Bryan Carmody Profile picture
May 11 23 tweets 6 min read Read on X
For the past couple of days, my timeline has been full of headlines about how medical students are avoiding states with more restrictive abortion laws.

The headlines are tantalizing… but are they true?

I’m skeptical.

(🧵) Screenshot of NPR story: “Medical residents are starting to avoid states with abortion bans.”
If you’ve somehow missed all of this, here’s a typical article (which happens to be the one I screenshotted above).

npr.org/sections/healt…
…and here’s the analysis by the @AAMCtoday, which used Electronic Residency Application Service data to find a decrease in U.S. MD graduates applying to residency programs in states with more restrictions on reproductive health.

aamcresearchinstitute.org/our-work/data-…
Their analysis was straightforward.

First, they broke states into 3 groups:

1. abortion remains legal (blue)
2. gestational age restrictions (orange)
3. abortion banned (red)

Then, they looked at how many ERAS applications were sent to residency programs in those state groups. Map from the Kaiser Family Foundation showing abortion bans by state.
Last year, applications from U.S. MD applicants were down across the board.

But applications ⬇️ most in states with abortion bans; a little less in states with gestational age restrictions; and the least in states where abortion remained legal.

The effect was largest in OBGYN.
Screenshot from AAMC report, showing a 4.2% reduction in applicants from 2022-2023 in states with abortion bans, vs. 1.9% decrease for states with gestational age limits, and 0.6% for states where abortion remains legal.
For OBGYN, MD senior applicants declined by 6.7% in states where abortion is banned, 2.6% in states with gestational age limits, and increased by 0.4% in states where abortion remained legal.
There are several things about the AAMC’s numbers that don’t make sense to me, but for now, I’m gonna set all that aside and take their figures as fact - because even then, there are some strange things.
Only *two* states saw an increase in applicants: Montana (30.3%) and Delaware (3.7%). Both of these are states where abortion remains legal.

But then again, so is the state that saw the biggest drop in applicants: New Mexico (-19.3%). The KFF Map of abortion by state.
Some red states saw big decreases - e.g., Texas dropped by 11.7%.

But Washington (blue) had an even bigger drop - in both relative (13.7%) and absolute terms (-1521 vs. -1499).
Similarly, abortion-banned Indiana (-9.1%) saw a smaller drop in applicants than abortion-legal California (-9.2%).

Oregon (-13.7%) had a bigger drop than Tennessee (-12.7%).

Colorado (-10.6%) had a bigger decrease than North Dakota (-9.8%).

Wait, what?
If the decline in applicants is driven by a desire to avoid abortion restrictions - or even with other political/social factors that are collinear with abortion restrictions - shouldn’t applicants be flocking to the bluest states?

Or is something else going on?
Beyond any state-by-state breakdown, the big story in the AAMC analysis is the decline in applications OVERALL.

So IMHO, we should also look at the state breakdown through the lens of what is driving this unprecedented decrease in applications:

And that’s preference signaling.
Dobbs wasn’t the only big thing that changed in residency application over the past couple of years.

Among other things, the AAMC also introduced geographic preference signaling, in which applicants can signal up to three regions of interest.
And when I look at the state-by-state results, they seem to conform better to the ERAS geographic signaling map than the KFF map.
ERAS geographic regions, 2023-2024.
Kaiser Family Foundation abortion status map.
If you’re unfamiliar with preference signaling, or the strategy behind effective preference signaling, I’d invite you to watch the guide for applicants I made last year:

The Applicant’s Guide to Strategic Preference Signaling
Briefly - because applications are expensive (and geographic preference signaling is free), it makes sense for many applicants to target most of their residency applications within regions for which they’ve indicated a geographic preference.
Thing is, you can only signal three geographic regions… and these regions aren’t created equally.

And yes, yes, yes - many applicants have strong desires to be in one geographic region or another. But ALL applicants have a strong desire to successfully match.
The ERAS regions that attract the *fewest* geographic preference signals from applicants are the “West North Central” (yellow), “East South Central” (pink), Mountain West (blue), and “West South Central” (dark red).

And these are the states with the greatest ⬇️ in applicants. ERAS geographic signaling map, again.
In contrast, the New England (purple), Middle Atlantic (orange), South Atlantic (light blue) and “East North Central” (green) have greater program density and attract more geographic signals - and applicants.

And these states were most protected against a decline in applicants.
I think this best explains why a state like Kentucky (-15.4%) did worse than West Virginia (-12.0). Both states are abortion-banned - but one is in a more favorable ERAS region.
Similarly, Minnesota (-12.5%) did much worse than neighboring Wisconsin (-6.3%) - but it’s stuck in the yellow region. Meanwhile, Indiana’s decrease (-9.1%) was similar to that in abortion legal states like Maine (-8.8%) and Connecticut (-9.3%).
None of this is to say that applicants aren’t motivated by the political climates in states where they apply. But that was true pre-Dobbs, too.

I’m certain some applicants are strongly motivated by this issue. I’m just skeptical it’s of the magnitude the AAMC suggests.
Totally forgot to say:

This probably also explains the AAMC’s otherwise-puzzling finding that the avoidance of abortion-limited states only applies to MD applicants, not DOs or IMGs. Screenshot from AAMC analysis showing increase in DO applicants in gestational-limited states and a disproportionate decrease in IMG applicants in gestational age limited states.
Although the greatest density of prestigious programs within ERAS regions is in the East Coast/Great Lakes, the regions with the greatest density of DO- and IMG-friendly programs are in other regions (where states tend to have more abortion restrictions). Image

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Bryan Carmody

Bryan Carmody Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @jbcarmody

Feb 19
UPDATE:

Last week, a Nepali doctor filed a class action lawsuit against the National Board of Medical Examiners, alleging discrimination based upon national origin and requesting that invalidated USMLE scores be restored while examinees appeal.

Today, the NBME responded.

(🧵)
The court will issue a ruling by February 21 - so the outcome of the suit remains uncertain.

Still, the NBME’s filing provides additional details on the scope of the scandal, how the cheaters were caught, and what’s likely to happen in the future.
First, the NBME confirmed the number of examinees involved.

According to their filing, 832 examinees have had at least one of their USMLE scores invalidated… so far. Screenshot from the defendant’s filing, noting that 832 individuals had at least one USMLE score invalidated.
Read 16 tweets
Feb 15
A brief update on the USMLE cheating scandal:

This week, a Nepali doctor and Match applicant whose Step 1, Step 2 CK, and Step 3 scores were all invalidated has sued the NBME.

The NBME will respond by February 19, and the court will render a decision by February 21.

(🧵) Screenshot from the lawsuit, noting that the plaintiff requests a decision by February 21, 2024, so that she can participate in the 2024 Match. The NBME has asked to have until February 19 to file an opposition brief.
The suit alleges that the NBME did not follow its own policies by invalidating scores before the involved examinees had an opportunity to appeal. Screenshot from the lawsuit, describing the USMLE’s procedure for investigating whether a person cheated, and allowing an appeal. The quoted section does not provide for identifying cheaters before the process is completed.
The suit also alleges violations of the Civil Rights Act due to discrimination based on national origin and ethnicity, based on the USMLE’s initial statement that the scandal was “associated with Nepal.” “In a press release, NBME explained that its new policies applied to test-takers ‘associated with Nepal.’”
Read 12 tweets
Feb 9
It’s always messy when a residency program loses accreditation - but this one seems particularly messy.

An update on the Crozer Chester surgery residency program closure…

(🧵)
Some background:

On January 8, the ACGME withdrew the accreditation for the Crozer Chester Medical Center general surgery residency program.

The hospital was told that the program need to shut down by January 12 - just FOUR DAYS later.

beckershospitalreview.com/hospital-physi…
The hospital leadership - as well as the Pennsylvania Attorney General! - reached out to the ACGME and asked for an extension while they appealed.

They agreed on January 30, and an appeal was filed - which would not be decided until ~June 30.
Read 19 tweets
Dec 7, 2023
Big changes may be coming for internal medicine residency applicants in the 2024-2025 application cycle…

The @AAIMOnline is recommending:

-⬆️ preference signals from 7 to 15 (3 gold, 12 silver)

-capping interviews at 15

-no interviews before November 1
Increasing the number of signals will likely function as a soft application cap, as we’ve seen in other specialties.

Many IM programs will still evaluate/interview applicants who don’t signal - but many others will receive so many signaled applications that they’ll stop there.
This will shift the dominant application strategy from “Apply to as many programs as possible” to “How do I most effectively allocate my preference signals?”

Applicants who signal wisely will benefit… those who don’t will suffer.

More here:
Read 8 tweets
Nov 8, 2023
If you want to become a pediatric subspecialist, you’ve gotta do a fellowship.

For most subspecialties, that fellowship lasts 3 years.

But most of that time isn’t spent learning clinical medicine. It’s spent doing research.

Ever wondered why these fellowships last so long?

🧵


Screenshot from NRMP website, listing typical fellowship length for  academic general pediatrics (2-3 years), child abuse (3 years), developmental-behavioral pediatrics (3 years), and neonatal-perinatal medicine (3 years).
Screenshot from NRMP website listing typical fellowship length for pediatric cardiology, pediatric critical care, pediatric emergency medicine, pediatric endocrinology, and pediatric gastroenterology. The length is 3 years for each subspecialty except for pediatric EM, which is 2-3 years.
Screenshot from NRMP website showing typical length of fellowship for pediatric hematology-oncology, pediatric hospital medicine, pediatric infectious disease, pediatric nephrology, and pediatric pulmonology. The length of fellowship is 3 years for each subspecialty except for pediatric hospital medicine, which is 2-3 years.
Screenshot from NRMP website, showing typical fellowship length for pediatric rheumatology (3 years) and pediatric transplant hepatology (1 year).
The short answer is, “because the American Board of Pediatrics requires it.”

But why?

For instance, in many specialties, a resident can choose between a 2-year clinical or a 3-year academic fellowship. The ABIM accepts either… but the ABP only accepts the latter.
To find the answer, we’ve gotta go back to the early 1980s.

At that point, peds heme-onc was a 2-year fellowship. But PDs felt trainees needed more training time to attain clinical proficiency.

So they asked the ABP if they could increase ⬆️ their fellowship from 2 to 3 years.
Read 13 tweets
Oct 25, 2023
For weeks, I’ve waited patiently for preliminary application data for the 2023-2024 residency application to be released by the AAMC.

And today they dropped.

So you know what that means.

Yup, that’s right.

It’s time to break it down, Winners & Losers style™️.

(🧵)
LOSER: Application Fever.

For years, the biggest theme in the ERAS opening data is some variation of “What incredibly large number of applications are being submitted now?”

Applicants are still submitting lots of applications… but many specialties are DOWN from last year.
Graphic showing the mean number of applications per applicant for the 2023-2024 residency application cycle.
Image
When you look at the overall application numbers, you’ve got to remember that these figures consider all comers (MD, DO, IMG). But there are important differences between these groups.


Mean applications per applicant, by specialty, for MD/DO/IMG applicants in 2023-2024.
Image
Image
Image
Read 13 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(