Asian countries framed the challenge differently. Recalling SARS (2003) and MERS (2012), they weren’t taking any chances. Masks *might* help in this new disease, so let’s wear them just in case. (e.g. Taiwan: https://cnbc.com/2020/07/15/how-taiwan-beat-the-coronavirus.html…) 10/

For many mission-critical weeks in early 2020, these bodies persisted in saying “there’s not enough evidence of benefit” and (without evidence) “there could be harms”, and insisting that these arguments justified inaction. 9/

But that’s what happened. Tragically, WHO along with Public Health England, CDC and many other bodies around the world all focused on two things: a) the lack of incontrovertible, definitive evidence and b) speculation about possible harms. 8/

A bit of cloth over the face simply doesn’t have the same risks as a novel drug or vaccine, and *doing nothing* could conceivably cause huge harm. Arguing for “caution” without engaging with the precautionary principle was scientifically naïve and and morally reckless. 7/

New drugs & vaccines may have toxic side effects worse than the disease itself. Hence, it’s appropriate to require definitive empirical evidence from RCTs of the benefit-harm balance before they're introduced. But critics inappropriately applied the same rules to mask studies. 6/

Critics of that early paper were right that the empirical evidence was weak. But they didn't engage—and 16 months later have still not engaged—with the moral arguments. They continue to argue that the best course of action in the face of empirical uncertainty is to do nothing. 5/

In this BMJ paper, we presented very limited and indirect empirical evidence (from non-Covid studies) and also *moral evidence* to argue for the precautionary principle: let’s all wear masks, *just in case*. https://pubmed.ncbi.nlm.nih.gov/32273267/ 4/

Was this association or causation? Early on in the pandemic, we didn’t know. But – important point - nobody in these Asian countries seemed to come to harm from wearing a mask. 3/

Let’s start with observational data. Countries that introduced mandated masking within 30 days of the first case (mostly Asian) had *dramatically* fewer Covid-19 cases than those that delayed beyond 100 days (mostly Western). https://ajtmh.org/view/journals/tpmd/103/6/article-p2400.xml… 2/

I’m struggling with how best to stress how fragile the global situation is, so I’ll be blunt: Each week >2.6 million cases and >53,000 deaths are reported to @WHO Reported, meaning that there are many others. Stop and think about that. (Short thread)

It’s July 4th calendar-related but yesterday was the first day where TSA passenger traffic exceeded the same-day 2019 level:

Serious cases among vaccinated: 7 people in 10,000 for #AstraZeneca 6 people in 10,000 for #Pfizer these numbers should get lower as we vaccinate more! Go get your #vaccine we can't afford another lockdown or surge in cases! #VaccinesWork

Updates to the wiki of COVID-19 vaccine misinformation: Safety syringes retract: https://hackmd.io/ovEzSQWcRp2bctQn8MYElQ#FACT-Safety-syringes-retract-after-an-injection… Fertility not affected by the vaccine: https://hackmd.io/ovEzSQWcRp2bctQn8MYElQ#FACT-COVID-19-vaccines-have-no-impact-on-your-fertility… @SciBeh @stefanmherzog @johnfocook @Sander_vdLinden @adamhfinn @julieleask @CorneliaBetsch @PhilippMSchmid

First phase 3 efficacy results for the Clover vax are out in a detailed press release. https://cloverbiopharma.com/news/83.html Great relief: it's a critical vax for the COVAX supply. Protein subunit vax, with a Dynavax adjuvant & alum. Records on it: https://zotero.org/groups/2528572/covid-19_vaccine_results/tags/SCB-2019%20Clover%2FDynavax/library… ...1/n HT @lutl88

How it started. How it’s going.

New @ONS #LongCovid estimates published today: 1.1 MILLION (1.7% of the whole UK population). Up from the summer estimate of 1.5%. 211,000 people with daily activities “limited a lot”. Greatest % in working age (35-69y). Rising prevalence in 17-24y. A tsunami of chronic illness.

On the J&J booster news, keep in mind: 1. Median follow-up since 2nd dose was just 36 days, 2. Efficacy vs moderate COVID was 75% globally, and 3. total number of cases in the US was 15. Please don't take this to mean that a 2nd dose provides long-term increase in protection.

Hmm okay. J&J two dose results are a bit more complicated, with different results globally versus within the US.

because of vaccine mandates we’re losing teachers who don’t believe in science, healthcare workers who don’t believe in medicine, and police who don’t believe in public safety…

Will you fall into the conspiracy theory rabbit hole? https://washingtonpost.com/opinions/interactive/2021/conspiracy-theory-quiz/?utm_campaign=wp_post_most&utm_medium=email&utm_source=newsletter&wpisrc=nl_most&carta-url=https%3A%2F%2Fs2.washingtonpost.com%2Fcar-ln-tr%2F34e4a0d%2F615dcde29d2fda9d41004e99%2F5faaf72f9bbc0f331650ee35%2F17%2F70%2F615dcde29d2fda9d41004e99… by @databyler @codingyan Good breakdown on some of the social forces (like ideology) that drive conspiracy theories. Despite the fact I study topic, still amazed how many believe this stuff.

This is the summary figure I meant to show. From the amazing resource that is http://CoVariants.org. Gracias, @firefoxx66! 31/

Vaccination coverage remains low at ~30%, and most countries are preparing for a rise in cases in the following weeks with the arrival of Delta. So, there's room for novel variants to emerge locally in the next months. We'll see how that goes. 30/ FIN.

But increased immune evasion may not be enough to overcome Delta's high transmissibility. As new cases fall to their lowest levels of 2021, we see Delta replace Gamma in Rio de Janeiro and Lambda in Lima. Delta is also expanding in Colombia and may replace Mu soon. 29/

So, it is possible that Gamma, Lambda, Mu, and local variants prevented Alpha from dominating Latam in early 2021, possibly because they could transmit better in populations with high levels of natural immunity from large initial waves of 2020. 28/

What about other countries? We see diverse mixes of regional variants plus Alpha and other imported lineages in Argentina, Chile, and Ecuador. Other countries report less than 1000 genomes, so it's hard to assess their situation https://medrxiv.org/content/10.1101/2021.07.19.21260779v1… https://auspice.cov2.cl/ncov/chile-global… 27/

Additional evidence on Mu is scarce for similar reasons to Lambda: (1) limited capacity for follow-up studies, and (2) these variants have not been a significant threat in high-income countries like Delta is. The recent VOI designation by WHO should bring new data on Mu soon. 26/

Data from PHE and a recent preprint suggest reduced neutralization by convalescent + post-vaccine sera to levels greater than Beta, potentially making Mu the most immune-evading variant yet. https://biorxiv.org/content/10.1101/2021.09.06.459005v1… https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1009009/6_August_2021_Risk_assessment_for_SARS-CoV-2_variant_VUI_21JUL-01.pdf… 25/

Colombia has had two epidemic waves in 2021. Gamma + 2020 lineages caused the first peak (Feb), and Mu drove the second peak (June-July). 24/

Finally, B.1.621 / Mu. It was first reported in Colombia, with E484K, N501Y, P681H, and a similar emergence pattern and export to other regional variants. https://virological.org/t/emergence-of-lineage-b-1-621-in-latin-america-and-the-caribbean/742… https://medrxiv.org/content/10.1101/2021.05.08.21256619v2… 23/

Evidence for reduced antibody neutralization to levels similar to Gamma and Delta through Δ246-253, L452Q, F490S: https://biorxiv.org/content/10.1101/2021.08.14.456353v1… https://biorxiv.org/content/10.1101/2021.08.25.457692v1.full.pdf… https://medrxiv.org/content/10.1101/2021.08.20.21262328v1… https://medrxiv.org/content/10.1101/2021.07.21.21260961v1… https://biorxiv.org/content/10.1101/2021.07.19.452771v3… 22/

Evidence for higher ACE2 affinity + infectivity https://biorxiv.org/content/10.1101/2021.07.28.454085v1… https://medrxiv.org/content/10.1101/2021.06.28.21259673v1… 21/

Most countries in Latam have limited capacity for follow-up lab + epi studies, so it's been challenging to assess the effect of C.37 mutations on transmission, virulence, or potential immune scape. Here's a summary of the early evidence from PHE. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1000662/8_July_2021_Risk_assessment_for_SARS-CoV-2_variant_LAMBDA_01.00-1.pdf… 20/

C.37 has accumulated additional mutations of interest appearing multiple times on the global tree: (1) Additional deletion in S:61-75, (2) Q675H, (3) I714V https://nextstrain.org/community/quipupe/C37_lineage… 19/

C.37 likely originated in Peru in late 2020, given its earlier growth and peak frequency. However, the earliest C.37 record is from Argentina from 8-Nov (EPI_ISL_2158693, although this sequence is identical to the second earliest case, from 29-Jan). 18/

By the time we first reported C.37 in late April, it had already been expanding in Chile and Argentina, in the presence of Alpha and Gamma. https://virological.org/t/novel-sublineage-within-b-1-1-1-currently-expanding-in-peru-and-chile-with-a-convergent-deletion-in-the-orf1a-gene-3675-3677-and-a-novel-deletion-in-the-spike-gene-246-252-g75v-t76i-l452q-f490s-t859n/685… 17/

We first noticed it in Lima in late December (<1% of cases then). By April, at the peak of Peru's second wave, it was already 80% of all sequenced cases. 16/

Lambda has its unique constellation of mutations in the Spike gene. Particularly interesting at first: Δ247-253 in NTD, and L452Q (similar to L452R in Delta) + F490S in RBD. It also shares a convergent deletion in ORF1a:3675-3677 with Alpha, Beta, Gamma, Eta, and Iota. 15/

C.37 evolved from B.1.1.1, a lineage imported from Europe early in the pandemic. By late 2020, B.1.1.1 accounted for 20%+ of sequenced genomes from Peru. 14/

C.37 has been exported to 30+ countries, mainly in the Americas and Europe. 13/

Now, on to C.37 / Lambda. I will explain more about it since I am most familiar with this story. https://medrxiv.org/content/10.1101/2021.06.26.21259487v1… 12/

Moving to Uruguay, which contained the virus for most of 2020. Cases rose in November with P.6, a B.1.1.28 sublineage with S:Q675H+Q677H. P.6 was replaced by P.1 by April, coinciding with a sharp rise in cases and deaths. https://go.nature.com/3jTTlrm https://bit.ly/3tunQHt 11/

Elsewhere in Brazil, P.2 / Zeta evolved independently from P.1 and had S:E484K but not N501Y or K417N. It peaked around December and has been exported to 40+ countries but has since been replaced by P.1 in most of Brazil. https://journals.asm.org/doi/10.1128/JVI.00119-21… 10/

Gamma has now generated multiple sublineages, has been exported to 80+ countries, and is arguably the most successful variant in the region so far. https://outbreak.info/situation-reports/gamma… https://cov-lineages.org/lineage.html?lineage=P.1… 9/

P.1 / Gamma is already known to everyone. Its emergence from Manaus and its spread have been described in detail by our colleagues at CADDE, Fiocruz, and others in Brazil. https://science.org/doi/full/10.1126/science.abh2644… https://nature.com/articles/s41591-021-01378-7… 8/

More on the global disparities in genomic surveillance in this post from @AndersonBrito_ https://twitter.com/AndersonBrito_/status/1431235752944340992?s=20… 7/

South America has a very limited capacity for genomic surveillance: We contribute less than 2% of sequences on GISAID (65k by September 8), which represent less than 0.5% of total cases for most countries. We are also very slow to upload sequences to GISAID/Genbank. 6/

So, we have had very high levels of transmission and it's not surprising that novel variants beyond P.1 would emerge in the region, usually in the presence of high seroprevalence from the initial wave of cases in 2020. https://go.nature.com/3DXhrJR https://bit.ly/3yVcGww 5/

Looking at excess deaths relative to previous years, many Latam countries appear at the top. http://ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386938… 4/

While much attention has focused on Brazil due to the magnitude of its epidemic (584k+ COVID deaths, pop. 210M+), relative to pop size, many countries in the region have had similar or worse epidemics. 3/

Let me first bring your attention to Latin America, an epicenter for COVID-19 since the start of the pandemic. We have around 8% of the world population (660M) yet accumulate 25+% (1.4M) of COVID deaths. http://ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386938… 2/

Hello from Peru. Recently, there's been some buzz about Lambda, Mu, and variants coming out of South America. Here's a summary of the variant landscape in the region where Delta is yet to dominate (but will soon), as we prepare for a new wave of cases in the following weeks. 1/n

We hope to understand how vaccines are helping some long haulers and not others. Understanding the pathophysiology of #longCOVID is direly needed to develop diagnostic tools and therapy. Thank you @thitran3 et al for this important study (end)

With input from @Survivor_Corps and @patientled, we are studying the impact of COVID vaccines on #longCOVID symptoms and correlating the changes in the immune responses to symptom changes. Led by @DaisySMassey @hmkyale (7/)

A prospective observational study by @gushamilton team showed small overall improvement in #longCOVID symptoms in vaccinated patients. (6/)

Overall, this study adds to growing evidence that vaccines can improve symptoms and lessen the disease impact in #longCOVID. What is the evidence? An important patient survey from @LongCovidSOS showed impact of vaccines on long covid symptoms. (5/) …https://3ca26cd7-266e-4609-b25f-6f3d1497c4cf.filesusr.com/ugd/8bd4fe_a338597f76bf4279a851a7a4cb0e0a74.pdf

What about severe adverse effects of vaccines in long haulers? Of the 455 long covid patients who received vaccines, 2 led to hospitalization, 2 led to ER visit, 13 had relapse of long covid symptoms. (4/)

In addition, disease impact of long covid on patients’ lives were significantly reduced (symptoms improved) in vax group (long COVID IT score of 24.3) compared to unvax group (IT score of 27.6). (3/)

The study found that the rate of complete remission from long COVID symptoms doubled in vaccinated patients compared to unvaccinated long COVID patients. Wow, vaccines appear to be helping long haulers with recovery (2/)

An important new study looks at how COVID vaccines impacts symptoms in #LongCovid patients. @thitran3’s team used data from ComPaRe long COVID cohort to emulate a target trial (1:1 matched vax:unvax) measuring outcome at 120 days after baseline. (1/) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3932953

So IMO, OAS is a super-interesting thing to study from the standpoint of immunodominance and selective pressures on viral evolution. Pre-pandemic I didn't care much abt what ppl call it. But the term carries negative connotations that I don't find very helpful for #SciComm.

Second, we earlier made some recombinant RBD that carried mutations at all of the key neutralizing epitopes. We wanted a serological comparison to estimate neut responses. But the mutant protein was less thermostable. So the virus *might* take a fitness cost if it did this.

But what happens if the virus mutates its way out of all earlier neutralizing epitopes? First, as @GuthmillerJenna points out, neutralization is not the only mechanism of protection. Antibodies (and T cells ofc) made to other parts of the virus can help.

Similarly, this paper by @TheBcellArtist and @florian_krammer showed that post-mRNA vaccination, you do get beta-CoV X-reactive memory responses, but it obviously doesn't keep you from making primary neutralizing responses to SARS2.

So this is all wrt flu. But there are some data for SARS-CoV-2. E.g this paper by @MahevasMatthieu showed that while you get beta-CoV X-reactive responses early post-SARS2 infection, later naive responses specific to SARS2 take over (*loved* this paper).

So for these reasons, I don't like the blanket term of 'sin.' Recalling memory B cells against conserved parts of the virus can be a 'blessing' too, as stated by @SCOTTeHENSLEY and @MichaelWorobey. http://perspectivesinmedicine.cshlp.org/content/10/10/a038471.long

Moreover, making memory responses against invariant parts of the virus doesn't prevent new primary responses against other parts. It's possible these new responses get slowed or are weaker (reason to be cautious in over-boosting), but it's not absent.

...making immune responses to parts of the virus that haven't changed can be a good thing. This paper by my colleague @MichaelWorobey and co-authors shows that such antigenic imprinting (IMO a better term than OAS) can be protective.

Since immune responses are competitions, this could be bad by competitively inhibiting new responses against parts of the strain that changed. There's evidence that as a result, repeated flu boosters confer less protection (tho this v. complicated). But...

boosters from working at all. First, what is OAS? It's a concept proposed by Thomas Francis wrt influenza. It means that you keep making immune responses aimed at the original flu you first encountered even when exposed later to diff strains. https://jstor.org/stable/985534

Okay, day 2 of kid quarantine (Btw, he is fine and tested negative so far)! Let's dig into original antigenic sin and why IMO the term (not the concept) is too negative. It could start to be a problem at the margins if we boost too often, but I doubt it would prevent updated

'Behavioural responses to Covid-19 health certification: a rapid review' @ProfJohnDrury The first rapid review on possible behavioural effects of 'vaccine passports' and similar schemes *open access* https://rdcu.be/cm6qf

Bogus, damaging “consent form” being peddled over in the UK. Ugh. The anti-vaccine movement is trying every grotesque tactic under the sun.

this is the Giesicke argument from Sweden- it makes sense only if you assume there will be no medical progress in the delay period. With vaccines and treatment improvements we know this to be empirically false.

So there we are - rising cases in kids & their parents' generation, falling cases elsewhere, A high burden of Covid continues. I don't know how vax in teens, some immunity from high infections this summer plays out against autumn & people returning to normal behaviour.

In fact many countries in Europe are not seeing big surges in children - a combination of vaccination (starting at different times over summer) and mitigations in schools (bubbles, masks, ventilation) & lower community case rates.

Other countries that vaccinated teens are not seeing big back to school spikes in teens (but some are in primary school kids). E.g. Ireland... where cases in teens much lower than England but primary school kids about the same...

Many of these cases could have been prevented with vaccination over the summer. It already looks like it (+ prev infection?) *might* be having an impact in 16-18 year olds... at least they're the only year groups where cases fell over last 2 weeks instead of climbed.

ONS infection survey (to 18 Sept) also shows cases highest and rising in school age children. over 1 in 50 children had Covid that week.

The number of children testing positive this term has already almost exceeded the number testing positive over the whole of the summer term. (158K vs 172K). Estimates of Long Covid in kids range from 2%-14% - even 2% of 158K is 3,000 children developing Long covid...

Hospitalisations in 6-17 year olds were dropping steeply at the beginning of September (corresponding to the late Aug flattening in cases?) but are now rising steeply again - only children have rising admission rates at the moment.

In England, case rates in 5-14 are higher than their July peak - about 1.4% of all 10-14 year olds tested positive last week! Cases also going up again in 15-19 year olds. And in 30-59 yr olds - the ages most likely to be parents of school children. Following Scotland?

In Wales, cases have been going up steeply in under 17s. In N. Ireland, we've seen similar record breaking rates in 5-14 year olds.

And we are seeing that hospital admissions in their parents' generation have been going up - in fact are higher now than they've ever been in the pandemic despite high vaccination.

Cases in Under 15s in Scotland remain high but have come down quickly over the last 10 days or so - and we are seeing a drop in admissions in children too which is good. But a *lot* of children were infected in the first month of term - over 5% of all children under 15.

Regionally, cases are high all over Wales and in England concentrated in the Midlands and Yorks - lowest of all in London (!). That pattern is seen in positivity rates too so it's not just testing.

By nation, cases are going up in Wales and England but dropping in Scotland and dropping more slowing in NI. Similar pattern seen in ONS infection survey (but Scotland not dropping yet - always takes a few weeks for drops in cases to show in ONS)

Hospitalisations are going down in all nations (good!) and deaths might just be starting to go down too - although we've been averaging over 130 a day for several weeks now and over 8,000 people have died since 1 July.

Yes hospitalisations and deaths are much lower than Jan peak - but they are still higher than a year ago and all these cases will be lead to many people develop long covid unfortunately, including some children.

Cases are going up again in the UK - and we've now had substantially more confirmed cases in the Delta wave than the Alpha one. In a few weeks we'll have had more cases than we had between Sept 2020 and May 2021.

On tests - the number of lateral flow tests done (or rather, reported!) has really varied with big "back to school" spikes that then drop off. We're seeing that again now - and the drop off is mainly in students and not staff. Hard to say how this affects case numbers.

There are big gaps in full vaccination uptake between the most and least deprived communities, and lower uptake in ethnic minorities. This hasn't really improved over the last 6 months - whatever is being tried doesn't seem to be working.

Vaccination update to start: we've got good vaccination coverage - and excellent in older age groups. Almost 60% of 16/17 year olds have had one dose of vaccine in England (higher in Scotland). BUT

THREAD (a bit delayed) on UK & covid: TLDR: flattish cases overall are masking differences between nations, regions & age groups. And we're still out of whack with Europe.

70% of long Covid sufferers develop damage to at least one vital organ like the heart or liver, a new U.K. study found

and just to note official government guidance https://gov.uk/guidance/covid-19-coronavirus-restrictions-what-you-can-and-cannot-do#wear-a-face-covering…Quote TweetProf. Christina Pagel@chrischirp · 3hhate to say it, but the Labour conference looks kind of super spready... 4 days of rooms this full plus evening drinking in the bar... and not a mask in sight. twitter.com/SadiqKhan/stat…

"We've been creative": Puerto Rico fares better at COVID shots than many U.S. states, with 62% of its population fully vaccinated. Gov. Pedro Pierluisi tells CBSN how Puerto Rico benefits from vaccine mandates, rejecting misinformation and avoiding politicization of the pandemic

The political divide over vaccinations is so large that almost every reliably blue state now has a higher vaccination rate than almost every reliably red state. https://nytimes.com/2021/09/27/briefing/covid-red-states-vaccinations.html…

Here's the full story about "vigilante medicine" on ivermectin Facebook. Antivaxxers are starting to wrap doctors and ICUs into their dark conspiracy theories, as they suffer at home with ad-hoc COVID treatments that don't work. I hope you read it.

Ivermectin Facebook is a wildly dark scene, an inverted reality where medical consensus is an elaborate conspiracy to kill you and random people on Facebook have the secret cure. As Dr. Aditi Nerurkar tells me, "They’re starting to target the messengers—nurses and doctors.”

They’ve developed elaborate conspiracy theories about doctors and nurses in the process. They believe ventilators and remdesivir are secretly drowning patients’ lungs, not COVID itself. QAnon boards have begun calling hospitals to harass workers for not prescribing ivermectin.

As they’re home self-medicating, antivaxxers are furious that friends and family won’t be administered ivermectin at the hospital. So they’ve developed directions on how to get loved ones out of ICUs: put them in hospice care, then get them the miracle cures from YouTube.

Obviously, keep taking the ivermectin and hydroxychloroquine, the ivermectin groups say. But also gargle iodine. Buy a nebulizer and inhale food-grade hydrogen peroxide. Anything but the vaccine.

Facebook bans explicit antivaxx groups, but they don't ban groups for quack "cures" that antivaxxers push instead. So in the last couple of months, Ivermectin groups have become the new hubs for antivaxx messaging. But there's a problem: Ivermectin, by itself, isn’t working.

A quick thread: It’s hard to explain just how radicalized ivermectin and antivax Facebook groups have become in the last few weeks. They’re now telling people who get COVID to avoid the ICU and treat themselves, often by nebulizing hydrogen peroxide. So, how did we get here?

Iranian media reports that at least 44 people have died from alcohol poisoning and hundreds have been hospitalized after consuming bootleg alcohol in an effort to treat the coronavirus. The Middle Eastern country, which has been especially hit hard by the coronavirus – with 8,042 confirmed cases and at least 291 deaths as of Tuesday – has struggled to prevent the spread of the virus.The majority of deaths attributed to the coronavirus in the Middle East are in Iran.A false rumor has circulated throughout the country that drinking alcohol can cure or prevent the coronavirus. Drinking alcohol is prohibited in the country.Some citizens, according to Iran Health Ministry official Ali Ehsanpour, drank alcohol that substituted toxic methanol for ethanol, using bleach to mask the color. Seven bootleggers have been arrested.In one part of the country, Khuzestan, more people have died from alcohol poisoning than from the coronavirus in that area, according to the state news agency IRNA. More than 30 people have died from poisoning, and 18 have died from the virus.The rumor also has circulated throughout Indian social media, reports NDTV and the Times of India, which the World Health Organization has debunked. One iteration of the rumor suggests that spraying alcohol or chlorine can prevent the coronavirus from entering the body.

The world has administered 6 billion doses of COVID vaccines to 3.4 Billion people For people waiting for more data before getting the shot The data is in We've vaccinated nearly half of all humanity The vaccines are safe

Quite something for UK to take such a stance against so many countries vaccine rollouts…especially those countries administering the *exact same* vaccines as UK (Pfizer/AZ/Moderna/etc) As we’ve come to expect, UK’s latest travel policy is as unnecessarily complicated as ever

Understanding, MULTIPLE countries are not happy with UK government tonight One foreign minister tells me: “In our first meeting with Liz Truss, UK new Foreign Secretary, we expect her to explain why our Pfizer jabs are supposedly different to Pfizer jabs in UK or Europe”

UK government confirm tonight that if a person has been vaccinated in Africa, or South America, or countries including UAE, India, Turkey, Jordan, Thailand, Russia… …you are considered “unvaccinated” and must follow “unvaccinated” rules = 10 day home quarantine & tests

Mike Yeadon's Covid vaccine claims: Fact Checked. Dose ranging studies DID take place Vaccines AREN'T dangerous in pregnancy You CAN refuse a vaccine Our sources http://fullfact.org/health/Mike-Yeadon-video

Check out this graphic we showed Friday night in the @CNNSitRoom - it’s so disturbing. Three months ago the U.S. was averaging about 200 Covid-19 deaths a day. Now it’s close to 2,000 deaths a day. The Delta variant is so dangerous. Be careful and get vaccinated.

These are risk levels that you pose to other people. They're compared with you as - a nonsmoker - a sober driver - a vaccinated person. Unvaccinated? 5x as likely to get sick, for 3x as long. Total risk to others? 15x a vaccinated person Details:https://election.princeton.edu/2021/09/15/your-failure-to-vaccinate-is-worse-for-other-people-than-drunk-driving/

The booster debate shows how much we need better real-life data collected here in the U.S. and we don't because public health has been chronically underinvested in and neglected. The debate amidst profound global vaccine inequity grows more frustrating.

Update:

Fascinating clip from Dr.Fauci where he says we should treat COVID like how we treat measles. Not flu.

Correcting a statistic I gave at the press conference today, 19 July. About 60% of hospitalisations from covid are not from double vaccinated people, rather 60% of hospitalisations from covid are currently from unvaccinated people.

A SCOTUS legal precedent about vaccines for the public good.

46 y.o. Brandon Haynes from #Louisiana died from COVID. "I asked him to take the vaccine & he wouldn't. The best way I can honor him is to request friends and family members take the vaccine.” Mom provided a vaccine clinic at his funeral #SoulsLostToCovid

10/ We've got to reset expectations about what COVID vaccines can or can't do. Our goal should be to turn COVID into a more benign disease like the flu... and we can do that through VACCINATION.

9/ Your risk of getting SARS-CoV-2 infection is a composite of: - immunity elicited by vaccination - how much virus is circulating in your community And how much virus is circulating in your community depends on what other people do, especially: - masking - vaccination

8/ Now let's say the incidence of SARS-CoV-2 infection is 18 per 100K among UNvaccinated persons. This is the overall incidence in NYC right now. With a vaccine that's 90% effective against infection, you'd have an incidence of <2 per 100K among vaccinated persons.

7/ Let's say the incidence of SARS-CoV-2 infection is 200 per 100K among UNvaccinated persons, as it is in some KY & TN counties right now. With a vaccine that's 90% effective against infection, you'd have an incidence of 20 per 100K among vaccinated persons.

6/ Vaccine effectiveness represents a RELATIVE RISK REDUCTION, not complete risk elimination.

5/ Remember:

4/ A vaccine doesn't have to induce sterilizing immunity to be highly effective.

3/ We might see something approaching sterilizing immunity SOON AFTER COVID vaccination, https://cdc.gov/mmwr/volumes/70/wr/mm7013e3.htm… but NOT long-term.

2/ We used to think vaccines for measles and smallpox provided sterilizing immunity. But as our tools to measure viruses have gotten better, we've learned that low-level infections can still occur after vaccination:

1/ Great piece by @TheAtlantic's @KatherineJWu on the myth of sterilizing immunity: https://theatlantic.com/science/archive/2021/09/sterilizing-immunity-myth-covid-19-vaccines/620023/… Sterilizing immunity: when the immune system 100% COMPLETELY blocks a virus or other pathogen from infecting someone

I’m fully aware this won’t stop “the vaxx gave me 5G” jokes. But #LetsVaxx anyway.

Because measles used to infect >99% of humans, this makes the humble measles vaccine perhaps the most important vaccine ever to be developed as it has reduced infections from all other pathogens. Biology never ceases to amaze! 8/

Thus, through a combination of mathematics, ecological modeling, and advanced immunological techniques we were able to discover that by erasing immune memory, measles once was partially responsible for ~50% of childhood infectious disease deaths. 7/

We found that measles eraaed up to 80% of people’s existing antibody repertoire (Immune Amnesia confirmed) and that they then had to build it back through vaccines or infections - placing them at high risk for disease for years. 2nd main paper here: 6/

We used VirScan (developed in Elledge’s lab) to monitor 100,000’s of antibodies in children who were naturally infected w measles and, w Diane Griffin at Hopkins, experimentally infected macaque primates. 5/

A few years later, with @RLdeSwart (involved w 1st studies too and who coined the term immune amnesia) and Steve Elledge and colleagues, we followed up by profiling the immunological memory in kids before and after they were infected w measles. 4/

We hypothesized then that measles was destroying B cells and plasma cells responsible for retaining immune memory. This is bc the measles virus specifically attacks immune memory cells via attachment to CD150 - a protein enriched on memory cells. 3/ https://nature.com/articles/ni0103-19… 3/

In 2015, along w Bryan Grenfell and colleagues, we discovered an exceptionally strong relationship between measles epidemics and all cause childhood mortality - following outbreaks of measles, child mortality was increased for 2-3 years. 2/

With all the news of vaccines & immunity, did you know measles infections destroy immunity and cause “Immune Amnesia”, increasing risk of all other infections Our research in ‘15 & ‘19 discovered this & the abbreviated story is written up nicely here: 1/

Vaccine passports work! @CBCCalgary just reporting 175% increase in AB vaccination bookings after BC implemented vaccine passport. Imagine the increase in vaccinations if,heaven forbid,we implement a similar strategy in AB? @CBCFletch -would be great to see the reference for this

Meta-analysis of 24 vaccine incentive programs suggest they don't work or effects are so small we can't precisely measure them (practically null)

This is what caving to political pressure looks like. Pfizer vaccine is leady and non-durable and risks are mounting. If we had tried to pulled this kind of sh**T in the Trump White...fill in blank. F.D.A. Grants Full Approval

Great to see numbers continuing to climb; seems like they’re climbing maybe as fast as supply allows!

For the under 60s, it is a bit more complicated. Under 50s have Pfizer with a gap of 3-6 weeks. Let's assume 4. For those in their 50s, they first had AZ and are now getting Pfizer - a 4-week gap doesn't work so well atm, but might work better in future. But for other ages?

With those assumptions, this is the projection for all age cohorts getting 2 doses... Australia might get 70% of the 16+ cohort double vaxxed in the middle of October (with some extrapolation by eye).

A ten week gap between AZ doses means that we expect almost 90% of those 70+ years old to be double dosed by early Nov (dashed lines). The actual double dose rates are soli lines. The 10-week gap looks an OK assumption. We'll see in a couple of weeks.

So that is first doses. What does it mean for second dose coverage in future? Well, let's play around with that...

In Australia, the over 60s are getting AstraZeneca. The recommended time between doses varies a bit. It was initially 12 weeks, but that has been reduced in some outbreak areas. Let's assume 10 weeks as a rough average.

Vax rates in the last week are highest in the youngsters. #WithaRocket

#COVD19 vax rates are still increasing in Australia. The over 70s are approaching 90% with first doses. The red line is the rate for all people >=16. That should reach 60% soon.

Six key words to remember when interpreting the selected variables in a multivariable analysis 1. Don’t 2. Pretend 3. They 4. Are 5. Causally 6. Important

NEW: in the last couple of weeks there have a *lot* of new studies out assessing vaccine efficacy, many of which have touched on the question of waning immunity. Unsurprisingly, these have prompted a *lot* of questions. Time for a thread to summarise what we do and don’t know:

Some good news from PHE's Friday Delta tech briefing update: two sublineages of Delta called AY.1 and AY.2 do not seem to be any more worrying that Delta in terms of transmissibility or vax resistance. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1012644/Technical_Briefing_21.pdf

I don't know why Florida is such a crazy outlier. My guess is a combination of: 1) The vax share doesn't reflect current population: US snowbirds + Latin American vaccine tourists 2) Low-masking culture among young ppl pre-wave 3) Heat + air-conditioning, something something

Adult hospitalizations since July 1 vs. vaccinations, by state: 1) The relationship between more vaccines and less hospitalization is pretty straightforward. 2) Holy moly, Florida. Among states with more than one shot per person, FL really is on its own island of pain.

Speaking generates more aerosols than breathing. Don’t take your mask of to speak!Quote TweetMaya Chavez@maya_chavez_ · 30 AugEvery single admin has pulled their mask off to address the staff.

Eight notable variants of SARS-CoV-2 have been found since September 2020. What do we know about them so far? This roundup includes a video showing how the first four variants of concern spread across the world #BMJInfographic http://ow.ly/roYK50G0lI6

there is a literature on arguments from ignorance (viewed from a Bayesian perspective) that can help here, I think https://sciencedirect.com/science/article/pii/S1364661320300206…Quote TweetNate Silver@NateSilver538 · 29 AugIn general I'm suspicious of the term "no evidence", which is rarely literally true. When people say that it often means they're setting the bar extremely high because they're clinging to arbitrarily-chosen null hypotheses, sometimes for political reasons. https://twitter.com/NateSilver538/status/1398013330774704133?s=19…Show this thread

Today, Mississippi surpassed New York in total coronavirus deaths per capita. For a long time, I didn't think we'd ever see any state end up worse than New York or New Jersey. One year ago today, Mississippi's death rate was less than half of New York's. https://nytimes.com/interactive/2021/us/covid-cases.html

the problem is the leaders, the media figures, the physicians who know better but would rather get in on the griftQuote TweetJonathan Howard@JHowardBrainMD · 31 AugBook review from 1911 of amazing book called “quackery and nostrums” and why no one should have been surprised many chose ivermectin and such over vaccines. https://jamanetwork.com/journals/jama/article-abstract/448782…Show this thread

For anyone who wants to know the source of this information, every element of it comes from the Ontario open datasets on vaccination, which are published daily (although the hospital/ICU data isn't updated on Sun./Mon.). https://data.ontario.ca/dataset/covid-19-vaccine-data-in-ontario

Explaining: The unvaccinated group should have much (much) higher natural protection from COVID by virtue of being younger on average; the fact this group is showing up in hospital much more often *despite this* tells you the vaccine is even more effective than we know.

That said: We don't have an age breakdown of these cases, and the huge demographic difference between the vaccinated group (much older + almost all seniors) unvaccinated (much younger + all young children) means this effectiveness number is 100% certainly higher even than this.

Here's the Tues. update of Ontario cases and hospitalization by vax status, w. pop. adjustment, the minimum needed to interpret this. The data shows the vaccine having a 92-95% effect on serious illness. We now have three weeks of data on this, and the result seems consistent.

Replying to @Marta_Mang and @chrstn_eSciBehReconfiguring behavioral science for crisis knowledge management: Help us foster proper science without the drag.scibeh.org

Marta.@Marta_Mang·26 Aughttps://pubmed.ncbi.nlm.nih.gov/33949091/ Do I know anyone who's worked on something similar to what @chrstn_e and colleagues describe here, ideally in the #cogsci/#BehaviouralScience field? #AcademicTwitter #AcademicChatterStrength is in engagement: The rise of an online scientific community during the COVID-19 pandemic...Many scientists, confined to home office by COVID-19, have been gathering in online communities, which could become viable alternatives to physical meetings and conferences.pubmed.ncbi.nlm.nih.gov

Quote TweetAAP News@AAPNews · 31 AugMyocarditis risk 37 times higher for children with COVID-19 than uninfected peers, according to a new @CDCMMWR: http://ow.ly/8HDS50G1Sjb

Oops the graph in the first tweet has some double-counting in December. Here's the corrected version:

Here's the python code to replicate the analysis. https://gist.github.com/jph00/635082450a8c9491dee52d5158c790a6

We now know that Delta is much more severe than previous variants. However we don't yet know what the impact on children will be, especially long term.Hospital admission and emergency care attendance risk for SARS-CoV-2 delta (B.1.617.2) compared...This large national study found a higher hospital admission or emergency care attendance risk for patients with COVID-19 infected with the delta variant compared with the alpha variant. Results...thelancet.com

This terrifying observation has previously been hidden by a lack of reporting of cases by age group. Thanks to vaccinations, cases amongst adults have been lower than the previous wave. But children are totally unprotected. Delta is ripping through them.

Jeremy Howard@jeremyphoward·30 AugI've been analyzing the UK covid data and I've just discovered something shocking. Cases in chidrens in England have just smashed all-time highs. Nearly double what they've ever been before. And rising VERY rapidly. Schools are about to reopen. With far fewer restrictions.

Someone asked me for a link, I should have included it, my fault, sorry. It's from here: https://web.iastate.edu/safety/updates/covid19

This, at Iowa State, is bananas.

And you can find links to all the studies I am talking about in this slide (and more!) here... 3/3

Here's a thread about the follow-up data from the V-safe pregnancy registry that I have just added to this summary... 2/

Updating my single-slide summary on the safety of #COVID19 #vaccines in #pregnancy to include new data from the V-safe pregnancy registry that came out while I was on holiday.... 1/

@NEJM joining the waning immunity debate. I'm going to push back a bit. Data from @UCSDHealth of vax effectiveness in health workers: 94% in June, 65% in July. Interpreted as "likely to be due to...delta and waning immunity over time, compounded by end of masking requirements."

Just adding that even though differences in sex weren't statistically different in this study, 65% of men seroconverted compared to 35% of women. This has come up in several other studies and needs to be researched further: https://twitter.com/ahandvanish/status/1387140392794079236… 7/

When paired with this other CDC study showing 65% are seronegative at 60 days, it looks like the *majority* of patients never seroconvert or serorevert early. This has big implications for care & research. Antibody tests can't be required for either! https://twitter.com/ahandvanish/status/1430614546356621315… 6/Quote Tweet

This also has implications for reinfections and vaccine efficacy as well - the study itself mentions that "seroconverters and nonseroconverters will probably also respond differently to vaccination." 5/

This study has huge implications not only for the many #LongCovid patients who have been denied care for not having antibodies, but also for how we think about protection from infection in society more broadly! 4/

B) Small viral load: people who had higher PCR Ct values (meaning the test had to run more cycles before returning positive) were also less likely to seroconvert. This implies mild and asymptomatic cases are less likely to seroconvert. #LongCovid 3/

They found two major risk factors that predicted a lack of seroconversion: A) Age: people under 40 are statistically less likely to make antibodies. #LongCovid 2/

New CDC study finds that 36% of COVID patients *never seroconvert*, meaning they NEVER make antibodies! #LongCovid This is a huge finding which we need to amplify broadly! Please retweet & send to providers, patients, support groups, #MedTwitter, etc. https://wwwnc.cdc.gov/eid/article/27/9/21-1042_article… 1/

um,

#GetVaccinatedNow

from the makers of ivermectin

Moreover, the case positivity is just getting worse and worse, nothing like this has been seen in over a year.

(and of course the true number of cases is much higher than reported given the high case positivity)

Cases may not be as high as in other states and vaccination (43% fully) is not as low as the gulf states, but this is the type of denial that lets COVID multiply.

Case positivity has reached this high even with a rise in testing.

Oklahoma - leading the nation in searches for ivermectin.

Oklahoma, this is bad. Case positivity - 81.98% Standing far above another other US state. 1/4

PS: PHE report here: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/990177/Variants_of_Concern_VOC_Technical_Briefing_13_England.pdf… Sanger data here: https://covid19.sanger.ac.uk/downloads I'm also kinda sick of people minimising this the whole time. The speed of spread has been so quick.

We need to stop obsessing about 21st June and start worrying about where we are *now* and how we can get cases down *now*. Oh and massively discourage international travel - not least to protect other, less vaccinated, countries from this variant. https://twitter.com/chrischirp/status/1398955136756617219?s=20… 7/7

Reported cases in England have risen 27% in last week. Hospital admissions to 26 May have risen 15%. Cases are going up steeply in many local authorities. https://twitter.com/julesmchamish/status/1399059877226504194?s=20… 6/7

There are still reasons why this might be higher other than just higher transmissibility - but note this estimate has got *higher* since previous report, as more data has come in. Also, SAGE's bad scenarios start from 40% more transmissible variant. https://twitter.com/Dr_D_Robertson/status/1396568757892489217?s=20… 5/7Quote Tweet

The PHE report last week also looked at "secondary attack rates" - chance that a close contact of a confirmed case will also test positive (NB contacts with no symptoms not necessarily tested) They estimate that person with B.1.617.2 is 67% more likely to infect a contact. 4/7

B.1.617.2 is now dominant in almost every region of England. Even NE and Yorks are catching up very fast. 3/7

Firstly, B.1.617.2 is now dominant in England. This data excludes traveller data and surge testing as much as possible. In absolute numbers, B.1.617.2 ("India") overtook B.1.1.7 ("Kent") about 15 May. PHE report estimates a few days earlier. 2/7

THREAD latest on B.1.617.2 variant in England: B.1.617.2 (1st discovered in India) is now dominant in England. Here is a thread summarising latest PHE report and Sanger local data. TLDR: it is NOT good news. 1/7