Ben White

From Paul Graham’s “The Brand Age“:

The way to find golden ages is not to go looking for them. The way to find them—the way almost all their participants have found them historically—is by following interesting problems. If you’re smart and ambitious and honest with yourself, there’s no better guide than your taste in problems. Go where interesting problems are, and you’ll probably find that other smart and ambitious people have turned up there too. And later they’ll look back on what you did together and call it a golden age.

My inbound contains more than enough AI doomerism. It’s more fun, however, to consider solving some problems.

// 03.31.26

One of my old college professors, Ruth Wisse, just gave the Jefferson Lecture (the highest honor given by the National Endowment for the Humanities) at the tender age of 89. One brief quotation, about addressing the cultural drift to grievance over gratitude during Harvard faculty meetings:

The best I ever did, more than once, was to say, “Please remember that democracy is not transmitted biologically.”

That is from an argument referencing the essential daily Jewish prayer, the Shema, and the criticality of reinforcing important ideas more than just prioritizing novelty of content. Things can both be imperfect—even deeply flawed—and still worth celebrating.

If there is to be enduring government of, by, and for the people, the people would have to be instructed and reminded to respect and confidently to perpetuate their precious inheritance.

Nihilism, learned helplessness, and conspiracy don’t have to be a dominant cultural operating system.

// 03.30.26

The American Board of Radiology: An Updated Unofficial Financial Analysis

03.25.26 // Radiology

Back in 2019, I wrote A Deep Dive into the Tax Returns of the American Board of Radiology. It’s now 2026, and a lot has happened in the world, so I thought it was past time to look at the ABR again and also provide some (but not exhaustive) additional context as to how the ABR compares to other members of the American Board of Medical Specialties (ABMS).

As before, this is for informational purposes. I don’t work for or with the ABR, I am not an accountant or tax attorney, and I am certainly even less knowledgeable about other medical specialties and their boards.

I’ll provide the data—which is derived from the ABMS 2024-25 report, non-profit tax returns (Form 990s) from the ABMS members, and comparative average salary data (averaging Doximity, Marit Health, and Medscape)—as well as some commentary. It is perhaps no great surprise that my commentary here is again relatively critical, but I also want to make the somewhat obvious point up front that organizations are made up of largely good people doing what they feel is a reasonable job. I would like to think we can do better as a specialty and more broadly as physicians. You can and should draw your own conclusions.

I actually started doing this update (and more with the ABMS) about three years ago, but the broader data collection across the ABMS was extremely tedious and time-consuming, and I abandoned it. Now, this time around, I was able to deploy modern LLMs and voila. If you see a mistake, let me know.
Read More →

Staffing for Average

03.23.26 // Radiology

In radiology, staffing for average volumes is the easiest way to staff, because most times things work out—in the same way that just-in-time manufacturing is the most efficient way to make products when everything is going well. Unfortunately, all systems designed for a typical day or ideal circumstances fall apart when confronted with the reality of unavoidable variability.

If most days produce a certain imaging volumes, some will be less and some will be more—and, less commonly, some will be way less and some will be way more.

If there is less work to do than typical—let’s say when staffing a typical call shift—then that radiologist is either less busy (so that they can relax but generate less revenue), or, if there is an integrated worklist and specific RVU demands, then they can read from a different bucket of work in order to hit a certain level of productivity.

In most cases without a productivity model, this can put the onus on the individual radiologist to choose how hard to work: they have the option of embracing the benevolence of the list gods and reading fewer cases, or—if they want to hit some predetermined productivity metric for reimbursement—they can still bust the stack somewhere else to make up the shortfall.

Dealing with a busy shift is a harder problem. You would need to be consistently busy in order to change staffing. Most people don’t have two FTEs of work for a shift that used to have one person working, so adding a second person is undesirable in that both are unlikely to be busy (and nobody generally wants to work more shifts, especially if those fall during evenings or weekends).

In some cases, when there is a consistent amount of extra work, dealing with the volumes can be achieved through some amount of overlap—particularly at the busiest parts of the shift—while mitigating the amount of time people might be twiddling their thumbs.

But the reality is there are many situations where a radiologist is busier than they want to be but has no recourse to get help, and the group or practice has no reason to—or ability to—change staffing on an ongoing basis.

This is a situation where there are two solutions:

One is a generous productivity component such that a busy shift beyond some predetermined threshold results in extra money. This doesn’t make the work any less stressful, but it does mitigate the perceived unfairness and frustration of the suffering, which occurs when the work is more than you can handle and you end up spending longer to do it without a commensurate increase in compensation for that extra effort.

The other solution is internal moonlighting. That’s where a radiologist or practice could activate internal moonlighting in some capacity to get list support. Some practices do this by having optional scheduled shifts, but this doesn’t actually solve the variability problem. True ad hoc moonlighting could be activated automatically via software if the number of RVUs hitting the list per unit of time (e.g., hour) reaches a certain threshold, if the number of RVUs on the list hits a predetermined number, if turnaround times fall behind some metric, or just if the person working wants help.

All approaches are valid, and a group could decide whether they should pay the extra money for high volumes, or if the person working should pay some amount of money for requesting relief from their call pay.

This is one of the use cases for LnQ (+ teaser video) that I mentioned in a previous article (as well as this “surge” staffing post), and it illustrates how a one-size-fits-all practice works less and less well in our current era of radiologist shortage and rapid job turnover—at a time when many people are feeling conventional positions because want to do teleradiology work during business hours reading outpatient imaging and mitigating stress.

We need to find ways to make sure that people don’t burn out. While generally removing work can be part of the equation, recognizing work, fairly compensating work, having backup, and making sure people don’t feel resentful is the other part.

The Intern Pivot

03.20.26 // Medicine

A reader asked me last year whether they should pursue applying to radiology as an intern because they felt like, in the end, clinical medicine wasn’t for them.

  • I have no idea. Maybe?
  • But also: Internships are hard, and grass is green.
  • Everything has good and bad parts.
  • It’s easier to blame the current context than acknowledge our own roles and attitudes in grading a subjective experience. As in, it doesn’t always have to feel this way.
  • Coffee and vibes are great, but easy to overweigh from a brief trip on rounds.
  • Whatever you do, every field is better with a craftsman’s mentality.

If Match Day results weren’t what you wanted, be aware: the mind is powerful. You will adjust, you will be happy, and you can find meaning in Plan B—hopefully so much so that, years in the future, you will look on this as the universe giving you what you needed over what you wanted.

Does anyone have official verbiage from RadPartners about the roll-out of the TBWU (their customized time-based work units)? Can anyone share a full chart of common exams and their new values? I’ve heard they have different weightings for ER/IP/OP exam settings, and that would also make sense, but the only partial example I’ve seen circulating online is small and doesn’t distinguish, so it’s not particularly reliable.

// 03.19.26

Paying Surveys for Doctors

03.19.26 // Medicine

Medical surveys are an easy way to make a few bucks at a good hourly rate (well, maybe at least for a resident), and there are multiple sites offering surveys to physicians. The caveat is that, of course, most survey sponsors are typically looking for board-certified physicians with multiple years of experience, particularly in sub-specialties. The less experience you have, the more you need to be prepared to get screened out of what seem like promising survey opportunities.

This article was originally posted way back on Feb 26, 2014 and last updated March 2026. This page contains referral/affiliate links (thank you for your support).

ENOS is a new healthcare panel with a novel premise:  members are paid instantly via Venmo, Paypal, or paper check—no delays or redemption thresholds. ENOS always pays for your time: Even if you’re ineligible for a survey after completing screener questions, they still send you $5. Readers get a $25 sign-up bonus.

ZoomRx is also excellent and has a nice app and better/shorter-than-average surveys. $25 sign-up bonus for the following specialties: Hematology/Oncology, Cardiology, Neurology, Gastroenterology, Psychiatry, Nephrology, Rheumatology, Allergy/Immunology, Pulmonology, Dermatology, Urology, Endocrinology, Surgery. They even pay for attempts when you screen out.

One of the biggest survey sites is Sermo (also an online healthcare community), which is now offering my readers a $10 welcome bonus. The survey experience has been recently revamped, and once you maintain a balance of $100 in honoraria, you get preferentially invited to more surveys.

One of my very favorites is InCrowd, which has a slick mobile-friendly site and will send you survey opportunities by email or text message. These are always of the very short and painless variety (the fastest of all in my experience), so the payouts are small, but it’s good money for the time and basically effortless. You do have to respond quickly before surveys fill up, but you even get a buck when you get screened out. Being referred (like signing up through that link) will earn you a $10 bonus after you answer your first two microsurveys.

M3 now has three separate very active research companies under its umbrella: M3 Global Research, M-Panels, and All Global Circle. You can earn $25 for joining one panel, $40 for two, and $60 for joining all three (for the following specialties:  Hematology-Oncology, Neurology, Gastroenterology, Nephrology, Cardiology, Urology, Surgery, Rheumatology, Obstetrics and Gynecology, Pulmonology, Allergy and Immunology, Family Medicine, Psychiatry, Dermatology, Ophthalmology, Endocrinology/Diabetes, and Pediatrics).

Curizon has been in the business a long time, but they recently completely revamped their website and platform. It’s a trusted site for well-paying healthcare surveys for physicians as well as other healthcare professionals. Every new registration is entered in a monthly drawing for $100.

At the resident level, one of my old favorites has been Brand Institute, which almost exclusively sends out short surveys about potential drug brand names. Payouts are always on the smaller side ($15), but each one is quick (about $1 per minute or more) and screen-outs are rare. So if you get invited to a survey, then you can generally complete it and get the honoraria. No BS. The main style/format is nearly always the same, so you pick up speed as you do more of them. And that honoraria size is also significantly larger than what one can generally pull as a non-physician (e.g. SurveySavvy, the biggest most popular survey site around, usually pays a measly $2 per survey). The website, however, is clunky and terrible. You’ve been warned.

Additional legitimate additional survey sites, many of which are significantly less active, are below:

From Effort to Intuition

03.18.26 // Miscellany, Radiology

True education isn’t just transmitting information. It’s the information filtered through experience that makes it real.

Experience matters because it allows us to convert all that expensive type-two thinking into long-term memory. It can, in a sense, help convert type-two into type-one thinking—or at least less expensive thinking—by taking a complicated world full of many discrete ideas and concepts and chunking them into a smaller number of discrete elements.

When radiologists read a scan, we do not evaluate each individual pixel as we gain experience; we are able to take in larger and larger swaths of structures as units for interpretation and pick up discrepancies that don’t match our mental model of what something should look like.

This is why chess grandmasters can, at a glance, recreate a chessboard after viewing it for only a few seconds, but novices can only remember a handful of pieces. It’s why “1-9-8-5” is a string of four numbers, but 1985 is my birth year.

One of the unappreciated components of a liberal arts education or general skills is that someone who is generally skilled or smart is actually somebody with a large volume of mental models and a large amount of relevant information stored in long-term memory that they are capable of bringing to bear on novel situations.

Learning to think is an organic, holistic process of taking on more and more things and doing the hard, meaningful-but-not-always-fun work to integrate them into the person you are. There is no shortcut for that. That’s why using AI to write something that’s actually important doesn’t really work for many people, and that’s one reason why a score on a standardized test is a helpful but woefully incomplete metric.

As a trainee, I disliked interpreting spine imaging. This is in part because there are a lot of discrete decision points to make at every level regarding canal and neural foraminal stenoses, the location of disc pathology, the degree of joint degeneration, and so on and so forth. That’s a lot of work if each decision requires deliberation, and it’s psychologically unfulfilling when your attending changes all your grades.

But then you read thousands of scans, and much of that type-two thinking is well chunked into a near-automatic type-one process. It doesn’t even feel like thinking—you see a canal, and you see a foramen, and you know what it is. That’s why, as an attending, reading a degenerative spine can have more in common with meditation than nearly any part of my job.

Learning curves are high, and there’s enough internal cognitive load that we can only learn so much at a time. When everything is new and challenging, everything is hard, and everything takes time and energy. Anthropic CEO Dario Amodei is, as yet, very wrong that “the most highly technical part of the job has gone away.” There is as yet no mastery of any topic—no inborn system-one ability or computer skill that you can simply offload parts of the scan to.

The experiences mediate the learning and move the learner down the pathway from effortful to intuitive abilities.

This is why, when a well-trained radiologist reviews a scan, only a small fraction of the findings actually require thought. The rest conform to the library of pattern matching they can bring to bear.

Information vs Education

03.17.26 // Medicine

Preparing for standardized tests by yourself using high-quality resources is both effective and a little bit soulless. There’s a reason why much of medical education could be streamlined in both time and cost to what amounts to an old-time correspondence course—that’s because it’s long been sold as an information problem, and the core question for schools has been how to best transmit the holy information to the student.

This was probably historically at least partially true, but in the 21st century, this conception misses the point: information is no longer the core skill that needs cultivation. It is social intelligence, human skills, and the bringing of information to bear for problem-solving. It’s critical thinking, and doing it while working with people. And yes, the information is important (it really, really is!)—but the information is not where very expensive medical schools really shine. With the advent of better qBanks, Anki decks, and commercial lecture products, we are increasingly choosing a factory schooling model over one that prioritizes a social experience of working with caring teachers, motivated peers that are important to you, and patients with meaningful tasks to provide motivation and centering.

The schools and students spend lots of money on pieces of paper and question banks and other forms of curated video content, and these are all wonderful things. My point is not to suggest that using better lectures and better questions is a bad thing for education—it’s certainly not. It’s that those things do not preclude the need for the other part—not in some token problem-based learning format or anything so prescriptive—but real, meaningful, non-tedious, in-person work.

Whatever it once was, it is no longer mainly an information problem. As the cliche goes, you can lead a horse to water, but you can’t make it drink. We need to encourage people to be motivated (or find motivated people), and we need to help curate and support sustaining that curiosity—that people can do and internalize the need to continue doing that type-2 fun of hard work. The fact that students would rather be at home studying instead of working in the hospital shows the fruits of our system: the triviality of many students’ clinical experiences (the clerkship as performance art) combined with the pressure of Shelf exams as the defining feature of their grades.

The world is full of heavy objects, and yet most people are not ripped. We are the limiting factor, and we need systems to help us and support us to be our best selves. Burnout is a system that snuffs out our soul’s flickering flame.

There was a time in my radiology training where people were encouraged to incorporate multiple-choice questions with audience response into their lectures instead of taking hot-seat cases where people were put on the spot. That is more comfortable, to be sure, but I can tell you the fear and anxiety of wanting to perform for your colleagues and mentors was much more inspiring. Not every time you’re wrong in public is unfair pimping/humiliation.

An ability to take those hits on the chin when you flub a case is also important. Psychological safety doesn’t mean never being challenged—it means being supported enough that you can bounce back from your inevitable failures. We have forgotten how important resiliency is, and we’ve allowed undergraduate medical education to remain dominated by the factory information paradigm while neutering the chance for more students to become respected members of the care team more of the time.

Staying Small

03.16.26 // Medicine

I would posit that good healthcare is more analogous to a restaurant than most large corporations. From Michael E. Porter’s Competitive Strategy: Techniques for Analyzing Industries and Competitors:

If close local control and supervision of operations is essential to success the small firm may have an edge. In some industries, particularly services like nightclubs and eating places, an intense amount of close, personal supervision seems to be required. Absentee management works less effectively in such businesses, as a general rule, than an owner-manager who maintains close control over a relatively small operation. Smaller firms are often more efficient where personal service is the key to the business. The quality of personal service and the customer’s perception that individualized, responsive service is being provided often seem to decline with the size of the firm once a threshold is reached. This factor seems to lead to fragmentation in such industries as beauty care and consulting.

Healthcare has seen those fragmentation factors dissolve since the 1990s and especially since the ACA.

In Redefining Health Care: Creating Value-based Competition on Results, Porter then argues:

Competition has taken place at the wrong levels, and on the wrong things. It has gravitated to a zero-sum competition, in which the gains of one system participant come at the expense of others. Participants compete to shift costs to one another, accumulate bargaining power, and limit services. This kind of competition does not create value for patients, but erodes quality, fosters inefficiency, creates excess capacity, and drives up administrative costs, among other nefarious effects.

Over the years since I returned to the area, the local university medical center has progressively moved away from individualized service into a predictably depressing corporate marketshare grab. If this is “The Future of Medicine Today,” the future is bleak.

If policymakers want to improve US healthcare, the easiest lever to pull first is to enable physician ownership and make it feasible to stay small without needing to opt out of the system entirely by going direct pay.