a regional level. If you are reading this, you probably live in BC’s Lower Mainland. (I haven’t enabled any analytics, but I’m guessing that the only people who read this blog are twelve or thirteen friends of mine, and I live in the Lower Mainland.) I don’t have firm numbers, but just by eyeballing plots in the weekly data summary, it looks like VCH’s vax rate is about five percentage points higher than the provincial average and Fraser Health is about two points higher. A vax rate of 88% and a vaccine effectiveness of 93% (which is what a booster should get you, at least initially) will get herd immunity for R0 of 6. A vax rate of 91% will get you almost to R0 of 7.
Fifth, we have new treatments which will dramatically reduce the number of hospitalizations and cut the costs by about half. (See The math for Paxlovid, below.) The province might decide that with the new treatments, it might make sense to go ahead and let COVID-19 spread. That’s not a very kind and gentle, socialist thing to do, but it is possible that the province will decide to do that. (I would be surprised if there aren’t a lot of jurisdictions in the USA which take that approach.)
Even so, a lot depends on people being able to start treatment within five days of symptom start. I can imagine that BC will hire a fleet of delivery people whose only job would be to take prescriptions to with positive tests, dispatched the instant that a positive test appears.
The math on herd immunity
The herd immunity equation says that herd immunity is reached when (1-(1/R0))/VE or more people have been vaccinated with a vaccine of effectiveness VE against transmission of a disease with a reproductive rate of R0.
Vaccine effectiveness studies are never about effectiveness against transmission. I will use effectiveness against symptomatic infection instead. Symptomatic vaccinated people are slightly less likely to infect, but it’s a small effect, and some asymptomatic vaccinated people are infectious. Thus using VE against symptomatic infection as a proxy for VE against transmission is probably reasonably close to reality.
Estimates vary on what the effectiveness of the vaccines are long-term (in part because the delay between first and second dose matters), but in a study in BC and Quebec, the mRNA vaccines seem to settle at about 85%-90% effectiveness against symptomatic infection. (Other studies of other vaccines or in other places generally find the effectiveness to be even lower, so 80% is kind of a best-case.)
COVID Classic had an R0 estimated to be 2.4 to 2.6, while a review from August found five studies estimating R0 for the Delta variant, ranging from 3.5 to 8. And there is no guarantee that it will not go up even further. (Measles, for example, has an R0 somewhere between 12 and 18.)
Here’s what share of the population needs to be vaccinated on the Y axis, vs. vaccine effectiveness against transmission on the X axis, for different values of R0:
British Columbia is currently sitting at about 80% of the population vaccinated with probably an 85-90% vaccine effectiveness against symptomatic infection. As you can see, that’s nowhere near herd immunity even for R0 of 5 (blue line).
But slightly more than 90% of the eligible population of BC has gotten a first dose. Won’t getting them second doses and vaccinating the 5-11 year-olds help? Yes, but not enough. My guess is that BC is going to end up at about 86% vaccinated with two doses within the next six months. As you can see from the chart, that’s just not quite enough with a 90% effective vaccine to get herd immunity for an R0 of 5.0 (blue line).
But won’t boosters help? Yes, they will. Boosters appear to get effectiveness against symptomatic disease up to the 93-94%ish range. That’s just barely good enough to get to herd immunity for an R0 of 5.
In order to get to herd immunity for an R0 of 8 with a vaccine which is 94% effective, we need to vaccinate 93% of the entire population — which includes infants and toddlers — and I just don’t think we’re going to get there.
The math on Paxlovid
- There are about 520K people who are completely unvaccinated, and about 200K people who have been infected. If every previously-infected person is unvaccinated, that’s still 320K people who will, given enough chances, get sick.
- The case fatality rate in BC in 2020 was 1.97%, so that’s 6,304 to 10,244 deaths still to come (versus 2,304 in the past 20 months).
- The case hospitalization rate in uninfected people is about 6.9%, so that’s about 22K to 36K hospitalizations (versus 12,143 in the past 20 months).
- Pfizer’s anti-COVID pill (Paxlovid) is 89% effective at avoiding hospitalization compared to placebo if they take it within three days, and 85% if they take it within five days. That means that the case hospitalization rate would go down to 0.7% within three days and 0.1% within five days. That would still mean 2414 to 5350 hospitalizations if everybody who caught COVID was able to find out and take a pill within 3-5 days.
- The US paid USD$529 (~CAD$670 per course for Paxlovid. A hospital stay for COVID costs CAD$23K. Not giving Paxlovid would thus cost the province about CAD$500M-828M. Giving Paxlovid to 320K to 520K people would cost CAD$169-275M plus CAD$55M-123M for the patients who would still be hospitalized, or CAD$224-398M, so it would cost a bit less than half (and be much much easier on patients and health care workers!). It seems reasonable to think it would also reduce Long COVID, although there are no studies on that yet.
Recommended Reading
This article talks about health care workers with Long COVID, and what a shock it is to them to see how poorly doctors treat patients.
This article talks about how one of the problems with Long COVID is almost definitional, just figuring out what is and is not Long COVID.
This article talks about how something very precious — from an economic standpoint, not a spiritual standpoint — to us has been ebbing during the pandemic: trust.