You (finally!) no longer need to hand anyone a negative test in order to travel internationally.

You also most definitely do not have to hand it to the FDA for their work solving the bottlenecks they created.

Paxlovid remains something that you will probably not by default get, but you can get it if you are sufficiently pushy. Public health has the opinion that if you are low risk, a large (51%?) reduction in your relative risk isn’t even worth continuing a study, so that tells you how much you should care about otherwise reducing your risk.

Executive Summary

  1. You can get Paxlovid and/or travel without testing if you want to, but you still can’t quite get permission to vaccinate your young child.
  2. Don’t leave your friends behind.
  3. Even if they don’t dance, they’re still friends of mine.

Let’s run the numbers.

The Numbers


Prediction from last week: 700k cases (+7%) and 2,300 deaths (+4%).

Results: 644k cases (-2%) and 1,917 deaths (-14%).

Prediction for next week: 650k cases (+1%) and 2,000 deaths (+4%).

As is suspiciously often the case, cases and deaths moved unexpectedly in the same direction, causing suspicion of a reporting issue, which was at least in part Pennsylvania not reporting. I’ve given the state the same numbers as last week, at which point things overall seem closer to stable.

That means two major states (Florida is the other) require adjustments due to declining data quality. That’s a pretty bad sign for our state capacity and competence.


This looks a lot like a combination of a slow decline and some sort of reporting hiccup I haven’t yet fully accounted for.


This includes a proper Florida adjustment, so it looks like the South is at least mostly stable.

New wave coming on in the UK with R0 up in the 1.4 range. Scary on its face, but I don’t expect things to get out of hand.

Travel is Back

Tests are no longer required to enter the United States. We were far behind the sanity curve on this one, as Europe has long since gotten over the testing requirement.

Canada is also lifting its requirements.

In other news, Justin Trudeau has Covid-19 (as does Tony Fauci), which is interesting timing. Remember when we used to worry when world leaders got Covid? Worth noting ‘following health guidelines’ means he’s likely not getting Paxlovid, which is pretty crazy. Mild symptoms all around.

United saw a 7% increase in international flight searches, both into and out of the United States, within the first 72 hours after the requirement was lifted. I can see this being a temporary bump or the extra searches mostly being idle curiosity. I could also see this being a lowball of the impact due to many people not yet knowing.

The testing requirement is super annoying. Getting a reliable test within the required window was often not only expensive but logistically annoying and stressful. If you did test positive, it could get that much more expensive. I was invited to two international trips, neither of which I could have gone to anyway due to the new baby, but if it wasn’t for that there’s a substantial chance the testing requirement would have tipped the scales on one of the trips.

Italian mask mandates on trains have enforcement proportional to distance travelled. Interesting range of possible explanations is available.

Prevention Prevention

It looked like NYC was going to require masking at (checks notes) outdoor pools? On the 13th the official website still said this:

I am at least glad they were not to be worn in the water. Later that day, they did update the rule to something reasonable.

I am willing to give some benefit of the doubt that this was ‘didn’t get around to updating guidelines’ and thus only treat this as a 2021 problem, but it was still pretty nuts even then. It also seems a lot like the extra attention shined on the situation may have caused a quick change, making it seem like a good idea to keep highlighting crazy restrictions that still remain.

If you’re wondering, yes, it’s still an ‘emergency’ here in New York, the emergency powers got extended yet again.

The Only Reason You Have to Hand it to the FDA is They Ruled That You Have to Hand It to Them

Which is why no, beyond the amount we are required to do so by law we absolutely do not have to hand it to them.

Requiring a huge amount of work to be done that doesn’t need to be done, and then doing it, is a cost. It is not a benefit, nor does this require that ‘it’, in any form, be handed to those in question. All of this was always very straightforward and should have been completed a long time ago, even in the world in which it was required in the first place.

The good news is this means I still got it. I am prepared to hand it out if I have to.

Paxlovid Chronicles

Paxlovid is often available if and only if you push for it hard enough. The pharmacists still can’t prescribe it, but sometimes the doctor will give you the prescription in fifteen seconds.

Paxlovid use among states correlates with vaccination rates, implying the main mechanism for differences is patient demand.

Meanwhile, hundreds of people die each day of Covid, and most people who could benefit from Paxlovid continue not to get it.

Also from same source, Paxlovid supply increased when it looked like demand would go up. We do not know how hard a cap there is on Paxlovid supply from lack of material components, but capitalism often solves such problems if we are willing to pay. We are however completely unwilling to pay. That’s going to be a problem.

How’s it going, Zeynep Tufecki?

A correct but not as hot as he thinks it is take from Nate Silver.

At this point I consider ‘we are only interested in Sacrifices to the Gods and don’t care about actually effective options’ to be a highly cold take. Win-win solutions are not something anyone with power believes in or cares about, so they get neglected. Or alternatively, they are used as excuses for everyone to try and leverage them into win-lose outcomes, and with all sides doing that nothing gets done.

I do also agree the whole thing is aggravating. I don’t consider the problem to be Covid-specific. To the extent it is Covid-specific, I accept the ‘do actual nothing’ compromise. The problem is that elites and those with power only see the zero-sum world. They do not believe in win-win, do not prefer better things or outcomes to worse things or outcomes.

The very idea of trade or useful production is disdained, and if someone is winning someone else must be losing. This is thus something that, reversed from its original intention, often passes for wisdom in these contexts.

Let that sink in. “There are no solutions.”

You can never even hope for a solution. To anything.

This is not what Sowell meant in the original context of this quote, or what I think Mandrola intended here, which was a warning to not ignore the costs of our actions (and inactions) and to try and minimize them. Which is a good idea. Some people very much need to hear a version of ‘there are always trade-offs’ and to remember not to neglect downsides.

But the version of this idea used by the powerful is now not only generalized far out of the original context, it is centrally the opposite of Sowell’s warning. Sowell was warning that we need to be careful to pick the choice with the biggest upsides and smallest downsides – he asks ‘at what price?’ so we ensure the price is a cost worth paying. The common interpretation among those with power is that upsides for me must always equal downsides for thee – that’s what it means to people that there is a trade-off – so the price is a benefit.

Thus, under that thinking, if something is claiming to be a solution, a win-win that does not require a trade-off but instead is a strict Pareto improvement expanding the production possibilities frontier – or in normal English, it’s strictly better than not doing it, it’s a good thing, everybody wins, yay, celebrate, good times, come on and just do it already – then no, the world does not work like that. It must be fake. It’s a disapproved-of scam, and we should only do scams that take advantage of other people instead.

Until that changes, getting funding for things like pandemic prevention will at best be like pulling teeth. Occasionally circumstances will allow it, when things are sufficiently dire, but by default no such luck.

The Paxlovid Paradox

It’s amazing how clarifying it is to have a cure for Covid-19.

What’s the original source?

June 14 (Reuters) – Pfizer Inc (PFE.N) said on Tuesday it would halt enrollment in a trial for its COVID-19 antiviral drug, Paxlovid, in standard-risk patients after a study revealed the treatment was not effective in reducing symptoms in that group.

The drug has emergency use authorization for high-risk groups in which it has been effective in reducing hospitalizations and deaths.

The new data, however, showed a 51% relative risk reduction in standard-risk groups, which the company said was not statistically significant.

The standard-risk population usually includes people who do not have health conditions that put them at risk of severe disease and who can recover without the drug.

A 51% relative risk reduction sounds like a big deal. If it’s not ‘statistically significant’ yet then the logical thing to do is keep enrolling patients until the sample size gets big enough to fix that.

If you think that a 51% risk reduction doesn’t matter for this group, and that seems to be what Gellad is saying here, well then. That tells you how much people in this group should care about Covid-19.

In Other News

FDA panel backs vaccine for children as young as six months. Summary of the FDA meeting. Report on the progress of the FDA as they move towards approval.

I declare defense production act! What happens next? Could it, perhaps, send a strong price signal not to expand production capacity? Might manufacturers not be all that excited by the prospect of having everything dictated to them indefinitely?

New Bob Watcher thread, largely about Long Covid. He naively buys the ‘this is going to cripple tens of millions of people’ interpretation of the studies, not seeing any flaws in the studies and also not checking to see if that interpretation matches the world he observes, and he continues to be cautious. Included for fairness between perspectives.

Zeynep has a thread on conditional probability, might be useful for those who need it but if you’re reading this you probably know it all already.

Not Covid

Saudi Arabia to spend a billion dollars a year on anti-aging research. That’s pretty cool. My worry is that they will have no idea how to do this and waste the money, so those who are in a position to do so may want to look into how to prevent this from happening.

Reminder that we continue not to prepare vaccine prototypes in case there is another pandemic, because congress will provide no pandemic funding of any kind.

A reminder this week that the Red Cross is so perverse that they will throw your perfectly good blood away, while there is a shortage of blood, if you have a health condition whereby donating blood benefits you. Because, you see, these donations are ‘coerced’ by the medical need to donate. Which means you’d better not let anyone benefit?

I do not understand why many people have this pathological an opposition to trade and to positive sum interactions. They literally seem to think that it would be unethical for both parties to benefit from a transaction – it wouldn’t be right for a charity to benefit unless the person causing the benefit suffers in equal measure, it would upset the balance of the universe or something. This is not the first ethically dubious move, shall we say, that we have noted recently by the Red Cross.

A study claiming that a theoretical portfolio of potential mRNA vaccines would not have a high net present value. In their analysis 94% of all costs are clinical trials versus 6% for manufacturing, yet human challenge trials and increased probability of success would/did not substantially enhance profitability and the only good lever to ensure strong profitability is price per dose. That’s pretty weird.

Reading the paper and seeing its assumptions does not make things less weird. They claim they assume 0.8 correlation between candidate vaccines for the same disease in one section, and seem to claim no correlation earlier although that doesn’t impact the conclusions. They see large cost savings to testing multiple mRNA vaccinations at once, helping to justify the use of a ‘mega fund’ to test multiple candidates simultaneously, which also in theory allows this large portfolio of candidates to sell financial derivatives. I don’t see why that detail is necessary, although I suppose it could make the portfolio seem marginally more attractive.

In general there are ass-pull assumptions all over the place, so it seems very easy to get whatever results you feel like getting, positive or negative. The whole exercise seems bizarre and pointless to me, but I bothered reading it so I figured I’d report back so you don’t have to.

AGI risk versus pandemic risk, predicting the baseline future results edition.


5 comments, sorted by Click to highlight new comments since: Today at 6:36 AM
New Comment

Italian mask mandates on trains have enforcement proportional to distance travelled. Interesting range of possible explanations is available.

In Germany long-distance trains historically invest a lot more effort in checking whether the passengers have valid tickets than happens for shorter distances traveled. 

I would assume the most likely explanation is that those systems for checking whether people have valid tickets get also used for checking whether people comply with masking. 

If a high-benefit, low-cost approach is available, then following that approach that would make sense, but it doesn't seem to me that that's politically feasible?

Whereas, it may be more politically feasible to defund high-cost, low benefit approaches.

I honestly can't understand this policy outcome from my model for US politics. Shouldn't the Democrats be all for more funding, with the Republicans being somewhat against it (fiscal conservatism) but not intensely so (not a hill worth dying on)? Instead we get both parties ignoring it entirely. I would really appreciate an attempted explanation, because it just boggles the mind.

Previously after Pfizer's 11/5 interim report on Paxlovid in high risk patients, you said an 89%, 95% CI: [64, 97] (n=774) was certain enough to conclude efficacy but not certain enough to stop the trial because the CI was uncomfortably wide.
After they had their 12/14 final report on Paxlovid in high risk patients, you said an 89%, 95% CI: [72, 96] (n=1379) looked good.
At that time they also shared the interim report on Paxlovid in standard risk patients, showing 70%, 95% CI: [-8, 92] (n=854).
Now after their 6/14 final report on Paxlovid in standard risk patients, you're saying 51%, 95%: [-44, 83] (n=1145) is just an issue of sample size?

I've really appreciated all of your analysis and curating for us here, but I feel like we're missing a lot about your internal model of evidence for these kinds of studies.

That looks quite straightforward to me. When you're looking for evidence on the reduction in number of severe cases, you need a larger sample in a population where the proportion of severe cases is very small to begin with.

The important n here is not really the total number of people enrolled in the study, but the number of people who would suffer severe effects. It's just that you don't know which ones those will be (and if you did they'd be called "high risk" anyway), so you need to also enroll a much larger number of people to get enough who end up having severe effects in the control group and hopefully not in the treatment group.

Their June 14 press release says that they ended up with only 10 people out of 569 in the control group who developed severe disease, which is obviously not enough to draw conclusions from and they should have known that in advance. Even if Paxlovid had been magically 100% effective, the study was barely large enough to give reasonable confidence that it had any effect.

The problem is that the risk of severe disease in most of the population is low enough to be hard to cheaply study, but high enough to have huge health and economic impacts since the incidence case rate is of similar magnitude to the total population per year. Halving COVID hospitalizations in the standard risk population would save on the order of a hundred billion dollars per year in net medical costs alone, without even considering quality and duration of life and productivity.

So yes, this should absolutely be followed up with larger studies. Even if larger studies somehow cost more than a billion dollars, they would on net be obviously worthwhile. They would either verify that the drug reduces risk by some decent percentage and everyone should take Paxlovid when they contract COVID, or reveal that it doesn't work well enough to be worthwhile taking for most people and so avoiding the cost of unnecessary medication and its side effects. In practice they would reveal additional information which would also be useful.