As reminders…

Alpha–Variant first identified in the UK

Beta–Variant first identified in South Africa

Gamma–Variant first identified in Brazil

Delta–Variant first identified in India

Also as a reminder:

–That’s right! We’ve got a new outbreak to track. A farmer in a rural part of Guinea presented on August 1 with fever, muscle pain, head and stomach ache along with bleeding gums for about the previous week or so. He died on August 3rd, and blood samples obtained after he was admitted to the hospital were positive for Marburg Virus.

Marburg, as you no doubt recall, is a cousin of Ebola, although generally has a lower mortality rate. That’s relative, of course, as mortality from Marburg has ranged from 50-80% depending on the outbreak, but generally trending towards the 50% end. Marburg is actually a little more rare than Ebola. Outbreaks have been -very- sporadic–the last I can think of was two utterly unrelated travelers in 2008 who had spent a lot of time around bats. Marburg was also the first of the filovirdae (Ebola’s family) to be identified, with the first cases definitely identified way back in 1967.

Because it is encountered less commonly than Ebola, there has been less emphasis on Marburg specific treatments. Unlike Ebola, there is no vaccine to Marburg, and no specific therapy. Some of the antibody treatments available for Ebola may cross cover Marburg, but there have been no trials done to prove this yet. Mostly because it’s so hard to find Marburg “in the wild”.

Like Ebola, Marburg is relatively tough to catch, with close contact, especially with body fluids, the best way to transmit the virus. There 173 known contacts being observed in Guinea with, so far, no new confirmed cases. The only contact they have lost to follow up is the farmers wife. So aside from losing the closest of contacts, tracing is actually going quite well so far. This will be followed for a few weeks to see if any new cases pop up.

–We now return you to your regularly scheduled slow motion coronavirus disaster…

–So in the US headlines this week is the ICU pressure in multiple cities and states from the delta wave. Far more of them, in fact, than I would have expected–especially given the lower hospitalization rate seen in CDC’s data from last week. ICU cases, and hospitalizations from COVID in general, remain overwhelmingly (95%+) unvaccinated patients who, I would wager, are catching SARS-CoV-2 for the first time. We’ll get to data on why that wager a little later. In general, hospitalizations and ICUs are trending with relative vaccination status. Your city or state is at higher risk to be near ICU capacity the less vaccinated it is.

That said, that correlation is NOT 100%. For example, less well nationally discussed is Hawai’i. Hawai’i, an island, has a very strict COVID testing policy to get in, generally more restrictions than many other states, and a good vaccination rate. Their numbers have followed the -exact- same trend (per capita) as Texas, and Hawai’i news stations are reporting some hospitals are at ICU capacity and curfews are being debated.

This is also starting to trickle over into “bed’s taken” situations. For example, there was a patient in the news in Texas who had been shot multiple times (but is stable) who was still waiting for surgery nearly a week later. Some places have also cancelled or delayed elective surgeries.

–So my optimism bias has bit me firmly in the ass again. I still keep expecting (hoping?) that there are enough recovered people out there to resist that kind of “bed’s taken” pressure on their local hospitals. At least for the delta wave, I have over-estimated how many people are now resistant, especially among the unvaccinated out there.

Again, all predictions wrong or your money back here in this newsletter.

–Now, the BIG reason I am off on the number of places seeing “bed’s taken” pressure was a surprise to me. I know that burnout among medical personnel has been high in this past year, and there have been people who left the field during that span.

And God, do I understand, fellow healthcare workers. Especially as the dial everywhere remains cranked to maximum stupid, and headlines just keep finding new ways to be depressing.

But a big reason for ICU pressure, and why you are seeing occasional reports of National Guard call outs to help out, is lack of staffing. This is particularly acute among nursing staff, especially for units like the ICU, that have to keep patients covered per nurse low.

I don’t know if that is a knock-on effect of vaccine hesitancy; among the professions, nurses and PhDs have been the most resistant to vaccination thus far (nurses are about 50% vaccinated; DOs and MDs are well north of 90%). I can’t imagine that they are staffing COVID ICU patients with non-vaccinated doctors and nurses, and the pool of available nursing staff may be low because of vaccine status.

I stress “could be” low because of vaccination status. Again, I don’t know for sure how much of a driver vaccine hesitancy among nurses in general is driving staffing issues, and limiting the total number of beds, or ability to flex expand beds, as demand increases. The other part really is a lot of people leaving, and/or few coming into nursing or other medical professions right now. I know the “float pool” for back up medical assistants, nurses etc. for my wife’s clinic is basically non-existent at the moment too, however anecdotal that may be.

At the start of the delta I wave, I “back-of-envelope” estimated that relatively few places would see “bed’s taken” pressure–but assumed that we were closer to fully staffed than we actually -are- in many places. The staffing issue seems unlikely to relent near term, at least not without a decline in new COVID cases (unlikely for a few weeks at least–epiforecasts has the US hovering around 1-1.1 nationally), and/or dragooning additional help by diverting other health care specialties/clinics/wards to increase available staff.

–As an update to last week, the rumor is that Mount Vernon schools, where Mount Vernon Doctor whose comments we addressed spoke, has gone to all virtual due to the large number of absences and necessary quarantines among students this past week.

–Also worth mentioning that restrictions not correlating with control of the virus, particularly the more contagious delta variant, are not unique to the US. Australia is still finding new cases, despite shockingly authoritarian measures that have seen videos of kids being maced for not wearing masks and Australian health authorities trying to ban speaking to other people. New Zealand locked down for one!!!! new positive on the island–and has still promptly found 10 more. China continues to be relatively shut down (we’ll get to supply chain effects of that later).

–In other news, the US mentioned that they would be offering booster shots starting at 8 months from the initial dose. The rollout is expected to follow the rollout pattern of the first shots, with high risk groups (elderly patients, immunocompromised patients) and healthcare workers (including nursing home workers, who just got a vaccine mandate) first.

A very popular reader question this week has been “what do you think about the booster shot?”

Well, the CDC has promised forthcoming data to help explain their rationale. I look forward to that data. We did mention a couple papers last week that we said could provide rationale for something like this. First, the Mesa County Arizona paper showed that breakthrough hospitalizations (which I consider the medically relevant metric) were more likely in older vaccinated patients than younger.

This could be for either of two reasons, or a combination of the two. First, older patients got the vaccine first. Protection may indeed be ebbing over time. However, we also know that older patients have more co-morbidities and a less responsive immune system over time. They may simply be more susceptible even with a vaccination, and especially if the vaccine only trains the immune system to be “close enough”, like with delta. There may still be some catch up in the immune system of a vaccinated patient, as the antibodies have to be adjusted to the small differences in the delta variant by your immune system. We mentioned this “catch up” period explains the occasional symptomatic breakthrough infection, and those symptoms may be severe enough in an already susceptible (elderly, co-morbid) population to put them in the hospital.

We also covered a petri dish experiment in France that showed antibodies in serum obtained from patients who got COVID a year ago were only able to neutralize delta 47% of the time. But, if those recovered patients got a vaccine in the past year, they neutralized delta 100% of the time. This suggests that COVID vaccination protection wanes over time with respect to delta–but does not prove it, as in petri dishes, the immune system has no chance to “catch up” and tinker the antibodies present to fight delta better like it does when you are actually exposed to delta. That the breakthrough rate is so low, counting only symptomatic positives with a vaccine, suggests that in vivo, when exposed to delta, the “live” immune system does far better than the 47% petri dishes suggest. For example, even though poorly interpreted by the authors, the Mayo Clinic study last week showed that the Moderna and Pfizer vaccines were 85% or better effective in preventing hospitalization with delta.

I also mentioned that I, personally, am 8 months status post my Pfizer vaccine. I have had no symptoms of COVID to suggest breakthrough in that time (and have not been tested by PCR for SARS-CoV-2 either). My antibody level for the anti-spike protein antibody generated by that vaccine remains super high. We have to dilute my sample 5 fold just to accurately measure how much of the antibody is there, since the assay was designed to expect a smaller amount of antibody present. That said, my personal results are -purely- anecdotal and an N of 1, so don’t read too much into it.

This week, we have a pre-print publication out of Oxford looking at vaccine effectiveness against delta in the UK. They looked at finely detailed data from 384,543 patients who had PCR tests during the alpha wave (December 2020 to May 2021) in the UK, with 16,538 positives, and 358,983 patients with PCR tests during the UK’s delta wave (17 May to 1 Aug 2021), with 3,123 positives. They have age, sex, if the patients were in the same household, location in the UK…

…what they don’t have, inexplicably, is hospitalization rate and outcome of those positives. Anywhere in here. I can’t find it in their data.

So everything on vaccine efficacy here is related to the ability of the vaccines to prevent a subsequent positive PCR test.

They call a PCR positive result for SARS-CoV-2, regardless if symptomatic or asymptomatic, a breakthrough “infection.”

Again, and I cannot stress this enough, a positive PCR result, absent symptoms, just proves the virus is there. Could be dead. Could be live. If asymptomatic and stays asymptomatic, prior studies have shown VERY likely dead and unable to infect anyone else.

PCR positive plus SYMPTOMS CONSISTENT WITH COVID is the definition of a CLINICALLY POSITIVE CASE, or “infection”, and -all- of the PCR tests for COVID I am aware of define a positive this way in the “information for patients and healthcare providers” we are required to provide with the test. You can look all of those bulletins up for all assays for SARS-CoV-2 approved or EUA approved by the FDA on their website.

I do not have a good explanation for why big name places ranging from Mayo to Oxford now continue to treat asymptomatic, screen positive PCR cases as a clinical positive. An “infection.” And not just “presence of the virus’ RNA”, which is all the PCR test really says at that point.

This failure to properly define a clinical positive grossly underestimates the clinical effectiveness of the vaccines, and is being used to drive misinformation about the vaccines themselves. Yet it keeps happening in the literature. I don’t get it. I really don’t.

Now, the Oxford guys included data on symptomatic versus non-symptomatic patients. As we have discussed before, and as I am sure you are aware by now, most patients with SARS-CoV-2 who will get symptoms get them in 3-5 days. Rare patients will get symptoms out to 14 days. That’s why the close contact quarantines are 14 days.

In Figure S5 in their data, where they present the rate of symptoms reported for patients with vaccines (1 dose or 2) and previously infected patients with a new positive PCR, they counted first symptoms out to 35 days after the PCR positive.

So think about that for a second. You have a virus where >90% of patients get symptoms, if they are ever going to, within 3-5 days of exposure, and all of them will within 14 days. But you are counting symptoms 2.5 times the incubation period as caused by that virus? You could literally have gotten a PCR positive screen in the UK, finished out your 14 day quarantine, gone back to your regular life, been fine on day 33, fine on day 34, got a cough on day 35.

And the authors would have counted that as symptom positive, PCR positive SARS-CoV-2.

Makes zero sense to me. So when they show that more patients in the delta wave are likely to report symptoms with a positive PCR result (a little over 50% overall versus right around 50% for the alpha wave), I don’t know what to make of that. Hospitalization data might have helped make some sense of it, but again, they, for reasons????, included every clinical detail BUT that in this paper.

However, I can tell you, even with THAT generous a definition for “COVID symptoms”, their data shows that patients who had caught COVID already, got better, and then had a second positive test appear less likely than the vaccinated to show symptoms on that second positive. But they had fewer of these in their data, so the 95% confidence interval is huge, and it may be no different than the vaccinated.

They show data in Table 1, however, that pretty conclusively shows there is NO difference in likelihood of subsequent positive PCR result between patients who got the vaccine and patients who had recovered from a previous COVID bout.

Yes, Hypothetical Reader, that does call into question vaccination mandates that do not ALSO include patients who previously recovered from COVID–as this data supports other studies, like the Israelis, that suggests natural immunity from prior infection is just as effective as the vaccine. Also think of that when New York restaurants and businesses are now required to exclude the vaccine hesitant (again, minority enriched)–even if they already had COVID and recovered, or have a medical reason to not be vaccinated.

Now, what they do have, that is useful to our discussion here, is spike protein antibody level prior to the positive PCR test. The median of that level was 30 days before the new PCR positive for SARS-CoV-2, so keep in mind, the antibody level may have changed (and dramatically for some patients if the level was drawn months before). But at least we have antibody level data for vaccinated patients to correlate with subsequent PCR positive results!

In Figure 4, they show conclusively that there is NO difference in viral load (the Ct value on the positive PCR test) and months since the vaccine, NO difference in viral load with age, and NO difference in viral load with spike protein antibody level.

What that total lack of correlation suggests to me is that there is NO difference in the likelihood of exposure to SARS-CoV-2 based on your antibody level. Thus, there is no antibody level line in the sand that can be drawn if your goal is to stop positive PCR tests.

But our goal is to stop hospitalizations from COVID. Their data are non-informative on that. They show a slight correlation to Ct value and likelihood of symptoms, but their definition of symptoms is so broad as to be clinically useless. They also show a lower Ct (thus higher viral load) for delta, but that is also to be expected of a more contagious variant–each exposure event “should” be transferring more viral particles.

So what does all of that mean? “Jeebus, doc, is there an answer to ‘what about the booster shot’ somewhere in here?” I hear you ask Hypothetical Reader.

My struggle is this. There is an awful lot of attention being paid to positive PCR results in vaccinated patients. We are counting these as clinically positive, even when many of them are asymptomatic, and NOT really a clinical positive at all. Further, we know that 95%+ of the hospitalizations are in the unvaccinated getting COVID for the first time. The vaccines ARE reducing hospitalizations and deaths among the vaccinated by 10 fold at least versus the unvaccinated. We know that antibody levels do NOT predict if your PCR test will be positive or negative. There might be an antibody level that predicts your risk of hospitalization with subsequent COVID. I have not seen anyone show data on that yet though. This would be useful, as we already use levels like that to help time boosters for other vaccines. And we could use it here to determine who needs a booster and when more precisely.

Right now, though, we have some evidence that at least mild symptoms are more common with the delta variant in vaccinated patients than previous strains. We have evidence that antibody levels decrease over time from vaccine, but again, we don’t know what that means in terms of stopping hospitalization.

The single best argument we have for boosters is the Mesa County, Arizona data–but that only argues for boosters for the elderly, who were already at higher risk. Even there, it’s not clear if that’s because they are at higher risk already, or if, because they were, on average, vaccinated earlier that vaccine efficacy is waning in everyone.

So to answer your question, Hypothetical Reader, if you are already in at risk group for actual COVID mortality (elderly, obese, the other major risk factors), a booster makes sense.

If you are not… well, that “forthcoming” CDC data had better be pretty compelling. Because so far, all I have is a bunch of papers that keep counting exposure like it’s a clinical case, without showing a very compelling difference in the hospitalization rates ESPECIALLY for those without pre-existing COVID risk factors.

In short, I am not running out to get a booster at my age and general health.

But don’t just take it from me! Business Insider got three other doctors to give quotes in an article titled “3 leading COVID-19 experts say there isn’t clear evidence that healthy, vaccinated people will need booster shots 8 months out.” A couple quotes from that article. Dr. Robert Atmar, running a booster trial at Baylor, said “”Will it keep more people out of the hospital? Maybe, but I don’t know that… Targeting the unvaccinated would have a greater effect, from a public-health standpoint, if those individuals could be persuaded to accept the vaccine.”

Also from the article:

“Dr. Paul Offit, a pediatrician at the Children’s Hospital of Philadelphia and coinventor of the rotavirus vaccine, said the goal of these boosters should be the same as any vaccination: to eliminate ‘the worst things the virus can do.’

Offit, like Moore and Atmar, said that aim would be better achieved by first vaccinating more people who haven’t got their first dose, rather than bolstering protection for those who have.”

–So it’s also not just me saying that the medical end is when you can keep people out of the hospital! At least keep enough of them out that the healthcare system is not at risk.

–Keeping that healthcare system staffed to keep capacity high was not a problem I had anticipated until the last couple weeks though…

–And I am still not entirely convinced that political leaders, ranging from the local to the state level to the national, have a clear idea of what the end looks like. Is it an end to positive tests (unlikely, at least for a long time yet)? A hospitalization rate among positives below a certain level? If so, what is that?

I am as confused as you are over what the public policy goals are for things like mandates, masks etc. “The science” is always claimed, but I think as we have illustrated these past couple updates, the headlines you see (and appear to drive policy) are not always supported by the data in the actual paper. In either direction, pro or against. Misinterpretation abounds. Half-baked studies, or studies missing key parts for actual clinical relevance abound.

–I am going to mention as an aside, there is a good mutation analysis study (that still manages to overstate its conclusions from its evidence in the discussion) out of Maryland showing that high vaccination rates, as you would expect, lead to fewer new variants in the vaccinated population. Vaccination is reducing the chances the virus gets to mutate by taking away hosts, and thus the error rate in viral replication that creates the variants to begin with. I know the idea has circulated in some corners of the internet that the vaccine is causing variants by selection pressure; the data says no.

–Finally, another op-ed from a virologist suggesting that the most likely origin for SARS-CoV-2 is a virus that was being studied in the Wuhan Virology lab that escaped. The best argument against this is the sequencing data our man from the Hutch found as a “ghost” on Google servers after China had somehow gotten it deleted, even from repositories of data for publication (since it was used in a couple early publications from China). That data suggests the virus was spreading outside of Wuhan before some of the reports of workers at the lab with symptoms and before the big open market superspreader event in China in late 2019. That said, still possible it was spreading elsewhere in China because it got out of that lab, just earlier than we might think.

Again, certainly possible. And again, our grandchildren might find out the truth. We won’t, not for many years.

–Which brings us to socioeconomics in the time of coronavirus…

The Pax Americana ended in Afghanistan this past week.

There’s a gal at my jiu jitsu gym who has been deployed to Afghanistan for the past several months, and got leave to come back a few weeks ago. Turns out it was quicker to come back to Indiana and get a COVID vaccine than to get a vaccination to our troops in Afghanistan. She was one of thousands of Americans trapped behind enemy lines by the Taliban advance, and her small group was fortunate enough to negotiate their way to the Kabul airport with their Taliban “keepers” to catch a plane out mid-week. There are still thousands of Americans and allies behind the line–to say nothing of good Afghanis who fought alongside us, and who trusted our sergeants and soldiers that we asked to look them in the eye and pledge the loyalty and assistance of the United States of America.

To know what went wrong in Afghanistan, you need only read Hackworth’s book “About Face.” You won’t believe me when you read the summary of that book, but you’ll also read the “Afghanistan Papers” published by the Washington Post in 2019, and you’ll understand that Hackworth was describing Afghanistan too.

20 years worth of military and political leadership, spanning periods of control by both parties, was utterly unworthy of the time, effort and sacrifice of those who served on the ground over there.

Again.

And the Bonfire of the Institutional Credibilities grows ever higher.

With the end of the Pax Americana, your local region in the world will return to its regularly scheduled history.

What that looks like, in a lot of places, will not make for happier headlines. For example, if I were Taiwan, I would be having some very serious conversations with Japan, India and the Philippines. Probably Indonesia too.

It’s clear that the US will be spending much of this decade sorting itself back out, and deciding on a direction. A lot more is possible now, both here and abroad, than you might think.

The end of the Pax Americana also complicates global supply lines. Globalization is only possible because the US, in essence, guaranteed free right of the sea for everyone’s goods by the might of the US Navy and by being the buyer of last resort as the world’s reserve currency. The catch was you had to use the dollar, and agree to open your markets. Again, as that now changes, more is possible than you think. It’s no accident that Palantir, a company built on AI analysis of huge databases (used by governments, big corporations, pharmaceutical companies, banks etc.) is now accepting Bitcoin and just quietly announced it had bought $50 million in gold.

Those are hedges for currency instability.

You couple geopolitically unsettled supply lines and higher geopolitical risk with still nearly random lockdowns and shutdowns and snarls in the supply chain, and “oddities” will start to appear. Inflation, from all the spending by -everyone- around the world to prop up shutdown economies continues to drive up food prices. As well as prices of everything, but food is non-trivial.

Supply disruption and lack of staff in transport, production and processing of food is already affecting food availability. Go Google “food shortage”, click to news, and set for just articles from the past week under tools. See what you find.

Droughts in the wheat belts of Russia have dropped global wheat stocks to 5 year lows. The upper Midwest of the US, another major breadbasket, has had sustained above average heat and less rain, with the USDA warning of lower crop yields. Tyson’s CEO at last earning call expected price hikes of up to 35% to be passed on to consumers. Biden just announced a 25% increase in food stamp allowance.

These things are not in isolation.

Deus impeditio esuritori nullus.

Any wonder retail sales looked abysmal and consumer confidence is cratering in the reports out earlier this week?

Again, if you are fortunate to have the means, best to look locally and consider a donation to your area food bank.

–Closely related thought piece this week… A history lesson from twitter:

Germany in 1920 pic.twitter.com/ChWirk9J5n

— Jared Dillian (@dailydirtnap) August 12, 2021

–Your chances of catching coronavirus most places in the world this week are equivalent to the chances I’m about to blow your mind with one re-posted tweet:

–Your chances of catching Marburg are equivalent to the chances that having linked one Pink Floyd classic already that I’m going to be able to avoid linking another…

<Paladin>