|I saw Richard Burton play Hamlet on Broadway!
Lots on masks, treatments, vaccines, and guidelines, my take on hospitalized with versus hospitalized for, and detours about what's going on in Italy and Denmark.
Treatment and prevention news
|Indians lined up to refill oxygen canisters.
Face masks 1: Finally, some real data on whether face masks protect the person who wears one. First, an international study found that mask mandates lowered death rates. Then a case-control study by the CDC, which found protection against SARS-CoV-2 infection to be minimal for cloth masks, 66% for surgical masks, and 83% for N95 or KN95 masks. Despite methodological weaknesses, I think this study disproves the earlier theory that masks keep infected wearers from spreading the virus but don’t protect uninfected wearers. Importantly, all the subjects said they wore masks consistently whenever they were in indoor public spaces.
Face masks 2: If COVID-19 is raging (97% of US counties are currently high-transmission areas), and if good masks protect, then everybody should be wearing them in indoor public spaces. For once the CDC and I agree. But states – especially blue states – are instead falling all over each other to revoke mask mandates. Since my husband and I will be visiting two of those states in April, California and New York, I’m scared. Of course you can only leave your mask off if you’re vaccinated. Yeah. Anyone who believes unvaccinated people are going to “keep” masking up, please raise your hand.
Bye bye lockdowns? Three economists made a huge splash on right-wing American media with a "working paper"claiming to prove that lockdowns did no good whatsoever. It gets widely described as a “Johns Hopkins study” though only one of the authors, who has no epidemiology or public health expertise, is from JHU. Are the trio right? Nope. Start with their definition of a lockdown: “The imposition of at least one compulsory, non-pharmaceutical intervention.” So school closures and international travel bans, whose benefits are indeed questionable, are just as much lockdowns as are stay-at-home orders? And it’s only downhill from there, as shown in excruciating detail by the research methodology expert F. Perry Wilson and others, and others, and still others. The paper depends on a single article whose findings were actually so contradictory to the “working paper’s” thesis that its author disavowed the citation.
Natural immunity: According to analyses in Qatar, having been infected with SARS-CoV-2 an average of 10 months earlier protected only 56% against reinfection with Omicron, as compared to 90% for previous variants. But that figure puts previous infection at least on a par with 2-dose vaccination. And protection against severe Omicron disease was an excellent 88%. But there’s zero evidence so far that simply having mild Omicron will protect you against other strains even short-term, much less forever.
Evusheld: This monoclonal antibody combo from AstraZeneca is being marketed as preventive treatment for immunosuppressed individuals, a concept I’ve long advocated. A single dose of RegenCov gave 81.6% protectionagainst COVID-19 for up to 8 months, but at this point an anti-Omicron product such as Evusheld or sotrovimab would obviously be a wiser choice.
Sotrovimab: Now this monoclonal antibody has been shown effective by intramuscular injection, much more convenient than intravenous infusion. But twice as many patients required hospitalization with intramuscular administration, even if the formal criteria the researchers laid out for “noninferiority” were not met. Me, I’d go for the IV version.
Bebtelovimab: This new single-dose intravenous monoclonal antibody product is Lilly’s latest offering. It’s predicted to be active against Omicron and the BA.2 subvariant. It’s already scored an Emergency Use Authorization for use in high-risk patients with early disease, and half a million doses or so should soon be available in the USA. But I the supportive evidence described in the FDA’s press release is strangely underwhelming. A placebo-controlled study that showed briefer illness was only in low-risk patients, not the high-risk patients for whom the drug is authorized. Neither of two trials in high-risk patients had placebo controls, and all the FDA can say is that the rates of bad outcomes were “Generally lower than the placebo rate reported in prior trials of other monoclonal antibodies in high risk patients.” It links to a clinical trial with no results posted. Very weird.
Paxlovid: At last Pfizer’s magic pill is hitting the street. But Italy is expecting only 600,000 courses of treatment in 2022, 1 for every 100 Italians. About 700,000 Italians over 60 had COVID-19 last month. Taking that very roughly as the number eligible for Paxlovid treatment (some older people will be too healthy to be eligible and some younger people will be eligible because of comorbidities), 600,000 is a drop in the bucket. There will be one course for every 16 Americans, and one for every 24 Brits, much more adequate. The New York Times ran a story describing the hoops one of its reporters jumped through to obtain Pavlovid for her mother. The tale was touching, and the reporter’s efforts impressive, but I confess some ethical qualms. The mother was already well protected by a vaccine booster, so one could argue the treatment would have been better given to someone else…
…a propos of who else, the head of Pfizer is now saying that not only will he allow generic companies to manufacture Paxlovid, but that Pfizer itself will provide the pill to low-income countries “at cost.” Since we know from other sources that manufacturing costs are estimated at $20 for a 5-day course, this is fantastic news.
…in Italy as of February 10th a grand total of 41 patients had received Paxlovid. Here the patient must go to a hospitalin person to be evaluated by an infectious disease specialist and be given the pills. For medically advanced countries this seems nuts: patient contacts physician, doc calls in prescription, family member picks up pills. Though in the mess that passes for an American health system even that is iffy. But a deep dive into the reasoning of the Italian authorities reveals a different issue: they don’t trust General Practitioners to know whether a patient’s disease or medication regime is incompatible with the pill. And from what I know of that category of docs, they’re right.
Corticosteroids: Italian madness: steroids, which are known to make outpatients worse, are now officially recommended for outpatients if they are sick enough to need hospitalization but no beds are available. Remember those families in India lined up to refill oxygen cylinders for their relatives? What an admission of defeat.
Remdesivir: After multiple well-conducted trials finding virtually no benefit in hospitalized patients, and with the World Health Organization recommending against its use, now a pair of studies have according to the NEJM’s highly-respected Journal Watch, provided “strong evidence that adding remdesivir to standard care improves outcomes among hospitalized adult patients with COVID-19.” I was shocked to look at the articles themselves. The first, a randomized open-label study from Canada, found no significant effect on mortality even though the comparison patients were slightly healthier. The second, an Italian trial that did report decreased mortality, wasn’t even randomized much less placebo-controlled. It seems absurd to me that these researchers didn’t bother to do a proper trial, and that Journal Watch’s commentator would consider their results convincing.
|Johnson & John's Leiden plant, no longer making COVID-19 vaccine
Novavax: The company finally requested Emergency Use Authorization from the FDA on February 1st, 3 months after beginning the process and a year after their first trial results came in. I think hoping this vaccine’s old-fashioned technology will persuade American novaxxers to repent is pie in the sky. And, of course, the trials are long outdated by now.
Vaccinating survivors: A giant Israeli study has shown that Pfizer vaccination cut the risk of reinfection among patients who had recovered from COVID-19 by 76% overall, though people over 65 did worse. A single dose of vaccine was just as effective as a double one, suggesting the Italian policy of giving COVID-19 survivors only one shot makes at least as much sense as the American approach. And a UK study among regularly-tested healthcare workers showed vaccination left survivors more than 90% protected against reinfection, with no waning over a full year.
Bye bye booster: A CDC study found the protection against hospitalization from Omicron provided by mRNA boosters to drop over 4 months from 91% to 78%. That’s not terrible, but it’s a far cry from the efficacy against earlier strains – for Delta, 97% and 89% respectively. And effectiveness against milder COVID-19 is likely to be still lower. These results are of particular concern for countries like Italy, Denmark, and the UK, which had intense early booster campaigns: a month from now about a third of older Brits will run up against the 4-month wall, as will 20% of older people in Italy including my husband and me.
Fourth doses: The big question is whether we should plunge ahead giving ever more boosters of the same vaccines, or wait for revised versions. Maybe fourth doses make sense for the immunocompromised, but I think monoclonal antibody products such as sotrovimab and Evusheld every few months is a better idea. For the rest of us, Israeli data remain uncertain. Antibodies skyrocket, but protection against symptomatic COVID-19 is at best halved. Dose 4 protects better against severe disease, but with only 23 days’ follow-up we have no idea how long it lasts.
Universals: Better than a whack-a-mole approach would be a single vaccine that could counter present and future SARS-CoV-2 variants, as discussed in a recent Washington Post article. The US Army vaccine that showed promise in animals should soon see results from a Phase 1-2 human trial, and groups at the University of Pennsylvania, Osaka University, and the University of Washington are working on others. Last May Duke researchers described a vaccine that neutralized several early variants in monkeys, but they still haven’t gotten around to human trials. Many “universal” vaccines actual involve bits of the spike protein taken from previous variants, which seems unlikely to yield protection from future ones. A group of researchers from the Italian central research institute, the Istituto Superiore di Sanità, are trying a vaccine that targets the N protein, a portion of the SARS-CoV-2 virus they seem to think is unlikely to mutate. So far they’ve only gotten as far as studies in mice. Some commentators think it may be good at preventing severe COVID-19, but poor against milder disease.
Kids: in Israel, 2-dose vaccination cut Omicron illness in kids by half but wore off in a few months – as did the booster… Twelve cases of post-vaccine myocarditis had been reported in American children ages 5-11 by December 17, 2021 – 4.3 per million in boys after the second dose. That’s less than a tenth of the rate in teenagers, but may be far enough from zero to make some parents uneasy.
Johnson & Johnson: The final results of their Phase 3 single-dose trial are very mediocre, only 53% efficacy overall. It peaked at 62% a month after vaccination but immediately started to drop, falling off a cliff around 6 months. Protection was stronger against severe disease (75%) and death (83%), so there should still be a role for this convenient, inexpensive vaccine in the developing world. But… J&J quietly paused production of its vaccine months ago, and won’t start up again until late spring.
Moderna: The World Health Organization may enable South Africa to produce this vaccine, with no help from the company itself.
Pfizer vs. Moderna: We already knew two doses of vaccine did very little to prevent Omicron disease, and the latest British data confirm that even boosters wear off disturbingly fast. Protection against hospitalization from a Pfizer booster dropped over 10 weeks from 90% to 75% – but Moderna won out as usual, remaining 90-95% effective.
CoronaVac: How about the tens of millions of people who got Sinovac’s vaccine in early 2021? It was mediocre to start with and becomes entirely ineffective by 6 months. A British group has tried giving various boosters to Brazilian Sinovac vaccinees 6 months after their second dose, and examined the antibody response. Additional doses of the original vaccine, AstraZeneca, and Johnson & Johnson did little to raise antibodies, but Pfizer hoisted them to levels that might be protective. Follow-up of the same research subjects should be able to tell both whether the antibodies translate into real-world protection and how long either lasts.
ARCoV: The two main Chinese vaccines used so far, from Sinovac and Sinopharm, aren’t very effective. Now they’re doing Phase 3 trials of an mRNA vaccine. In Phase 1-2 trials it wasn’t as good at stimulating the immune system as Moderna or Pfizer, and gave more side effects, but – who knows? – it could perform well enough in the real world to let China continue to avoid buying foreign products.
Gaping gaps: The CDC is considering increasing the standard interval between the first two doses of mRNA vaccines from 3 or 4 weeks to 8 weeks, on grounds of greater efficacy and lower risk. I’ve been advocating for some time lengthening the dose gap for males under 30, to lower the risk of myocarditis. After reviewing the Canadian efficacy data that were presented to the CDC on February 4th I can’t agree with broadening that advice, especially for Pfizer. By 2 weeks after the first dose the Moderna vaccine gave more than 80% protection against pre-Omicron variants, and it is relatively long-lasting. A single dose of Pfizer doesn’t even reach 60% efficacy pre-Delta and only 33% against Delta, versus 90%-plus for two doses. And its efficacy will start fading before the second dose kicks in – probably why Pfizer was so ineffective among the elderly in the UK, where the gap was 12 weeks. I find the manuscript from Canada itself unconvincing. It does show improved effectiveness following a second dose given after an 8-week interval, but does not report the crucial data, 2-10 weeks following the first dose. A longer gap should be less of a problem with young people with robust immune systems, so I stick with my position: increase the gap to 8 weeks for younger males, but otherwise leave it unchanged.
What’s up with the Danes?
Let’s say you’re a national health authority and you’ve been watching your country’s COVID-19 cases, hospitalizations, and deaths climb steadily for 4 months. What do you do?
If you’re Danish, you scrap all mitigation measures and make like it’s 2019. No more masks, no more Green Passes, no more self-isolation if you’re sick. WOW!!! I’m not impressed with their justifications, nor is epidemiologist Dr. Eric Feigl-Ding, who tweets: “Danish 🇩🇰 political leaders have completely lost their frigging minds.” Maybe, just maybe, it’s all about putting economics (ski tourism) over public health. And surprise, surprise! After a mere two weeks there were already surges in cases (up 43%), hospitalizations (up 30%), and even deaths (up 14%), though deaths ordinarily lag behind other indicators by weeks.
True, case numbers can be inflated by high testing rates, but Denmark does only twice as many tests as Italy, and finds 7 times as many cases. Deaths, too, are inflated by Denmark’s peculiar definition (which it may soon scrap): any death within 30 days of a positive swab is attributed to COVID-19 even if it’s in a car crash. But the death rate is on its way up rather than – as elsewhere in Europe – sharply down.
Exactly how confusing are US guidelines? Now we know. Researchers presented 338 well-educated American adults with several hypothetical testing scenarios and randomized them to receive either: 1) official FDA-authorized instructions for how to act on the results, 2) experimental instructions designed by the researchers, or 3) no instructions at all. Those who received the authorized instructions were the least likely to behave according to CDC guidelines, which in any case the researchers found inconsistent and hard to follow. Even the best policies won’t have any impact unless the target audience understands them; communicators must carefully design and test messages whenever stakes are high.
As for how off-base those CDC guidelines are, take the one allowing COVID-19 patients to be out and about after 5 days. It turns out that more than 40% of vaccinated health care workers who feel enough to work after breakthrough infections still have positive antigen tests after day 5.
|Pharmacy rapid-test tent
We expats were charmed to see the Washington Post run an article about the rapid testing tents that have sprouted outside Italian pharmacies. But the journalists fail to mention that much of what they describe is illegal – all those people “coughing, sniffling…” Sidewalk tents and private laboratories can screen people who feel fine, but are strictly forbidden to test anyone with possible COVID-19 symptoms. Symptomatic people must either get tested on the public system or have a private lab make an expensive house call. Yes, the nurses taking swabs at those pharmacy tents often neglect to ask, and patients often lie, but it would have been nice if the journalists had done their homework.
Only 7% of the Italian population over age 12 are unvaccinated, but they accounted for 56% of ICU patients at last call.
Several Italian intellectuals, featuring philosopher Massimo Cacciari, have constituted “The Dupre Commission.” The group started out gently, as befits a bunch of eggheads, saying they’re not no-vaxers but different-vaxers: the mRNA vaccines lose effectiveness so fast that Italy should switch to the one from Sinovac, which “Has a lower initial efficacy but lasts longer.” False! Chinese researchers found that 6 months after a second dose of Sinovac neutralizing antibodies were practically gone, and only 20% of vaccinees still had any kind of antibodies. Contrast with Pfizer, which after six months is still at least 50% protective, the highest level Sinovac ever attains. After softening us up, the Dupre bunch moved on to minimizing vaccines’ benefits, exaggerating their risks, disparaging Green Passes, and finally arriving at true nutcase stuff: the pandemic was pre-planned by Big Pharma, vaccines will only make it worse, etc., etc.
Mystery: a month ago Italy was running as many COVID-19 deaths per capita as the United States (now they’re doing somewhat better) despite more Italians being vaccinated, twice as many boosted, strong vaccine passports, a slimmer and healthier population, and universal indoor masking. Fewer than half as many Italian patients per capita were in ICUs, but they were much more likely to die. In Germany there are more ICU patients but far fewer deaths. Here are a few hypotheses for why (Italian experts share my suspicions):
1) Twice as many Americans per capita had had COVID-19 in the previous year.
2) In late January more Italian cases were due to the deadlier Delta variant (7.6% vs. 1.4% in the US as of January 24th). But that proportion was still higher in Germany (11.6%), where they’re doing fine.
3) 23% of Italians vs. 16% of Americans are over 65, and age is the biggest risk factor for severity. But, again, look at Germany, where 22% are elderly…
4) Then there’s those red zones in the US where families insist COVID-19 be left off death certificates, and coroners play along.
5) Non-COVID-19 patients can spend days parked in Italian ERs, risking catching the virus. But even in the US, more than 2% of non-COVID-19 patients get infected with SARS-CoV-2 during their hospitalization.
6) Patients can die in Emergency Rooms waiting for an ICU bed.
7) Italian ICUs are not always world-class, in my experience.
8) Some Emergency Room staff are triaging selected fragile patients straight to nursing homes, where they’re more likely to die.
Hospitalized with, hospitalized for?
|An Italian COVID-19 ward
There’s a raging debate everywhere over how many SARS-CoV-2-positive inpatients should be counted as COVID-19 hospitalizations. See this excellent Washington Post article, where 3 California clinicians describe their personal experiences.
People who come to Italian hospitals for other reasons but test positive in the ER are counted as cases and sent to COVID-19 wards, but – as in some US states – they are no longer counted as “COVID-19 hospitalizations” unless their condition deteriorates while they’re in the hospital. Assuming their hospital course will be adequately monitored…
ICU occupancy may be artificially high if a patient with, say, a heart attack tests positive incidentally. But most infected individuals do eventually develop symptoms, and those sick enough to be in the ICU for sepsis, strokes, etc. are at extremely high risk of COVID-19 complications or death.
…Death, though, is a very reliable parameter. Italian authorities, for example, list 4 clear and sensible criteria for listing COVID-19 on a death certificate:
- A positive PCR swab for SARS-CoV-2
- A clinical picture compatible with COVID-19
- Absence of any other obvious cause of death
- No period of remission between illness and death
|Tweeted by Dr. Satoshi Akima
Deltacron: As suspected, it doesn’t exist.
BA.2: This Omicron subvariant is now dominant in Denmark, South Africa, and several Asian countries. It’s harder to detect and may be a bit more transmissible, though perhaps only from unvaccinated patients. It unfortunately resiststhe anti-Omicron monoclonal antibody sotrovimab, though it remains susceptible to Lilly’s bebtelovimab (Evusheld wasn’t tested). We have no real-world evidence it causes worse illness, better evades vaccines, or reinfects Omicron survivors. Overall it looks more like Alpha than like Delta or Omicron, so even if it does take over the world it may not cause much damage. If the WHO isn’t yet worrying, neither am I.
Omicron: Is it really milder? Yup. Canadian researchers matched 9087 Omicron patients one-to-one with 9087 Delta patients on sex, age, vaccination status, time since most recent vaccine dose, and onset date; Omicron patients turned out to be 43% less likely to need hospitalization, 81% less likely to wind up in an ICU, and 88% less likely to die. Guess that settles it.
A Washington State bar advertised an event on Facebook specifically aimed at spreading COVID-19: “Come see the show, maybe catch the virus, or just stay home and whine. Tickets 10 bucks or 6 with proof of omicron positive test.” Four staff quit, 3 bands cancelled their engagements, and attempts to pass the post off as a joke fell flat.