Economists do occasionally publish papers with which I agree: ⊕ For the other ones, look here.

…I find that vaccines saved 748,600 lives through June 2023. That is, without vaccines, cumulative mortality from COVID-19 would have been closer to 1.91 million over this time period. In answering the second question, I find that behavioral efforts to slow the transmission of the virus before vaccines became widely administered were critical to this positive impact of vaccines on cumulative mortality. For example, with a complete relaxation of these mitigation efforts, vaccines would have come too late to have saved a significant number of lives. Earlier deployment of vaccines would have saved many lives.

Which yet again shows that out of the two extremes, John Snow and GBD, Snow was the more correct one both ex ante and ex post. Yet instead of taking an “L”, GBD proponents keep saying that we should not have locked down. Yes, shutting down outdoor playgrounds for a full year was ridiculous, but stopping mass gatherings and any goings on in tight public spaces until we get a working vaccine? Absolutely! Only next time, Operation Warp Speed should be set at Warp 5, not 0.1.(ᔥTyler Cowen)

The Washington Post’s Andrew Van Dam on the average US doctors' salaries:

The average U.S. physician earns $350,000 a year. Top doctors pull in 10 times that.

I will write more about this later but for now I will just note how frustrating it is to read an article that has a premise and conclusion that I completely agree with (America doesn’t have enough doctors so the ones that it does have are compensated way above average) backed up by mishandled and misreported data (first the article doesn’t say whether the “average” is mean or median — it is the median, which is actually good — then doesn’t explicitly mention that the median in question is of the adjusted gross income at the household level, not of individual compensation: the median total individual income is $265,000).

At least the article linked to the NBER paper with all the data, which in turn completely validated my recent quip about economisits. Frustrating throughout, especially if you try reading the comments.

Economists are prone to making hypotheses about other fields that make perfect sense to them and others outside of the field, but that can be easily refuted to anyone with an iota of relevant field-specific knowledge. And not just economists. This very sentence is, in fact, one such hypothesis.

But to get to the point: Alex Tabarrok at Marginal Revolution wrote a post titled Left Digit Bias in Medicine which excerpted his WSJ review of Random Acts of Medicine which is a book about “The Hidden Forces That Sway Doctors, Impact Patients, and Shape Our Health”, which is, believe it or not, part of the title. ⊕ I hate, hate, hate what book titles have become. This is the literary equivalent of the Amazon product name trash recently discussed on ATP. SEO for books as an externality of Amazon’s dominance — who would have guessed?

Now, I haven’t read the book yet, and the WSJ is behind a paywall, so I only have the excerpt to go by, but it is long and it is sufficient. I won’t quote from it — 4th level of abstraction would be too much — but I will copy the figure and make a few comments. You should read the blog post itself, it is good.

1. This is the biggest and the most obvious regression discontinuity I have seen, and it has a reasonable explanation. Kudos.
2. Age is plotted as a continuous variable. This is not how doctors see the patient’s age in their medical record. It is shown as an integer, not a fraction, so someone who is a day shy of their 40th birthday will look just the same as someone who just turned 39. And if the guidelines say you should do something for a 40-year-old but not for a 39-year-old, that’s what most doctors — let’s hope — will do.
3. This is therefore not left-digit bias.
4. While the date of birth is also part of the medical record, it is rarely if ever looked at by MDs — except just prior to an invasive procedure that requires a timeout. It is often checked by nurses prior to administering medications, and they are often the ones who will note that an inpatient’s birthday is coming up.
5. Tabarrok has buried the lead in his blog post. Regardless of the cause, the discontinuity is there and can be used as pseudo-rendomization for a natural experiment of the effect of “testing for heart attacks” (I will guess by that the others meant an ECG and troponin levels) on outcomes. The entire last paragraph of the excerpt is about that, and I 100% endorse the idea.

Two things keep me logging back in to Twitter X: DMs from people who should know better, and all of my colleagues who insist using it. But how else was I supposed to learn of this petition for ABIM to eliminate their “maintenance of certification” grift?

Yesterday’s EconTalk was with Lydia Dugdale on the Lost Art of Dying, which is the title of Dr. Dugdale’s book but also a translation of Ars moriendi, a 15th century Latin text about the good death. The episode is in this year’s Top 5, and I wish I could dwell into this. Ars longa…, as they say.

Visiting Montauk beach at Calvert Cliffs, a family member had one mission: to find a shark tooth. Millions of years ago, this part of Chesapeake was warmer and mostly under water. Many a shark dropped a tooth or a hundred during that time; today, they tend to drift to the shore with some regularity.

Searching for a speck of black in a tapestry of white-gray brought to mind Annie Dillard’s Pilgrim at Tinker Creek, more specifically the chapter about learning to see, and yet even more specifically, her discovering praying mantis egg cases This is a longer blog post from The Examined Life about writers and insects; scroll down for the Pilgrim… excerpt. everywhere she looked, once she learned what one looks like.

My own learning-to-see training started with watching birds — not organized or consistent enough to be called birdwatching — and realizing in short order that not every brown-gray bird smaller than a robin is a sparrow, that blue jays, cardinals, and woodpeckers are actually quite abundant even in urban areas, and that those blue jays, as magnificent as they are, usually sound like nails on a chalkboard. The beach makes for even better training grounds. For novices like us there are mermaid’s purses and loggerhead turtle tracks — we saw both during our Outer Banks excursion — things alien enough to immediately be recognized as something. The mental exercise consists of discovering what that something is.

Not so with shark teeth, especially not with the small ones you are more likely to come across during a daytime summer stroll, as opposed to a planned break-of-dawn winter expedition. Is it a spiky piece of iron ore? A fossilized crab claw? Tooth of a mammal? Who knows!? Short of finding a 6-inch dental behemoth, casual beachgoers like us will come up with a million reasons why this black triangle isn’t an actual tooth, and why this other may be, without ever knowing if they are correct. Annie Dillard could put that insect egg casing in a jar and see dozens of tiny praying mantisses scuttle out and devour each other. I can put my black triangle in a dish and look at it until the Sun implodes, and it will continue being that same black triangle, possibly melted.

Unless, of course, we find an expert to tell us why these ridges here mean that it comes from a shark’s jaw, or why this dent over there means it is actually part of a crab. And, knowing that, we will know with certainty — conditional on us trusting the expert — what those two particular artifacts are, but could hardly extrapolate to other pieces of black material found on the beach, and most certainly not to those nestled on the forest floor, or buried in the desert sands, or hiding under the carpet of a 3-story walk-up.

This is in fact very much how medicine works: sometimes, the symptoms are clear enough and occur often enough that you may know as well as an MD that there is a urinary tract infection brewing. But too often — most of the time, in fact — the problems are subtle and chronic and may not develop into something recognizable until it is too late — in which case you better find an expert — or, maybe, never amount to much of anything — in which case you need that expert even more, the most valuable part of medical expertise consisting of the knowledge and experience needed to muster the confidence to say that something is just a piece of rock.

Update: Two months later, we went back and found some.

I recently attended a residency graduation party at an academic medical center, for the first time since the pandemic. Two things struck me:

1. So. Many. Awards. For the residents. For the faculty. For the ancillary staff. There were nearly as many awards as there were graduating residents.
2. No roasting of the graduating house staff, or even a hint of humor of any kind. This used to be the highlight of any graduation party.

Award inflation is akin to grade inflation: they have become currency for further post-graduate training and, more importantly, faculty promotion. With the recent focus on diversity, equity, and inclusion, a whole new spectrum of accolades has opened up. So yes, there is a reason for all those plaques being thrown left and right, but it was funny nevertheless to see faculty speed through the list of graduates, then spend the next hour patting themselves on the back.

The lack of a proper roast was more concerning. Has the environment become so fraught that the residents are concerned about offending anyone? Humor is to dialogue what beavers are to a river: sometimes a nuisance, but also the hallmark of a healthy ecosystem. Or should I say good humor; when done lazily and as an afterthought, roasts too often devolved into a series of racial and sexual stereotypes. I imagine that is why some places have done away with them, which is also a lazy, unimaginative thing to do — you would think that with all the stress on DEI, the graduates would if anything be more capable of doing a character/personality rather than race/orientation-based roast.

What I hope DEI workshops did not teach them is that they should go out of their way to avoid making people uncomfortable. Sometimes people should be uncomfortable, and making them squirm just a little bit at the highest peak of their career-to-date is the best time for it. They will have the entire rest of the night to pat themselves on the back.

I have always admired prolific writers like Matthew Yglesias and Scott Alexander — both now on Substack, and not by accident — for their ability to produce tens of thousands of words daily, ⊕ My admiration being tampered somewhat by ChatGPT and other LLMs, which are about as intellectually and factually rigorous as Alexander, and slightly less so than Yglesias; some sacrifices do have to be made in the name of productivity. on top of the random bite-sized thoughts posted on social media. There are only so many words I can read and write in a day, and for the better part of the last year, my language IO has been preoccupied by helping clean, analyze, interpret, and write up the results of a single clinical trial, which are now finally out in The Lancet Neurology. ⊕ Yes, my highest impact factor paper to date is in a neurology journal. Go figure.

The paper is about our clinical trial which used the body’s own immune system to treat autoimmune disease — and a particular one at that, myasthenia gravis — via technology that up until now has only been used against cancer (CAR T cells). It has made a decent impact since it came out less than two days ago. It got a write-up in The Economist, for one. Endpoints News as well. Evaluate Vantage got the best quote — it is at the very end of the article. And there is a whole bunch of press releases: from National Institutes of Health, University of North Carolina, Oregon Health and Sciences University, and of course Cartesian Therapeutics.

What went on yesterday reminded me that Twitter is not going anywhere any time soon: all of the above releases were to be found only there, not on a Mastodon instance, the journal’s own media metrics do not — and can not, at least not easily — trawl the Fediverse for hits, and I can’t just type in “Descartes–08”, “myasthenia gravis CAR-T”, or “Cartesian” into a Mastodon search box and get anything of relevance. ⊕ One could, of course, argue that you wouldn’t get anything of relevance on Twitter either, most of the discussion consisting of people who have barely read the tweet, let alone the article. And one would be correct. And while most of the non-Web3/crypto tech world has moved out, it looks like people in most other fields, from medicine to biotechnology to the NBA commentariat, are maintaining substantial Twitter presence.

This will, of course, have no impact on my commitment to staying out of the conversation to the extent possible while maintaining a semi-regular schedule of 500-character posts, which may now, IO bandwidth having opened up, become a tiny bit longer. Thank you for reading!

Washington Post’s Katherine Ellison on the striking decrease in mortality from lung and breast cancer in the US:

There are many and varied explanations for the progress, says Memorial Sloan Kettering oncologist Larry Norton, including “better early diagnosis, better imaging, better blood tests, better preventive measures and better treatments, including precision medicine with gene-profiling of patients’ tumors.”

The rest of the article focuses on treatments — immunotherapy in particular — ⊕ And yes, of course dostarlimab was mentioned. Once a darling, always a darling. and cancer survivorship, but in discussing decreasing deaths from lung and breast cancer the article missed an opportunity for some education in cancer epidemiology.

The two sources chosen to present these data, ASCO’s cancer.net for lung and Breastcancer.org for breast are lacking in two ways: they are an impenetrable wall of text without much context, and they both have an agenda. Now, it happens that I agree with ASCO’s agenda — I am a dues-paying member — and don’t know enough Breastcancer.org to form an opinion, but neutral parties they are not. If only there was a tax-funded, publicly available database which could help us visualize trends in cancer statistics.

Now it so happens that the CDC maintains such a database, with its very on visualization tools, and it is exactly what we need. It will even make your PowerPoint slides for you! And yes, deaths from both lung and breast cancer have been steadily decreasing for the past two decades.

But is it because of better prevention, early diagnosis, more effective treatments, or all three? Looking at cancer incidence — the number of newly diagnosed cases per year — may help some. Better prevention would lead to decreased incidence, early detection would lead to an increase, a combination of the two may cancel each other out leading to a flat line, and any change in treatments would not affect it at all.

A slight initial dip in female breast cancer incidence followed by an even slighter increase make me think that early detection — all those mammograms — is superimposed on better prevention. The case is less ambiguous for lung cancer: the incidence is plummeting. In both cases, “prevention” was initiated by the 1964 Surgeon General’s report on tobacco smoke which led to massive anti-smoking campaigns from the 1970s onwards. The results weren’t immediately obvious — not having to air out all your clothes after a night out notwithstanding — but cancer rates started dropping after 20 years, and 50 years later we are reaping the full benefits.

Note that in lung cancer the mortality slope is steeper than the incidence slope. And while this may be explained by early detection and better treatments, it is possible that at least some of the improvement over newly diagnosed lung cancers is due to non-smoking associated lung cancer being generally less aggressive and occurring in younger and healthier people than tobacco-associated cancers. What could help unravel these different components — and highlight the increasing importance of cancer survivor healthcare — would be a prevalence curve: how many people in the United States are currently living with a particular cancer. Alas, those data are not available.

If you thought interpreting those four curves was interesting, do go back to the CDC database and check out the incidence and mortality curves for thyroid cancer — that poster child of over-diagnosis — and prostate cancer, the incidence of which fluctuates ever which way with changing screening recommendations but with mortality marching downwards for the last 20 years.

After seeing a friend and collaborator yet again plant foot firmly in mouth, I begin to see a pattern. A course on ergodicity should be a requirement for a public health degree, since the masters of public health keep getting it wrong (see also: the screening colonoscopy debate).