How safe will you feel when vaccinated?

[Marcie]

@twig93

I think the 1 in 5,000 if you’re vaccinated is total cases because they did 100% testing, right?

Did you read the NYT piece? That’s based on confirmed breakthrough cases. Keep in mind that in many places, vaccinated people are testing less often than unvaccinated. The article even says

@Actuary321

(These numbers don’t include undiagnosed cases, which are often so mild that people do not notice them and do not pass the virus to anyone else.)

(NB: David Leonhardt quietly admitting that asymptomatic spread is largely a myth)

@twig93

I’m not sure it’s reasonable to include the first couple of months in that calculation when most of the country (let’s say most counties in the country if you want a more specific metric) didn’t have a single case.

Fair enough. I did say it was quick, back of the envelope stuff. Instead of your calc, then, let’s try to get even more apples-to-apples. Let’s see what good ol’ Mr Leonhardt based his number on.

@Actuary321

The estimates here are based on statistics from three places that have reported detailed data on COVID infections by vaccination status: Utah; Virginia; and King County, which includes Seattle, in Washington state. All three are consistent with the idea that about 1 in 5,000 vaccinated Americans have tested positive for COVID each day in recent weeks.

“In recent weeks” isn’t very clear, and I didn’t see David’s sources for this data, so I can’t easily check his math, but just to look at total cases over a similar timeframe, I’ll use the Johns Hopkins data tracker (linked in previous post) & look at cases since ~end of July. (Because he doesn’t show his work, this is going to be a bit rough on my part, too)

Utah: as of 9/7/21, total reported cases = 474,086
As of end of July (7/30), total = 432,467
That’s 41,619 cases in 39 days ≈ 1067 per day, in a state of ~3.34 million people ≈ 320 per million per day (1 in 5000 = 200 per million).

Virginia: 9/6: 786,910; 7/30: 694,384
Diff: 92,526 in 38 days ≈ 2435 / day
Population: 8.67 M
Cases/1M /day: 281

King County WA (JHU doesn’t appear to have county-level data, so switching to NY Times tracker
Not as easy to get cumulative numbers as of certain dates from this, so I’m just eyeballing the 7-day moving average, which has been hovering around 600 cases/day for about a month.
Population: 2.30 M
Cases/1M /day: ~261

So in each of these places, the total population number (including breakthroughs) is a little higher than the 1 in 5k Leonhardt estimates. Since each of these places is roughly 2/3 fully vaccinated (give or take - this is where the math gets rougher bc I’m tired), you’re probably looking at roughly ~350-500 cases per million per day for unvaccinated (which may include partially vaxxed, too). So looking at ~1 in 2-3k per day, or about a 1/5 - 1/8 annual chance of catching covid unjabbed vs ~1/13 for jabbed.

That’s a bit higher, but not what I would call “far higher.” Factor in the big bump in testing unvaccinated kids due to back to school & absurd quarantine rules, and the difference gets a little muddier.

[Lucy]

@Marcie

Did you read the NYT piece? That’s based on confirmed breakthrough cases. Keep in mind that in many places, vaccinated people are testing less often than unvaccinated. The article even says

In my experience, the people who are most concerned about covid are testing the most. Some of my friends are getting tested before and after every social thing they do. (twice after, typically) And hey, they are also vaccinated. I don’t have any stats about the relative testing rate of vaccinated and unvaccinated people, but your claim doesn’t pass the smell test.

@Marcie

(NB: David Leonhardt quietly admitting that asymptomatic spread is largely a myth)

While asymptomatic spread is rare, presymptomatic spread is routine – in fact, it likely accounts for the vast majority of spread, since the disease is most infectious a couple of days before it becomes symptomatic, and rapidly becomes less infectious after symptoms appear.

@Marcie

So in each of these places, the total population number ( including breakthroughs) is a little higher than the 1 in 5k Leonhardt estimates.

I haven’t looked at those estimates. But I’ve looked at break-thru and total cases in my state in the last 6 weeks, and my best analysis indicates that the vaccines are still highly effective. This is backed by observing that while infection rates are rising in my state, death rates have barely budged. There’s an element of “death is a lagging indicator”, but the difference is too large and too persistent for that to explain it. “Old people who are likely to die are mostly vaccinated” is almost certainly most of what’s going on.

Kids are unlikely to die even if they aren’t vaccinated, but the growing studies on persistent brain damage after covid would worry me as a parent. If I lose a few IQ points, I suppose I could retire. If a kid does, it reduces their outlook for a long time. Not to mention the heart damage, blood clots, kidney damage, and lung damage that covid can cause. I’m sure all of those are less likely in kids than in older people. But how much less likely?

[Marcie]

@Lucy

There’s an element of “death is a lagging indicator”, but the difference is too large and too persistent for that to explain it. “Old people who are likely to die are mostly vaccinated” is almost certainly most of what’s going on.

Agreed. I believe Israel (and UK?) data also agreed with this. The degree to which older people who are more likely to die are also more likely to be vaccinated, along with the degree to which testing has bumped up more among younger people with the start of school + the likelihood of cases among students & teachers to be mild, make looking at the data pretty thorny. I agree with your conclusions here, but while the shots are still performing well, I think there’s a sense that they were oversold - outside of the very healthy trial population, efficacy is not as high as the 95% seen in the trials, and the shots do not block infection or transmission as was implied in a lot of the early messaging. You might blame Delta for this, but it’s virology 101 that viruses mutate. Failure to plan & all…

@Lucy

the growing studies on persistent brain damage after covid

Can you link these? I’d be curious to look at them. I remember seeing one a while back that had n = ~29 (?) that were solicited from far & wide, with little in the way of controls (pre-covid baseline, other kids), so it seemed premature to infer any strong signal from it.

@Lucy

Not to mention the heart damage, blood clots, kidney damage, and lung damage that covid can cause. I’m sure all of those are less likely in kids than in older people. But how much less likely?

Great question. A similar question is: these are all adverse events that the shots can cause, too. I’m sure those events are rare in children, but how rare? Are they more or less rare in us kids than they are in adults? Are they rarer with the shots or with covid?

[Lucy]

That’s why there are pediatric trials – to make sure the cost/benefit is worth it for kids. So we don’t yet know whether side effects from vaccines are more or less risky than covid for kids under 12. But we do know that side effects from vaccines in kids 12-18 are less common than the same problems from covid. That is, one of the most common side effects in young people who get mRNA vaccines is inflammation of the heart – but that’s MORE common among unvaccinated kids who are exposed to covid than it is to vaccinated kids.

I’ll like the brain damage studies I’ve seen in a minute.

[Lucy]

lower IQ among those infected
Cognitive deficits in people who have recovered from COVID-19 - EClinicalMedicine (thelancet.com)

Here, we analysed data from 81,337 individuals who completed the full extended questionnaire in order to test the hypothesis that those who had recovered from COVID-19 would show objective cognitive deficits when performing tests of attention, working memory, problem solving and emotional processing. We also determined whether the extent and/or nature of cognitive deficit related to severity of respiratory symptoms as gauged by level of medical assistance, positive verification of infection via a biological test, or time since illness onset.

Note that solicitation for the study didn’t mention covid, and the “did you have covid” questions were asked after the IQ part had been completed. (I think the researchers piggy-backed on to a big IQ study being done for other reasons.) There was a clear dose response: those ventilated showed a larger decline than those just hospitalized, who showed a larger decline than those who tested positive but stayed home, who showed a larger decline than those who asserted they had covid but hadn’t had a positive PCR test.

all results were looked at both raw and normed for a variety of demographic factors, and the results were slightly more significant with norming.

The cognitive test wasn’t technically an IQ test, and differences among the groups are shown as fractions of a standard deviation away from the uninfected norm. Results showed:

There was a significant main effect (F(5,81,331) = 9.6867 p = 2.915e-09), with increasing degrees of cognitive underperformance relative to controls dependent on level of medical assistance received for COVID-19 respiratory symptoms (Fig. 2a-Table S4). People who had been hospitalised showed substantial scaled global performance deficits dependent on whether they were (−0.47 standard deviations (SDs) N = 44) vs. were not (−0.26 SDs N = 148) put onto a ventilator. Those who remained at home (i.e., without inpatient support) showed small statistically significant global performance deficits (assisted at home for respiratory difficulty −0.13 SD N = 173; no medical assistance but respiratory difficulty −0.07 SDs N = 3,386; ill without respiratory difficulty −0.04 SDs N = 8,938).

Now, you might say “only 44 people were on respirators”, but nearly 9000 were mildly ill, and the dose-response is striking. And they claim statistical significance of all those results.

Reduced Brain Volume in images of the brain as compared to same person pre-infection:
Brain imaging before and after COVID-19 in UK Biobank | medRxiv

This one is not yet peer reviewed, but the methodology looks pretty good to me:

UK Biobank scanned over 40,000 participants before the start of the COVID-19 pandemic, making it possible in 2021 to invite back hundreds of previously-imaged participants for a second imaging visit. Here, we studied the possible brain changes associated with the coronavirus infection using multimodal MRI data from 785 adult participants (aged 51–81) from the UK Biobank COVID-19 re-imaging study, including 401 adult participants who tested positive for SARS-CoV-2 infection between their two scans. We used structural, diffusion and functional brain scans from before and after infection, to compare longitudinal changes between these 401 SARS-CoV-2 cases and 384 controls who had either tested negative to rapid antibody testing or had no COVID-19 medical and public health record, and who were matched to the cases for age, sex, ethnicity and interval between scans…[lots of technical stuff about which parts of the brain were smaller in people who had been infected]
…
Unlike in post hoc cross-sectional studies, the availability of pre- infection imaging data mitigates to some extent the issue of pre-existing risk factors or clinical conditions being misinterpreted as disease effects. We were therefore able to demonstrate that the regions of the brain that showed longitudinal differences post-infection did not already show any difference between (future) cases and controls in their initial, pre-infection scans.

About 400 each infected and controls.

[Lucy]

In general, you can follow
Covid – effects other than mortality - The Sandbox / Covid19 - GoActuary

[magillaG]

@Marcie

I think there’s a sense that they were oversold - outside of the very healthy trial population, efficacy is not as high as the 95% seen in the trials, and the shots do not block infection or transmission as was implied in a lot of the early messaging. You might blame Delta for this, but it’s virology 101 that viruses mutate.

I remember experts being very careful to caution that efficacy might not be as high in reality as in trials. I noticed this because it is common that drugs usually do not perform as well in reality as in trial. Possible mutations were specifically cited as one reason to be cautious.

Remember the CDC got a lot criticism because it told vaccinated people to continue masking.

The CDC did report efficacy reports on the performance of the vaccine “in the field”. After that also showed the vaccines to be very effective, then they allowed people to stop masking. Which also got criticism.

The CDC has not been perfect. But i think a lot of the criticism has been unfair. We are in a situation where the science is being learned as we go along. To make recommendations, the CDC has to balance trying to estimate the “probabilities” of what is true about the virus, which should have some objective truth, and then combine with with the utility of different actions, which will be different for everybody.

Also, I’m not sure this 1-in-5000 statistic is very helpful. I think the actual probability of getting covid would be highly variable. I care about my probability, given what I am doing. As an analogy, something like 10,000 people die a year from drunk driving, which is a small fraction of the population. This isn’t a useful statistic for me to decide whether to drive drunk.

[NerdAlert]

My city is reinstating the mask mandate in a few days. Commence with the freakout.

[ao_fan]

even NYC isn’t putting in a mask mandate other than for a few specific things like public transit.

I mean, unvaccinated people aren’t allowed to participate in society at all, but no mask mandate! yay!

[Marcie]

Are they going to require at least surgical/medical masks, per the recent terribly flawed Bangladesh mask study? Limited to just red/purple or blue/green?

[Lucy]

Ooh, do the fun-colored masks work better? I just got a shipment of fun-patterned KF94 masks.

[Marcie]

It was something they measured.

To be fair, I haven’t really looked at this part of the paper, but I strongly suspect these effects have to do with increased mask usage, not effects of different colored masks on covid seroprevalence. Also, note the size of those confidence interval bars.

[Mountainhawk]

Salem MA has an a mask mandate for a few weeks. Goes through November 13th at a minimum.

[Marcie]

Hope I’m wrong, Ms Hawk, but I’m going to predict it does little to nothing to stop or even dull the fall/winter seasonal wave that’ll peak around then.

[Maphistos_Sidekick]

Salem MA does tend to get a bit crowded in October.

[Mountainhawk]

It’s more for the Halloween tourists. Salem is more than 80% vaxxed of eligible population and more than 70% of total population, so I imagine we’ll be in pretty good shape once the young kids can get vaxxed — probably heading towards 90% by the end of the year.

[The_President]

@Marcie

Hope I’m wrong,

I know you are looking for that irrefutable evidence that you are wrong, but it’s still a pretty cheap I hope im right precaution even if you want to ignore the anecdotes that don’t qualify as the real study you wish could happen.

[Mountainhawk]

There actually was a randomized study in rural Bangladesh that showed masks worked:

Randomized study of mask-wearing illustrates their importance in preventing...

[JFG]

Oh, Marcie will be along shortly to shred it.

[Dara]

Too late.

@Marcie

terribly flawed Bangladesh mask study