Coronavirus

[Cyberyeti]

I have noticed over the past week or two that every article on Covid vaccine effectiveness seems to top the "most read article" list on the BBC website. I would have thought the a single-number for effectiveness (e.g. today's headline --- "Janssen single dose Covid vaccine 66% effective") provides no informational value to the average (i.e. non expert) citizen.

 

The Covid vaccination centres are unlikely to offer people a choice, right? And even if they did, would the recipient be in able to make an informed choice? In other words, is the fixation on the percentages a needless distraction?

 

Would love to hear opinions.

 

I think all it means is that's another 30M doses coming in to the UK of a decent vaccine, the effectiveness number means little as long as it's decent. Also the ones that work in different ways may give options to some patients for those that can't take some of the others.

[cherdano]

I have noticed over the past week or two that every article on Covid vaccine effectiveness seems to top the "most read article" list on the BBC website. I would have thought the a single-number for effectiveness (e.g. today's headline --- "Janssen single dose Covid vaccine 66% effective") provides no informational value to the average (i.e. non expert) citizen.

 

The Covid vaccination centres are unlikely to offer people a choice, right? And even if they did, would the recipient be in able to make an informed choice? In other words, is the fixation on the percentages a needless distraction?

 

Would love to hear opinions.

One number??

We need efficacy against symptomatic disease, against asymptomatic infection, against transmission, against severe disease, against hospitalisation, against death. For old-fashioned covid-19, for B1.1.7, for B1.351. By age, risk group, gender.

 

Also, frequentist confidence intervals, CIs based on a Bayesian model with minimally informative prior, or CIs based on cherdano's model (Bayesian model with a "best guess" prior based on clinical Phase 1/2 results - antibody titers etc. compared to other vaccines). Now we are talking.

 

More seriously? Let BBC readers enjoy some good news, we had enough bad news to digest!

[pilowsky]

One number??

We need efficacy against symptomatic disease, against asymptomatic infection, against transmission, against severe disease, against hospitalisation, against death. For old-fashioned covid-19, for B1.1.7, for B1.351. By age, risk group, gender.

 

Also, frequentist confidence intervals, CIs based on a Bayesian model with minimally informative prior, or CIs based on cherdano's model (Bayesian model with a "best guess" prior based on clinical Phase 1/2 results - antibody titers etc. compared to other vaccines). Now we are talking.

 

More seriously? Let BBC readers enjoy some good news, we had enough bad news to digest!

 

Confidence interval - the gap between reality and fantasy.

 

This definition would certainly explain a lot about how politicians manage disasters.

[y66]

https://www.slowboring.com/p/good-vaccines

 

When you read in the press that a vaccine is 90 percent effective vs another one that’s only 70 percent effective, do you know what that means? It turns out that what everyone is measuring in their Covid trials is the share of people in the control group who develop symptoms vs the share of people in the treatment group. That focus on symptoms has a bad news aspect that’s been widely publicized — they measured symptoms rather than doing constant PCR tests so some of the vaccinated people may have had asymptotic infections, and it’s possible that vaccinated people can still spread the virus.

 

Less widely publicized is the good news: None of the people in the Pfizer/Moderna treatment groups died or even fell seriously ill and had to be hospitalized.

 

This is typical of vaccines. For virology reasons that I don’t really understand, flu vaccines have very low efficacy as measured in this way. One of the main reasons doctors recommended them anyway is that the immune system head start they provide greatly reduces the severity of flu infections even when it doesn’t stop them. These days they vaccinate kids against chickenpox, so kids mostly don’t get chicken pox. But even more remarkable, when they do get chickenpox these days it’s a “sick for a few days” kind of thing not “miss weeks of school while suffering in agony.”

 

This is a really big deal with regard to the lower efficacy we are expecting from the AstraZeneca and Johnson & Johnson vaccines. A vaccine that’s only 70 percent effective at blocking infection would be expected to generate a larger than that reduction in hospitalizations and an even larger reduction in death. Which is to say that especially as a solution for the non-elderly, a vaccine like that is actually really good. Giving everyone under 65 a 70% effective vaccine sounds a little lame, but would eliminate almost all the loss of life among the non-elderly and also take a huge burden off America’s hospital system. These vaccines are really good.

[y66]

The CDC ordered software that was meant to manage the vaccine rollout. Instead, it has been plagued by problems and abandoned by most states.

 

https://www.technologyreview.com/2021/01/30/1017086/cdc-44-million-vaccine-data-vams-problems/

[hrothgar]

Writing software is hard

 

Writing software that needs to be ready immediately and works perfect is especially hard.

 

The Obama administration got burned during the early phases of the Obamacare rollout.

I'm not surprised that we're seeing similar problems here.

 

Don't get me wrong. I'd like to be able to blame this all on Trump /Jared / whomever. However, I'm not sure that that's quite fair this time around.

[y66]

Writing software is hard

 

Writing software that needs to be ready immediately and works perfect is especially hard.

 

The Obama administration got burned during the early phases of the Obamacare rollout.

I'm not surprised that we're seeing similar problems here.

 

Don't get me wrong. I'd like to be able to blame this all on Trump /Jared / whomever. However, I'm not sure that that's quite fair this time around.

It may be premature to assign blame. You can't build decent software if the requirements suck or if the requirements are still changing or the runtime environment is unpredictable. But I don't think it's premature to say that Deloitte's software sucks or, given their history, that this is not surprising.

[pilowsky]

If you aren't blaming Trump and his band of alternate 'geniuses' then you are mistaken.

 

Everything is 'hard' when you don't know how to do it.

Countries - with actual functioning health care systems - have been giving vaccinations for more than 100 years. Even before software was invented.

Computers with hard drives that people could operate easily did not arrive until after 1990. Access to computers is still limited.

 

Neil Armstrong landed on the moon in 1969. The computing power available to him was equivalent to an HP-125

The Federal system for managing healthcare in the USA looks like it was designed by Kafka and implemented by Mervyn Peake.

Read Catch-22, The House of God etc to understand why the American system (I use the word advisedly) is a failure.

 

To distribute a vaccine you need is a little book with a few columns and a pen.

A steady supply chain would also be helpful.

 

What you don't need is software. Using software impairs access to poor people. In the USA the 'average Joe' is finding it damn near impossible to get a vaccine because of the 'software'.

 

In any event, I'm sure that when they get the injection the software that bills them will work just fine.

 

Except with Trump of course. He can't pay his bills because he's under audit, and someone called Eric is running his company for him. But Eric has been cancelled so don't hold your breath for anything useful to happen.

 

In any event, you seem to be suggesting that while they were spending time making the vaccine nobody in government thought that it would be a good idea to think of a way of distributing it.

 

The US Federal government is now so lacking in expertise that when Trump was rambling to Johnathon Swan the graphs he was misinterpreting were obviously downloaded from ourworldindata.org - everyone's go-to source on this Forum.

 

This means that Trump was relying for his advice on some staffer googling the internet to get help from a University in the United Kingdom instead of taking advice from the vast intelligence infrastructure specifically designed to help any administration cope with disasters.

 

The fact that Trump failed to appoint a Minister of Science (different name in the USA) for more than a year is emblematic of the catastrophe.

 

Not everything needs good software: interest, empathy, determination and perseverance are also useful.

 

Was Polio eradicated with software? How about Cholera in London.

[hrothgar]

 

In any event, you seem to be suggesting that while they were spending time making the vaccine nobody in government thought that it would be a good idea to think of a way of distributing it.

 

 

Actually, that is precisely what happened here in the US.

 

The Federal government almost completely ignored issues surrounding distribution (particularly what would happen once supplies were handed off to individual states)

 

This is what REPUBLICANS are saying

 

“That comprehensive vaccination plans have not been developed at the federal level and sent to the states as models is as incomprehensible as it is inexcusable.”

 

Just today, it was announced that the Federal Government "lost" 30 millions doses...

 

https://pbs.twimg.com/media/Eqp_aprVkAERND1?format=jpg&name=900x900

[hrothgar]

 

To distribute a vaccine you need is a little book with a few columns and a pen.

A steady supply chain would also be helpful.

 

 

How do you think that supply chain management is done?

 

Hint: It involves software....

[pilowsky]

How do you think that supply chain management is done?

 

Hint: It involves software....

 

How do you think it was done prior to 1990?

 

Hint: it doesn't involve software...

[pilowsky]

Actually, that is precisely what happened here in the US.

 

The Federal government almost completely ignored issues surrounding distribution (particularly what would happen once supplies were handed off to individual states)

 

This is what REPUBLICANS are saying

 

“That comprehensive vaccination plans have not been developed at the federal level and sent to the states as models is as incomprehensible as it is inexcusable.”

 

Just today, it was announced that the Federal Government "lost" 30 millions doses...

 

https://pbs.twimg.co...pg&name=900x900

 

Actually, That is exactly what I said. You seem to have that singularly American skill of being an 'irony-free zone'.

 

See those spaces

 

in between the lines

 

try to read them.

 

It will hurt a bit at first.

 

But then you will get better at it.

[shyams]

Actually, That is exactly what I said. You seem to have that singularly American skill of being an 'irony-free zone'.

 

See those spaces

 

in between the lines

 

try to read them.

 

It will hurt a bit at first.

 

But then you will get better at it.

This is utter bullshit used to backtrack from previous misstatements or poorly constructed points.

[pilowsky]

This is utter bullshit used to backtrack from previous misstatements or poorly constructed points.

 

I would not be the first to utter bullshit around here.

[kenberg]

The vaccine roll-out.

I'll say a bit from personal experience. I suspect my experience is common.

 

 

Background: I'm 82, Becky, my wife, is 73, we are both in reasonable health I am often told I am in great share for an 82 year old which is sort of giving with now hand and taking back with the other. I have spent more time tin doctor's offices in the last ten years than in the seventy years before that. A different order of magnitude really. I live in Carroll County Maryland. It's a short walk to get to see cows, it's a short drive to get into Baltimore. OK, enough background.

 

 

I filled out a form expressing interest in the vaccine for the county, and one for the county hospital. At first they were overwhelmed but within a few days they got it together and I have been acknowledged as in their list of people to get it. I am 1B (being over 75), Becky is 1C(between 65 and 75). The original statement, maybe 3+ weeks back, was that 1A people would be seen within a week and then they would start on the 1C. Oops. Not so.

But now it gets complicated.

 

Among my various medical interactions, I was involved with the University of Maryland Medical System, and that puts me in their database. I also expressed interest in the vaccine to UMMS and, last Monday evening, I got an email saying I could come in on Tuesday or Thursday. I got my first vaccine (Pfizer) on Tuesday. A very impressive operation, many people in a large area, including a large area where we were to stay for fifteen minutes after the shot to see if there was any adverse reaction. There wasn't. I will get the second dose Feb 17.

Ok, now I should get off of the other lists that I am on. I should, but as near as I can tell, I can't. Perhaps they have access to the UMMS database and can see that I am already with them? Maybe, but I doubt it.

 

And, as I say, I expect this to apply to quite a few others. We octos, most of us, get to know more than one or two docs. A bridge playing friend has been to Johns Hopkins for knee issues, he has now received the vaccine from Johns Hopkins. You get the idea.

 

A recent article in the Carroll County times spoke of how few people over 75 in the county have received the vaccine. They were relying on data from the county health person. But am I in his data set as someone who got it, from UMMS, or am I in his data set as someone who has expressed interest to the county health office but has not yet received the vaccine? I sus[pect it is the latter but do I know? No, it beats me. And the County Commissioners were concerned with the number of 0ver 75s that have not been vaccinated. No doubt there are some over 75s, perhaps quite a few, who have not received the vaccine, but I seriously doubt that the data that they have is (ok, are) accurate. Perhaps a long way from accurate. But I don't know.

 

As for Becky, she is 1C rather than 1B, and now the county is being somewhere between vague and pessimistic about when they will get to 1C. And she has not been involved with UMMS or any other such good luck.

Ok, as of yesterday, I do the grocery shopping, not Becky. We can cope.

 

I appreciate the vaccine, I really do, and I can appreciate that there are problems with a roll-out of this magnitude. Still, it's a little troubling. It's as if the vaccine Santa decided I have been a good boy but Becky has been a naughty girl.

 

This is not really a letter of complaint, just a statement of how things are from what I have seen up close.

[y66]

Very happy to hear you have joined the Vaccination Class.

[y66]

https://messaging-custom-newsletters.nytimes.com/template/oakv2?campaign_id=9&emc=edit_nn_20210201&instance_id=26625&nl=the-morning&productCode=NN&regi_id=59211987&segment_id=50744&te=1&uri=nyt%3A%2F%2Fnewsletter%2Fbbb05a80-af7c-5037-b53f-35ed99b5d44b&user_id=2d8b72dd84a9ff194896ed87b2d9c72a

 

The news about the vaccines continues to be excellent — and the public discussion of it continues to be more negative than the facts warrant.

 

Here’s the key fact: All five vaccines with public results have eliminated Covid-19 deaths. They have also drastically reduced hospitalizations. “They’re all good trial results,” Caitlin Rivers, an epidemiologist at Johns Hopkins University, told me. “It’s great news.”

 

Many people are instead focusing on relatively minor differences among the vaccine results and wrongly assuming that those differences mean that some vaccines won’t prevent serious illnesses. It’s still too early to be sure, because a few of the vaccine makers have released only a small amount of data. But the available data is very encouraging — including about the vaccines’ effect on the virus’s variants.

 

“The vaccines are poised to deliver what people so desperately want: an end, however protracted, to this pandemic,” as Julia Marcus of Harvard Medical School recently wrote in The Atlantic.

 

Why is the public understanding more negative than it should be? Much of the confusion revolves around the meaning of the word “effective.”

 

What do we care about?

In the official language of research science, a vaccine is typically considered effective only if it prevents people from coming down with any degree of illness. With a disease that’s always or usually horrible, like ebola or rabies, that definition is also the most meaningful one.

 

But it’s not the most meaningful definition for most coronavirus infections.

 

Whether you realize it or not, you have almost certainly had a coronavirus. Coronaviruses have been circulating for decades if not centuries, and they’re often mild. The common cold can be a coronavirus. The world isn’t going to eliminate coronaviruses — or this particular one, known as SARS-CoV-2 — anytime soon.

 

Yet we don’t need to eliminate it for life to return to normal. We instead need to downgrade it from a deadly pandemic to a normal virus. Once that happens, adults can go back to work, and children back to school. Grandparents can nuzzle their grandchildren, and you can meet your friends at a restaurant.

 

As Dr. Ashish Jha, the dean of the Brown University School of Public Health, told me this weekend: “I don’t actually care about infections. I care about hospitalizations and deaths and long-term complications.”

 

The data

By those measures, all five of the vaccines — from Pfizer, Moderna, AstraZeneca, Novavax and Johnson & Johnson — look extremely good. Of the roughly 75,000 people who have received one of the five in a research trial, not a single person has died from Covid, and only a few people appear to have been hospitalized. None have remained hospitalized 28 days after receiving a shot.

 

To put that in perspective, it helps to think about what Covid has done so far to a representative group of 75,000 American adults: It has killed roughly 150 of them and sent several hundred more to the hospital. The vaccines reduce those numbers to zero and nearly zero, based on the research trials.

 

Zero isn’t even the most relevant benchmark. A typical U.S. flu season kills between five and 15 out of every 75,000 adults and hospitalizes more than 100 of them.

 

I assume you would agree that any vaccine that transforms Covid into something much milder than a typical flu deserves to be called effective. But that is not the scientific definition. When you read that the Johnson & Johnson vaccine was 66 percent effective or that the Novavax vaccine was 89 percent effective, those numbers are referring to the prevention of all illness. They count mild symptoms as a failure.

 

“In terms of the severe outcomes, which is what we really care about, the news is fantastic,” Dr. Aaron Richterman, an infectious-disease specialist at the University of Pennsylvania, said.

 

The variants

What about the highly contagious new virus variants that have emerged in Britain, Brazil and South Africa? The South African variant does appear to make the vaccines less effective at eliminating infections.

 

Fortunately, there is no evidence yet that it increases deaths among vaccinated people. Two of the five vaccines — from Johnson & Johnson and Novavax — have reported some results from South Africa, and none of the people there who received a vaccine died of Covid. “People are still not getting serious illness. They’re still not dying,” Dr. Rebecca Wurtz of the University of Minnesota School of Public Health told me.

 

The most likely reason, epidemiologists say, is that the vaccines still provide considerable protection against the variant, albeit not quite as much as against the original version. Some protection appears to be enough to turn this coronavirus into a fairly normal disease in the vast majority of cases.

 

“This variant is clearly making it a little tougher to get the most vigorous response that you would want to have,” Dr. Francis Collins, director of the National Institutes of Health, said. “But still, for severe disease, it’s looking really good.”

 

What would an expert do?

The biggest caveat is the possibility that future data will be less heartening. Johnson & Johnson and Novavax, for example, have issued press releases about their data, but no independent group has yet released an analysis. It will also be important to see much more data about how the vaccines interact with the variants.

 

But don’t confuse uncertainty with bad news. The available vaccine evidence is nearly as positive as it could conceivably be. And our overly negative interpretation of it is causing real problems.

 

Some people worry that schools cannot reopen even after teachers are vaccinated. Others are left with the mistaken impression that only the two vaccines with the highest official effectiveness rates — from Moderna and Pfizer — are worth getting.

 

In truth, so long as the data holds up, any of the five vaccines can save your life.

 

Last week, Dr. William Schaffner of Vanderbilt University told my colleague Denise Grady about a conversation he had with other experts. During it, they imagined that a close relative had to choose between getting the Johnson & Johnson vaccine now or waiting three weeks to get the Moderna or Pfizer vaccine. “All of us said, ‘Get the one tomorrow,’” Schaffner said. “The virus is bad. You’re risking three more weeks of exposure as opposed to getting protection tomorrow.”

[cherdano]

tldr: Here are all the important vaccine trial results in one single table

[thepossum]

tldr: Here are all the important vaccine trial results in one single table

Agree about tldr

 

Here's another link of the trials and status that WHO knows about. There seem to be others.

 

Do we really need this many

 

https://www.who.int/publications/m/item/draft-landscape-of-covid-19-candidate-vaccines

[johnu]

Agree about tldr

 

Here's another link of the trials and status that WHO knows about. There seem to be others.

 

Do we really need this many

 

https://www.who.int/publications/m/item/draft-landscape-of-covid-19-candidate-vaccines

The clearly obvious answer is Yes, we needed that many.

 

While there was optimism about developing a covid vaccine, there were no guarantees that any one company would be successful. For example, last week, Merck announced they were discontinuing development on 2 vaccine candidates because the results were disappointing. At least of couple of the Chinese vaccines may not even be as effective as the minimum CDC standard of 50% effectiveness, so would be considered failures if that is true. Japan is developing a vaccine but it isn't likely to be available until 2022, if development even continues. I'm sure there are other failures that have not been as well publicized.

 

Some vaccines may not be approved until much of the world population is already vaccinated, which really matters as the infections are getting out of control world wide. It matters that we have started vaccinating millions of people, even if the situation is still a disaster.

 

In different scenarios, the handful of vaccines that are currently in emergency use may have been failures, and some other approach may have been the successful one but wasn't taken because somebody decided that we would put our faith in just a handful of candidates. So yes, we needed to explore as many possible vaccines as we did.

[pilowsky]

Trials serve multiple purposes.

 

 

The 'sacred' purpose is to ensure that a method does what it is meant to do and does no harm.

And, if it does harm, the harm is acceptable ("I heard that the level of political killings in El Salvador is acceptable - My question is Acceptable to whom" JK Galbraith)

 

But there is another reason that big Pharma do trials. First, they want to put their compound into the hands of 'opinion leaders' - an old fashioned term for social influencers I guess.

These people - who run the trials - then ensure that their colleagues use the compound.

Big Pharma is not a 'virus' it really does like it when its product works. But more important to them is that it sells.

 

This may be less obvious with a vaccine, but I have no reason to doubt that it still operates.

 

After the compound is in the community comes the post-marketing surveillance.

 

As one Professor remarked when I was training - every patient is in a trial.

 

Good post-marketing surveillance should IMHO be part of the original trial.

 

Unfortunately, this approach doesn't suit the current paradigm where academics get 'likes/citations' for individual publications.

 

Science, in general, has failed to keep up with technology.

[thepossum]

Can someone please explain the apparent lack of variance between the effectiveness of the vaccines at Phase 3 so far :)

[barmar]

How do you think it was done prior to 1990?

 

Hint: it doesn't involve software...

It's true, they eradicated polio without software to manage the vaccination effort.

 

But things are different now. The population is much larger, it's an international effort, some of the vaccines require fancier technology to transport and keep.

 

It's not fair to compare different time periods. You can't say that personal computers and refrigerators are not necessary because they didn't have them in the 19th century.

[pilowsky]

It's true, they eradicated polio without software to manage the vaccination effort.

 

But things are different now. The population is much larger, it's an international effort, some of the vaccines require fancier technology to transport and keep.

 

It's not fair to compare different time periods. You can't say that personal computers and refrigerators are not necessary because they didn't have them in the 19th century.

 

Polio existed in the 1800's true. Polio was not brought under control until the 1950's and '60's. There are two vaccines - the injectable one used in America was first and was very effective. I had the oral Sabin vaccine - a pink liquid on a plastic spoon: delicious.

 

You may have read it, but I recommend Waldrop's book "The Dream Machine; JCR Licklider and the revolution that made computing personal".

 

I started University in 1976 there were no personal computers.

 

Neurotrophic viruses are still a problem. I even co-authored a paper about one of them.

[cherdano]

Can someone please explain the apparent lack of variance between the effectiveness of the vaccines at Phase 3 so far :)

On some level, they all do the same thing: somehow produce copies of the coronavirus spike protein inside your body, so your immune system learns to fight it.

In the case of mRNA vaccines, they just spam cells in your body with the RNA code to produce the spike protein. (This is part of the normal replication process of viruses within the body - hijacking your cells with their RNA to reproduce copies of the original virus.) In the case of AstraZeneca, Sputnik, Johnson & Johnson, they changed the genetic code of another (harmless to humans) virus so that it also includes the coronavirus spike protein. In the case of Novavax, they basically grow lots of spike proteins in their factories, which you then get injected directly.