Covid 8/18/22: CDC Admits Mistakes

Two Covid-related things happened this week that I did not expect.

The CDC admitted that it had failed us during the pandemic, withdrew at long last many of its remaining recommendations and promised reforms to be less academic and otherwise do better.

A new paper found potential biological markers for Long Covid, claiming it can be identified via tests that match patient self-reports almost all the time. This points towards potential progress in treatment, and more generally in Long Covid being much more concretely A Thing that might exist and could be reasoned about. There are still flaws, and even without flaws there is still much work to do here.

Most of the reasons not to be concerned remain, so up front: I do not think that this should substantially change anyone’s level of Covid precautions.

Executive Summary

1. CDC finally fully gives up on six foot social distancing and other measures.

2. CDC admits some failure, promises reforms that seem potentially promising.

3. New study finds potential biological markers for Long Covid.

Also I think someone said something about over the counter hearing aids?

Let's run the numbers.

The Numbers

Predictions

Prediction from last week: 650k cases (-5%) and 3,200 deaths (+0%).

Results: 602k cases (-12%) and 3,183 deaths (-1%).

Prediction for next week: 560k cases (-8%) and 3,200 deaths (+1%).

I do not know why cases declined more than expected but result seems robust so I see no reason to expect it not to continue at least somewhat. There won’t be enough time for the decline to impact deaths yet, so I’m mostly going with the null prediction there. There is not much meaningful uncertainty here.

Deaths

Cases

Decline cuts across all four regions. We are fully into the BA.5 era with nothing on the near-term horizon to replace it, so things should be quiet for a few months at least.

Physical World Modeling

A guide to buying the right HEPA air filter.

Bob Wachter sees no sign of anything that might replace BA.5. Thread also reminds us that it is a common mistake not to take into account the correlation between the Covid status of people who choose to be together in a group. Trevor Bedford however sees logistic growth in BA.2.75, although only with R0 ~ 1.3 (versus ~1 for BA.5) which isn’t that much of an advantage for a new strain taking over. My guess is Trevor is right and BA.2.75 will displace BA.5 over time, but that we will barely even notice.

Some sad news.

I opened a Manifold market on whether she’ll experience a rebound, to see how common people think such events are.

This also brings up the question, what happens when you’re against actual physical world modeling and instead blindly follow CDC rules?

They do know he had Covid-19 plus a rebound within the last few weeks? That it is absurd to think that being a ‘close contact’ puts him at relatively high risk for Covid-19 given that timeline?

No. Of course not. This is not a man to concern himself with whether or not an action makes physical sense.

Almost no children under 5 are getting vaccinated, even weaker ‘than experts feared.’ Doses are being discarded due to lack of demand. This should not be a fear so much as a revealed preference. If we had approved these doses sooner, I am guessing we would have had much higher uptake, although still nothing that would have satisfied experts. At this point, people don’t care enough, especially given the logistics are frequently annoying.

This raises the question of why we insisted in so many crazy precautions for these same young children for so long, in ways that I strongly believe did serious damage to their development and well-being. They were never at risk and everyone knew this well enough not to bother doing much about it when finally given the opportunity.

Mina says if you had Covid last month, hold off on your booster. Seems right.

The Long Long Covid Saga

I noted last week that I didn’t get a chance to look at this (paper):

It has been two years, and finally someone is lining up biological markers with patient reported outcomes. That is a big deal. The fact that this, as far as I could tell, hadn’t happened yet was certainly rather glaring. This is saying in its key findings that yes, we can prove physical ways in which there is a difference between those with and without Long Covid, independent of what they tell us but correlated to it.

Another box checked. It never made any sense to me that, to the extent Long Covid is real, it wouldn’t be heavily correlated with case severity.

I don’t know how ‘clean’ clusters need to be to count as clusters, but this doesn’t seem to me like an impressive clustering, for whatever that is worth.

So to be maximally non-technical the theory is that Long Covid is the result of a screwed up immune response, as has been commonly theorized.

Skipping a bit for space.

This is certainly still much more interesting findings than we have seen before, holding out more hope for a real explanation. My chances of something real happening some of the time has gone up.

It is still small sample size and I’d still like to see it replicate before I draw too many other conclusions. All the other calculations about how sufficiently large effects would show up in population statistics are still binding.

Next, do out of sample tests to confirm?

While importantly doing the study while also getting measurements before infection.

I realize that is asking for something pretty difficult at this point in the pandemic, or at least tricky. And it puts you back at square one plus time required for infections. But yeah, that’s what it is going to take, I think, at this point, to rule out the usual correlation versus causation issues fully.

I could be convinced to look into this more by a sufficiently high bid of various kinds.

I also realize that’s all at least somewhat of a cop out, but on reflection I am comfortable with that, at least for now. I do encourage those who understand the science here to chime in for comments.

Most important is practical takeaway - should you change your behavior? My answer there is still essentially a no, not a substantial amount.

In other long Covid news, from Nature, still no treatments for Long Covid. It turns out if something is amorphous and of unclear amount of physical reality, and presents as a grab bag of lots of different problems, it’s hard to figure out what to try doing about it. Which, if this was as big a deal as many claim, should not have been stopping anyone - better to try pretty much actual everything we’ve got and see what sticks. That, alas, is not how we work, but the lack of urgency here reveals pharma’s perspective.

Masks On, Masks Off

CDC more or less says pandemic over, move along, nothing to see here.

(CNN) The US Centers for Disease Control and Prevention says the nation should move away from restrictive measures such as quarantines and social distancing and focus on reducing severe disease from Covid-19.

In new guidelines released Thursday, the agency no longer recommends staying at least 6 feet away from other people to reduce the risk of exposure -- a shift from guidance that had been in place since the early days of the pandemic.

…

The new guidance also does not advise quarantining people who've been exposed to Covid-19 but are not infected.

The nothing to see here, alas, still kind of includes your face.

But the guidance does keep some measures the same. It encourages testing for people with symptoms and their close contacts. It also says people who test positive should stay home for at least five days and wear a mask around others for 10 days. It also continues to recommend that people wear masks indoors in about half the country.

I am confused why one would continue recommending that.

One constant is that any move by the CDC will always be seen as a move aimed at regaining the public’s trust:

Chin-Hong thinks some states, like California, will continue to go beyond the CDC's guidance in their own recommendations, but by and large, he thinks these reflect the prevailing attitudes toward the pandemic. He sees it as a move by the CDC to try to regain the public's trust.

Another constant is that most moves will be seen by the public as damaging to their trust, since even when it is the right move it highlights older incorrect moves. I consider it important in such situations not to use ‘we made the situation better via doing less to make the situation worse’ as an opportunity to say ‘but you made the situation worse though.’ Incentive compatibility is crucial. A little bit of ‘buy you are still making the situation worse even if you’re doing it less’ still is necessary at times.

Another common theme is the worry that perhaps no one will listen to them ever again.

"What the CDC is, in my opinion, trying to do, they are trying to still be relevant, and maybe when they say something, people will listen to them instead of being completely 180 degrees away from what behavior is anyway," Chin-Hong said.

Fair enough, although if that was the goal I wouldn’t continue to recommend indoor masking.

School-based advice improved markedly.

The agency removed the recommendation that kids in different classrooms avoid mixing, a practice known as cohorting. It also removed advice that kids who are contacts of someone who tested positive for Covid-19 take regular tests -- and test negative -- to remain in the classroom, which was known as test-to-stay.

Once again: Resist the temptation to incredulously go ‘they only now removed this crazy restriction’ and instead be happy the crazy restriction is lifted.

Most of all, please please do not go around using the inconsistency in such situations as an opportunity to dunk on those involved. Seriously, don’t do this:

It is vital that we are able to take yes for an answer from people who have previously answered no. What changed? Their minds changed. The social and political situations changed. The path of the pandemic changed. People came to their senses. We need to ensure they are happy with their decision to do this.

Eric later offered this thread of clarification that things did change and also of what he didn’t want, but what he does want are deep, painful hearings about why the reversal wasn’t done sooner, a call prompted by this, but this totally doesn’t mean the CDC can’t change its mind. Except when it does this too late, yes, well. No line of retreat, then.

What he actually centrally wants is the very reasonable ‘let dissenters dissent and reason with them’ and also the government making better decisions faster. I am certainly all for both of those. And of course there is always a ‘if not now, then when?’ response to those who seem to continuously say ‘now is not the time’ on such matters. I sympathize, but I believe the point must stand.

CDC also seems to have ditched the six foot rule in favor of reasonableness.

The correct response, again, is ‘thank you for your good decision.’

Next Time, CDC - Next Time

CDC also quietly acknowledged that their response to Covid-19 ‘did not meet expectations,’ and there will perhaps be an attempt to do better next time.

“For 75 years, CDC and public health have been preparing for Covid-19, and in our big moment, our performance did not reliably meet expectations,” Walensky said in a statement. “I want us all to do better and it starts with CDC leading the way. My goal is a new, public health action-oriented culture at CDC that emphasises accountability, collaboration, communication and timeliness.”

The whole post is a clear case of the very good news of an admission of bad news. Scott Gottlieb approves and has a good summary.

I agree that these are positive, badly needed changes.

Despite that, at least for now, no, we are totally doing the same crazy stuff over and over again and if this doesn’t make you want to throw things you need to work out to improve your throwing-things capacity.

Not that you have to respect that decision, of course (direct link). Would be a real shame if everyone in position to monitor the situation knew how to monitor it.

In Other News

UK goes first and approves Omicron-specific Moderna vaccine. Better too late than even later, I suppose.

Vaccines are great, and there is some truth to the claim that if you still test positive but have a mild case then the vaccine worked, but also you do not know the counterfactual so you don’t know that anything changed. It is not reasonable, when the vast majority of outcomes are already good outcomes, that a good outcome after being vaccinated means ‘the vaccine worked’ in a way that implies that the vaccine should get the credit. It’s not like half the time unvaccinated people get seriously ill when they get Covid-19, especially now when most have been infected before.

Where to get tested for Covid-19 these days in NYC. The test-to-treat where you can get Paxlovid right away are pretty great, and there are free at-home test kits available. The tents offering testing are still everywhere, presumably because they are good business (and given what I saw on my insurance forms after using them they are very good business) and there is almost never a line anymore. They require insurance now.

This, except as a warning.

Prize for ‘Good Billionaires Index” up to $700.

A thread of threads of calls for Doing More because The Pandemic Never Ends.

Latest call for investigation into alternate possible origins for Covid-19. I strongly agree at this point that we do not know the origins of Covid-19 and that this is due in large part to a failure to investigate. And that we continue to do exactly the types of work that could damn well create another similar pandemic in the future, with many things in the same reference class being far, far worse if they were to happen.

In the absence of better information, we must treat such research as if there is a decent chance it created Covid-19.

If you say that this level of risk is acceptable in order to advance the state of human knowledge in important areas or to develop worthwhile technologies? That we should value improving and saving the health and lives of humans?

All right, punks. Deal. Your terms are acceptable. Let’s take that principle and apply it to every rule and regulation and restriction you place upon every form of research, of science, of engineering, of doing anything that improves or saves people’s lives, anywhere, for any reason.

Otherwise, or actually either way, let me humbly suggest something not to do.

At a minimum, I suggest the Hollywood test, which is where you ask ‘if this was the opening scene in a proposed Outbreak 2 would you consider the people involved too stupid to be believed by the audience and ask for a rewrite?’ and then if the answer is yes then do not do that then you f****** genocidal morons.

Will MacAskill has been hyping germicidal UV on his book tour and it is known to physically work. We don’t know if it is a good long term plan given potential long term effects because we haven’t bothered to figure it out, but getting sick also has long term effects, so this is definitely understudied.

Cuba didn’t do so well with the pandemic. The replies are full of mood affectation, but yes, if you go with a vaccine that does not work well your pandemic had a bad time. Seems hard to draw too many other conclusions about Cuba given that baseline.

As an avatar of the “safety” mindset: Things that make you go hmm.

Nothing… here… is… unsafe. Is the complaint that too many people want to travel via airplanes, by far the safest mode of travel known to mankind? Is it that when there wasn’t enough staff they slowed things down to keep it that safe? Is it that if we didn’t allow people to work there wouldn’t be such a high demand for labor, and workers not allowed to work don’t count as being out sick? Is it that if we’d Done More less people would be out sick tonight and that would make it all worth it? Is it ‘now look what you made me do keeping all these people home due to exposures’ somehow? Could be anything.

A fun little story of musical chairs.

I am weary of the tendency to link every stupid thing we did to every stupid thing someone later does, and the polio crisis has been coming for a while, not merely since 2020, but this runaround is all quite insane.

You know what is even more insane than playing musical chairs to get the kids into the office? The fact that if the wrong person noticed you doing this in another context, you could get your kids taken away from you.

Monkeypox

Post on the pros and cons of the ACAM2000 vaccine.

An MSM study in Paris, France found 13 of 213 completely asymptomatic men (6.5%?!) tested positive for Monkeypox.

However, note that this was not a randomly selected MSM group, it was those who were already getting HIV medication. In that group, the symptomatic infection rate was, shall we say, rather high.

During the study period, 706 MSM visited our clinic, 383 had symptoms suggestive of MPXV infection (40% had anal lesions), and MPXV infection was confirmed in 271 of those with symptoms (Table).

That’s more than half of the population being somewhat symptomatic, and that half having a >50% positive test rate. Thus, even though a 6.5% asymptomatic positive rate sounds scary, this still has ~90% of the positive tests being from someone symptomatic. Given that the symptoms help spread the virus quite a lot, that means that almost all the actual risk continues to be in those who have clear symptoms.

All the questions at the news conference were about Monkeypox, not Covid-19. A full sweep would have been surprising, but there were indeed a few Covid-19 related things shared. Dr. Ryan seems a bit too surprised here.

On to the actual answers:

Stop. Math time. 12 deaths in 35,000 cases is a case fatality rate of 0.03%. Given the state of testing, the IFR is going to come in quite a bit lower than the CFR. The situation in France where testing was adequate is the exception, not the rule.

While this still is not nothing, this once again emphasizes that as an individual you are not in that much danger statistically speaking. Yes, we may have many doublings ahead of us, but this has so far killed about one person in a billion.

Always keep this front of mind when you see things like “Inside America’s rapidly unfolding monkeypox crisis — and the mistakes that made it worse.”

Once again, if you are not in the MSM category, major behavioral modifications simply do not make any sense, and any stress or worry is misplaced.

From a different Kai thread, before we resume, some numbers to drive it home, note that the one case in a child was retracted:

I continue to not understand the vision that those nations with more resources and ability to spend money don’t get more and quicker access to scarce resources than those with less resources and less ability to spend money.

It confuses the issue when there is indeed something urgently needed, as in:

Starvation is very bad. We should prevent it, and avoid complicating the issue by comingling it with concern over vaccines for something that has killed a total of 12 people being inequitably distributed. Keep eyes firmly on the ball.

Endemic is a state of mind?

I notice this says ‘when’ rather than ‘if.’

More suggestion that those with monkeypox isolate from their dogs. Plausible, although I continue to suggest we first isolate those with monkeypox from other humans.

Should you worry about getting monkeypox from thrift clothes? It’s not literal zero risk, but it can be fixed if anyone bothered to wash the clothes. If you aren’t confident a place would do that before selling them to you, I wouldn’t have been excited to be buying anything there in the first place.

We are not using the vaccine distribution system we got for Covid-19 for the monkeypox vaccine, and the whole thing is quite the giant mess. Also it seems no one has any idea how to ‘take a fifth of a dose out of a vial’ as this is beyond our powers of measurement.

Math Is Hard

Exhibit one, people choosing between 50% chance of 50 million dollars versus 100% chance of 1 million dollars, two times. If this happens to you, say ‘hang on’ and you call me first. I am confident we can work something out.

I do realize that if you can’t make deals at all and things are sufficiently bad that the answer is technically non-obvious, but also the answers included ‘flip a coin first and then take $1 million if you got the $50 million the first time.’ So yes. Math hard.

Exhibit two, dads not teaching their children so well. The CC is covering up “MEN.”

If dads are +28 while all men are +4, what percentage of men do you suppose are dads?

Yet I saw this several times and this made it onto national television.

China

This, too, is a choice.

The continued obsession with surfaces and other not-especially-relevant secondary vectors of infection indicates an unwillingness to learn or to optimize actual Covid-19 prevention over looking like you’re doing Super Hardcore Covid Prevention. Testing fish and blaming this on ‘trade with foreign boats’ is a distinctly lower threshold of sanity.

This points out a strange dynamic.

If you knew what techniques work to prevent Covid-19, there would be hope that you could take reasonable actions, or at least boundedly expensive actions, and prevent Covid-19. It would still be a bad idea because the price is too high, but you could do it.

If, however, you have no idea what works to prevent Covid-19, and your system is incapable of learning which techniques work, now you have to do vastly more to get the same result. Since containment is all-or-nothing, that means either going straight to crazytown, population you, or giving up. If you can’t give up for whatever reason, then, yes. Well.

Not Covid

Paper: Car seat requirements as contraception, for every death prevented you prevent 150 births.

Using census data and state-year variation in laws, we estimate that when women have two children of ages requiring mandated car seats, they have a lower annual probability of giving birth by 0.73 percentage points.

I often hear statements like ‘there is no way for government to do anything that substantially increases the birth rate.’ Well, at a minimum we can stop decreasing the birth rate. Also, whenever there’s a number like this that can vary a lot for many reasons, and you tell me it’s easy to lower but impossible to raise, I am not going to believe you. What else is super low hanging fruit?

PredictIt is being shut down. I was going to write a full post on this but Scott Alexander and Chris from Karlstack covered all of it and there isn’t much more to say.

Why it is important, in order to get ahead in Russia and many other places, not to show any beliefs at all. See Motive Ambiguity.

Polio continues to be back thanks to insufficient vaccination levels. If you need to, get vaccinated.

Paper claims that often used random number generators that look fine are sufficiently biased that they mess up chemical Monty Carlo simulations. Proposed solution is to test a source of randomness by simulating well-understood results via Monty Carlo simulation to see if you get the right answer. Which I love because that answer seems like it has to not be The Way but I suppose it would work.

The case for California state capacity as an EA cause area. I am continuously dismayed by the lack of emphasis on American prosperity, good governance and state capacity in such places, so I can’t complain too much about the additional bit of locality here via the focus on California. I am more interested in the federal level, but California does have unusually terrible governance and state capacity, and it is large. I have reached out to the author for the promised more concrete details, although of course I am not so focused on California.

DC childcare workers required by law to have college degrees, presumably so that when liberal arts majors end up working in daycare they can tell themselves they are not overqualified? I will accept DC merely doing this to itself rather than imposing such a requirement on the rest of us. Pretty crazy to tell people without college degrees they aren’t qualified to take care of kids, a message they can understand - not going to college does not mean you are stupid. Also, it seems the break-even number of children where ‘hire a human’ being cheaper than daycare is… two?

SBF explains ‘the financial circle jerk.’ Not investment or medical advice. Pass the salt.

Did you know cholesterol is correlated with bad outcomes? Here’s another study to that effect that tells us exactly nothing but will be used to tell you to have less joy.

How to solve a Rubik’s Cube via learning 3,915 algorithms for the final layer. I notice I am more confused that this didn’t happen earlier than I am impressed or surprised that it happened now. I mean of course that’s the way to do it, right?

Thread: Against against flirting. Also, cause area: Bring back classic OKCupid.

DeSantis shows us the path to occupational licensing reform, perhaps. The replies tell us why we should not hold our breath for occupational licensing reform.

All-male teams do not suffer reduced effort when teammates dislike each other, but all-female teams do, as do all members of mixed teams:

We study the effect of likeability on women’s and men’s team behaviour in a lab experiment. Extending a two-player public goods game and a minimum effort game by an additional pre-play stage that informs team members about their mutual likeability, we find that female teams lower their contribution to the public good in the event of low likeability, while male teams achieve high levels of co-operation irrespective of the level of mutual likeability. In mixed-sex teams, both women’s and men’s contributions depend on mutual likeability. Similar results are found in the minimum effort game. Our results offer a new perspective on gender differences in labour market outcomes: mutual dislikeability impedes team behaviour, except in all-male teams.

This abstract hints at a possible ‘burying of the lede’ but even setting that aside this is a clear case of multiple equilibria, but also interesting incentives. Note that there is no sign in any of these groups of increased effort when teams did like each other. Thus, not knowing how much you like your teammates strictly dominated knowing.

The all-male teams did the best by ignoring this never-helpful information. Well played.

It is greatly frowned upon and mostly illegal to do so, but this suggests that if you have men who dislike each other but most work together there is an obvious step one might take.

Another suggestion here is that the bigger the group, the greater the probability of dislike reducing effort, thus one should expect higher effort on smaller teams, which is indeed actionable.

An ongoing concern.

Good News, Everyone

The FDA will now allow hearing aids to be sold over the counter. The FDA announcement on this is good, but it would be much better if they realized the result generalizes. Something about a Simpsons blackboard with the line “I will not ask why the FDA required this before” written on it. This only took five years after a bill requiring the FDA to do this, and over one year after Biden signed an executive order for them to do it. They might even be available as early as mid-October.

By report of a former CIA director, there is no one worse out there anywhere in the world than today’s Republicans, and we’re holding up all right, so the world is safe? And also, if you want to beat them in elections, look how much room there is for improvement on your messaging.

That does, of course, go both ways.

Good debate, everyone.

I used to think it was the height of hubris to think I knew better than the people who did a given thing for a living, especially at high levels of high stakes. Man, was I wrong.

The cookies are important. I recommend Levain Bakery.

Permitting reform is in the works, and look at what we could do if we got some permitting reform instead of our usual not reforming and tacking on of protectionist requirements that make construction costs prohibitive.

The first thing we’ll do is give all the lawyers brain implants?

I too did not expect this to be the first use of the technology. The safe bet is always porn. What is being proposed here for lawyers implies a transformational improvement

Israel and Turkey resume full diplomatic ties.

Game ad shows correct gameplay for once, customers lose their minds.

Germans, amid Google searches for firewood, now support keeping their existing nuclear plants open 67-27 and are now going to do that.

Next up, they closed three other reactors a few months ago, how about we reopen those?

If we want to expand wind farm production, we don’t have to spend money, all we have to do is make it legal to build them. Padme says we’ll be doing that, right?

A reminder that you can and should do various stuff without worrying too much about whether it is the best possible stuff to do, and that comparative advantage is a thing.

Some of you should do one thing, some of you should do the other. The world would be a better place if this type of public advice from people like Sam Altman and Paul Graham was followed more often, and those that did follow it were more rewarded and funded than my model says they are, including by those same people. Failing that, it’s still better to favor Doing of the Thing, even if it’s not the Best Possible Thing. Much, much better to actually Do a Thing for real that you find interesting than forcing yourself to fake-do the theoretical Best Possible Thing. It’s totally fine to go your own way, don’t let anyone tell you different.

There is context. I prefer to ignore it.