Scathing Call Out of NRMP Violations

[ridethecliche]

Just saw this pop up on the interwebs recently and thought it made a pretty compelling point re: programs violating the NRMP rules.

Discuss..

Assuring Integrity in the Residency Match Process : Academic Medicine

(as scathing as one could theoretically be in a journal lol)

---

Members don't see this ad.

 

[Gastrapathy]

Guess it was a slow day on the wards.

 

[NotAProgDirector]

It's an interesting article. Not sure I really understand all of their outrage.

I agree with the "only offer interviews to the number of people you can fit". That's just plain decency.

The post interview communication thing is more vague. I agree that programs shouldn't be digging for where students will be ranking them. And no employer should be asking about religion, children, etc. But some of the email examples in that article are totally fine IMHO. Completely banning post interview communication is also stress inducing.

Second visits - If programs decide they want to have a second visit to help decide whom they are interested in, then that's their right. I don't, but others might. Banning them seems extreme.

And how would we police all of this? Anonymous reporting is not very helpful.

Overall, I think it would be best to focus on the interview issue. Although, if we decrease the stress due to this issue, I'm sure something else will be the biggest source of stress.

 

[Raryn]

It's an interesting article. Not sure I really understand all of their outrage.

I agree with the "only offer interviews to the number of people you can fit". That's just plain decency.

The post interview communication thing is more vague. I agree that programs shouldn't be digging for where students will be ranking them. And no employer should be asking about religion, children, etc. But some of the email examples in that article are totally fine IMHO. Completely banning post interview communication is also stress inducing.

Second visits - If programs decide they want to have a second visit to help decide whom they are interested in, then that's their right. I don't, but others might. Banning them seems extreme.

And how would we police all of this? Anonymous reporting is not very helpful.

Overall, I think it would be best to focus on the interview issue. Although, if we decrease the stress due to this issue, I'm sure something else will be the biggest source of stress.

Honestly, what would improve stress the most is if the arms race ended. People are applying to more and more programs and ranking more and more programs every year. Look at the last page there - In 15 years we've gone from the average applicant ranking 8 programs to 12.5 - an increase of more than 50%.

On an individual level, it makes complete sense - but it just makes the system crappier for everyone involved. You have to go on more interviews, the programs have to interview more people, etc.

I don't know how we'd reverse it short of limiting the # of programs one could apply to though.

 

[Donald Juan]

It's an interesting article. Not sure I really understand all of their outrage.

I agree with the "only offer interviews to the number of people you can fit". That's just plain decency.

The post interview communication thing is more vague. I agree that programs shouldn't be digging for where students will be ranking them. And no employer should be asking about religion, children, etc. But some of the email examples in that article are totally fine IMHO. Completely banning post interview communication is also stress inducing.

Second visits - If programs decide they want to have a second visit to help decide whom they are interested in, then that's their right. I don't, but others might. Banning them seems extreme.

And how would we police all of this? Anonymous reporting is not very helpful.

Overall, I think it would be best to focus on the interview issue. Although, if we decrease the stress due to this issue, I'm sure something else will be the biggest source of stress.

I agree, the example given of a program director emailing an applicant seems very friendly and not unethical or against the rules, although the authors imply that it was just a ploy to get info out of the applicant.

 

[NotAProgDirector]

Honestly, what would improve stress the most is if the arms race ended. People are applying to more and more programs and ranking more and more programs every year. Look at the last page there - In 15 years we've gone from the average applicant ranking 8 programs to 12.5 - an increase of more than 50%.

On an individual level, it makes complete sense - but it just makes the system crappier for everyone involved. You have to go on more interviews, the programs have to interview more people, etc.

I don't know how we'd reverse it short of limiting the # of programs one could apply to though.

There's been lots of talk about this. I don't see any simple solution.

One option is to limit applications. That seems unfair to applicants. It will force applicants to decide whether it's "worth" applying to a program, or whether they are "wasting" an application. There would need to be an exception for people couples matching.

Another is increased transparency from programs -- minimum USMLE scores, for example. But that's complicated, programs might have a "minimum" score of 220 but basically never invite anyone less than 240 (made up example, don't panic!). Or they might have different minima for US grads and IMG's.

A "third" is to combine these two options -- limit apps, and force programs to be more transparent. Perhaps the best thing about this plan is it makes everyone unhappy.

An interesting solution is to come up with a list of strengths, or interests. There might be 20 or so of these -- Research (clinical and/or lab), Community based care, social activism, rural vs urban, VA association, community hospital association, etc. Each program would pick some limited number of these as "strengths". Each applicant would pick some limited number as "interests". ERAS could show applicants which programs best matched their interests, and programs could pick whom they wanted to interview based on interest / strength match (although programs could ignore this completely if they wanted to). Perhaps this would limit applications.

The last option I've discussed is some sort of early decision process. One application to a program you're very interested in as a binding application -- if they take you, you're matched. Programs would only be able to fill some maximum of their spots this way (perhaps 30%). This would remove some of the most competitive applicants from the pool up front, saving all of those interview slots. How to time this is unclear -- it would be best to have the early application process complete before applications for the regular cycle go out. But at a minimum would need to have the early decision round complete before interviews for the regular cycle start. Likely would require pushing the match timeline forward in time -- which creates a huge problem for programs that need to get visas for their interns.

 

[enalli]

There's been lots of talk about this. I don't see any simple solution. One option is to limit applications. That seems unfair to applicants. It will force applicants to decide whether it's "worth" applying to a program, or whether they are "wasting" an application. There would need to be an exception for people couples matching. Another is increased transparency from programs -- minimum USMLE scores, for example. But that's complicated, programs might have a "minimum" score of 220 but basically never invite anyone less than 240 (made up example, don't panic!). Or they might have different minima for US grads and IMG's. A "third" is to combine these two options -- limit apps, and force programs to be more transparent. Perhaps the best thing about this plan is it makes everyone unhappy. An interesting solution is to come up with a list of strengths, or interests. There might be 20 or so of these -- Research (clinical and/or lab), Community based care, social activism, rural vs urban, VA association, community hospital association, etc. Each program would pick some limited number of these as "strengths". Each applicant would pick some limited number as "interests". ERAS could show applicants which programs best matched their interests, and programs could pick whom they wanted to interview based on interest / strength match (although programs could ignore this completely if they wanted to). Perhaps this would limit applications. The last option I've discussed is some sort of early decision process. One application to a program you're very interested in as a binding application -- if they take you, you're matched. Programs would only be able to fill some maximum of their spots this way (perhaps 30%). This would remove some of the most competitive applicants from the pool up front, saving all of those interview slots. How to time this is unclear -- it would be best to have the early application process complete before applications for the regular cycle go out. But at a minimum would need to have the early decision round complete before interviews for the regular cycle start. Likely would require pushing the match timeline forward in time -- which creates a huge problem for programs that need to get visas for their interns.

A huge underlying issue is that students are notoriously bad judges of how strong of an applicant they are--you really only know when you start getting interviews. Also, even if you know you are an average or weaker applicant, there isn't a set ranking of programs that will definitely interview you. It really is a complete crapshoot. My worry is that by limiting the number of applications, some students (who would have gotten 5-8 interviews from 60-100 applications) will be shut out of interviews completely.

How about just limiting the number of interviews you can accept?

You would need a completely centralized interview invitation and scheduling system but I envision it like this:
- Students apply to as many programs as they want, but there is an earlier deadline (say September 30)
- On a specific date (say October 15), programs offer interviews through the system
- Within 2-3 days, students select up to 7 invitations to accept
- There will be a second round of interviews offered (for example, by October 25)
- Students get an additional 3 interview slots and can now schedule up to 10 interviews total
- After this point, programs can offer interviews at any time, but students cannot attend more than 10 interviews

The chosen dates and number of interviews are just an example, but you get the idea. This system would still overall limit interview costs for everybody and decrease the length of rank lists.

 

[evilbooyaa]

There should not be limitations on applicants. I understand it's a rat race and neurotic people want to do 10-15 interviews in IM to make sure they match, but there is not a great solution for that.

There should be teaching on the minimum number of interviews necessary to guarantee application success (that number is like 5-6 for a USMG in IM).

The issue is not the number of applications, but rather the amount of interviews. Limiting interviews is not a great solution either, IMO.

Having an early decision style of interview would probably be the best way to manage most of the sheer numbers.

 

[enalli]

The issue is not the number of applications, but rather the amount of interviews. Limiting interviews is not a great solution either, IMO.

Can you elaborate on these two statements? They seem to contradict each other.

 

[evilbooyaa]

Can you elaborate on these two statements? They seem to contradict each other.

For programs and applicants - the main issue is not the number of applications students put out, but rather the number of interviews attended. If student A applied to 80 places, got interview invites at 40 of them, but only went on 5 interviews, it wouldn't be a huge issue from the program's or applicant's standpoint. Student A would likely end up at one of those 5 places. If program X has 4 spots per year and interviews people with the mindset of student A, then (assuming that rank preference is random, which is a big assumption obviously) the program would only have to interview 20 people to ensure that they filled their 4 spots per year.

I won't re-run the numbers with student B applying to 80 places, gets 40 interviews, and goes on 20 interviews, but you can easily tell the number of applicants a program has to interview per spot just went up exponentially.

Let me rephrase - the issue is the number of interviews (not applications). However, to limit an applicant from only going on X (let's say 10) interviews is not within the applicants' best interests. Not infrequently, due to scheduling, the 11th nterview you go on is a program that is going to go higher than 11th on your rank list. People find out about interview dates at various times throughout the ERAS cycle. Some programs interview earlier or later than the most coveted programs (especially in smaller fields) so as to allow those applicants to interview with them as well. Some programs purposely interview on the SAME day as multiple other programs to FORCE applicants to choose one over the other (assuming interviews have been obtained at both programs).

As an applicant, especially one applying to a competitive specialty, if you have borderline stats, you might get 10 interviews, but be in the lower 1/3rd of the rank list at all 10 of those programs. Match day comes along and due to the (proposed) hard limit on interview you don't match. Whose fault is it? Yours', your advisors', or the NRMP's?

 

[gutonc]

A huge underlying issue is that students are notoriously bad judges of how strong of an applicant they are--you really only know when you start getting interviews. Also, even if you know you are an average or weaker applicant, there isn't a set ranking of programs that will definitely interview you. It really is a complete crapshoot. My worry is that by limiting the number of applications, some students (who would have gotten 5-8 interviews from 60-100 applications) will be shut out of interviews completely.

How about just limiting the number of interviews you can accept?

You would need a completely centralized interview invitation and scheduling system but I envision it like this:
- Students apply to as many programs as they want, but there is an earlier deadline (say September 30)
- On a specific date (say October 15), programs offer interviews through the system
- Within 2-3 days, students select up to 7 invitations to accept
- There will be a second round of interviews offered (for example, by October 25)
- Students get an additional 3 interview slots and can now schedule up to 10 interviews total
- After this point, programs can offer interviews at any time, but students cannot attend more than 10 interviews

The chosen dates and number of interviews are just an example, but you get the idea. This system would still overall limit interview costs for everybody and decrease the length of rank lists.

I really like this idea. I suspect there are a lot of unintended consequences from it that I'm not smart enough to either predict or understand. But it seems like a reasonable, rational way to "stop the insanity".

 

[NotAProgDirector]

How about just limiting the number of interviews you can accept?

This has been discussed also. As you mention, it would require a centralized interview system. Getting everyone to use it (and meeting everyone's needs) is usually felt to be impossible in my informal discussions.

Then, you get to the exceptions. Couples in the couple's match. Research candidates looking at MSTP programs. The hyper competitive fields like Ortho and Derm (although one could argue that as long as all applicants were limited to 10 interviews, that would work out OK).

Plus, Although limiting interviews might help programs somewhat, most programs complain about the number of applicants. Just decreasing interviews may not make them happy.

But, it's an idea that has some merit.

 

[wholeheartedly]

This has been discussed also. As you mention, it would require a centralized interview system. Getting everyone to use it (and meeting everyone's needs) is usually felt to be impossible in my informal discussions.

Then, you get to the exceptions. Couples in the couple's match. Research candidates looking at MSTP programs. The hyper competitive fields like Ortho and Derm (although one could argue that as long as all applicants were limited to 10 interviews, that would work out OK).

Plus, Although limiting interviews might help programs somewhat, most programs complain about the number of applicants. Just decreasing interviews may not make them happy.

But, it's an idea that has some merit.

The idea of an early match sounds pretty intriguing.

 

[Gastrapathy]

Early match could lead to the entire match collapsing. GI match collapsed and took years to be reinstated

 

[mvenus929]

Early match could lead to the entire match collapsing. GI match collapsed and took years to be reinstated

Tell us more.

 

[Gastrapathy]

As the best candidates come off the top, the incentive to stay in the match declines.

https://web.stanford.edu/~niederle/Gastroenterology.pdf

 

[BoardingDoc]

Another interesting article I found: The Gastroenterology Fellowship Match: A Decade Later

 

[NotAProgDirector]

An early stage to the match wouldn't collapse it -- because the number of spots that could be filled early would be limited.

But I'm not convinced it would help anything. If we assume 30% of the spots would be allowed to be early, then 30% of applicants might apply and get an early spot. But that also means that 70% of people would likely apply and then not get a spot -- and I expect that would drive those 70% of people to apply to even more spots.

 

[enalli]

As the best candidates come off the top, the incentive to stay in the match declines.

https://web.stanford.edu/~niederle/Gastroenterology.pdf

This was an interesting read. The only real reason they give for the collapse of the match was the fact that there were fewer applicants than available positions. However, they mention that this was the case for ID as well, and the match didn't collapse for them. I wonder what made ID different. Is it just that they were used to not filling, while GI programs got scared at not filling for the first time ever?

 

[Gastrapathy]

@aProgDirector
I’m not so sure. Early decision will pull top tier applicants away from top programs. The appeal of a simplified process and sure thing will make that inevitable. Those programs will realize that they could fill with all their top choices simply by withdrawing from the match. If you look at how quickly the GI match collapsed once it weakened, I think a couple top programs deciding that early decision changed the playing field against their interests could quickly start the spiral.

 

[Gastrapathy]

@enalli the match collapsed because a small percentage of programs got nervous and abandoned it. That started a spiral and quickly no one could afford to be the last program left holdin the bag.

 

[Raryn]

@aProgDirector
I’m not so sure. Early decision will pull top tier applicants away from top programs. The appeal of a simplified process and sure thing will make that inevitable. Those programs will realize that they could fill with all their top choices simply by withdrawing from the match. If you look at how quickly the GI match collapsed once it weakened, I think a couple top programs deciding that early decision changed the playing field against their interests could quickly start the spiral.

Unlike fellowships, residencies can't withdraw individual spots from the match anymore. Not without fully and completely withdrawing from the whole thing - which is a pretty high bar to cross.

 

[Gastrapathy]

I get it but fundamentally the match is asking the most and least competitive programs to act against their best interest. Making that worse could make them decide to change behavior. Why would being all out be bad for a top 20 program?

The biggest thing protecting the match is inertia. It really sucked for the GI applicants in the early 2000s. You had to take that first exploding offer. Imagine if UCSF or Brigham came calling and offered slots to their top applicants with 2 weeks to decide. They would fill. Then next year 3-4 more programs join them. Then next year, the match is dead.

 

[Raryn]

I get it but fundamentally the match is asking the most and least competitive programs to act against their best interest. Making that worse could make them decide to change behavior. Why would being all out be bad for a top 20 program?

Not really. The match is just as beneficial to the programs as the applicants. I mean yes, it fundamentally uses applicant preference as the primary goal, but programs get to rank their applicants at their leisure after interviewing whoever they want - which gives them the class of the most preferred candidates who wanted to go there.

 

[Gastrapathy]

You don’t think it would be a lot less work to interview and offer in a single step? Sort the files, interview your top 20, hire the ones you like, repeat until full. Way less time and maybe they get a few people who wish they went somewhere else but who cares. The programs sure didn’t mind having no match in GI.

 

[BoardingDoc]

Not really. The match is just as beneficial to the programs as the applicants. I mean yes, it fundamentally uses applicant preference as the primary goal, but programs get to rank their applicants at their leisure after interviewing whoever they want - which gives them the class of the most preferred candidates who wanted to go there.

Disagree. The match fundamentally favors the applicant. If programs could use arbitrary time constraints, individual application schedules etc... They could significantly increase their edge in obtaining the fellows they want. The downside as has been illustrated above is that this leads to an arms race in terms of application schedules. It's an interesting application of game theory. If you can out edge everyone else, you win and everyone else suffers. If everyone plays fair in the match, you don't do as well as you could, but you do better than you would if everyone "cheats" but you aren't the best cheater.

 

[NotAProgDirector]

@aProgDirector
I’m not so sure. Early decision will pull top tier applicants away from top programs. The appeal of a simplified process and sure thing will make that inevitable. Those programs will realize that they could fill with all their top choices simply by withdrawing from the match. If you look at how quickly the GI match collapsed once it weakened, I think a couple top programs deciding that early decision changed the playing field against their interests could quickly start the spiral.

As mentioned, for fellowships this might be a risk. For residencies, they must be all in (or all out), and all US grads must be in, so there really isn't any choice.

But, what you point out is that any time you change the rules, you shuffle the deck of winners and losers. Some people will do better. Some will do worse. Whether your overall solution is "good" or not will totally depend on whom you ask. You are correct that super-top programs might actually do worse in this system, as applicants they might have gotten decide to "play it safe" and match elsewhere in the early decision round.

 

[kchan99]

How about a two-stage match? In stage 1, applications and interviews would be limited. Then once the results are out the presumed more competitive candidates would have matched and are no longer interviewing. In stage 2, there will no limit on number of applications or interviews. This also give unmatched applicants to extremely competitive specialties time to switch specialties if there are few spots remaining in the extremely competitive specialties.

 

[Mad Jack]

It's an interesting article. Not sure I really understand all of their outrage.

I agree with the "only offer interviews to the number of people you can fit". That's just plain decency.

The post interview communication thing is more vague. I agree that programs shouldn't be digging for where students will be ranking them. And no employer should be asking about religion, children, etc. But some of the email examples in that article are totally fine IMHO. Completely banning post interview communication is also stress inducing.

Second visits - If programs decide they want to have a second visit to help decide whom they are interested in, then that's their right. I don't, but others might. Banning them seems extreme.

And how would we police all of this? Anonymous reporting is not very helpful.

Overall, I think it would be best to focus on the interview issue. Although, if we decrease the stress due to this issue, I'm sure something else will be the biggest source of stress.

It can be tough to gauge how deep into the match list you'll go. Psych, for instance, used to struggle to fill and now even backwoods programs are often barely cracking 4 applicants per seat to fill

 

[NotAProgDirector]

How about a two-stage match? In stage 1, applications and interviews would be limited. Then once the results are out the presumed more competitive candidates would have matched and are no longer interviewing. In stage 2, there will no limit on number of applications or interviews. This also give unmatched applicants to extremely competitive specialties time to switch specialties if there are few spots remaining in the extremely competitive specialties.

This is very similar to an early decision round. In a true eraly decision process, each applicant would get only 1 application. In what you're suggesting, they would get some limited number of applications and they wouldn't be binding in any way. As you say, it might allow some competitive applicants to match out early, and allow those applying to Ortho/Derm/Plastics/Etc a chance to do so, then retool their application if they don't match.

But:
1. Sounds like in your system programs wouldn't be limited in how many spots they could fill in the first round. In that case, programs might fill most/all of their spots in the first round. All you might have done is limit the number of applications. If you do limit the percent of spots filled in the first round, then I don't know that it really helps -- as there would still be lots of Ortho/Derm/Vasc spots in the second round.
2. Timeline becomes a real problem. When exactly would this first and second round be?
3. Would I even look at second round applications? Or just try to get people matched in the second round from my first round applicants who didn't match (who clearly are more interested than second round applicants)?

 

[NotAProgDirector]

I was thinking about the timing of a two staged match, and it's always problematic. Either the first stage gets pushed really early and it's almost impossible to interview people, or it's late enough that even people ultimately matching in the first stage end up going on a bunch of interviews in case they don't get a spot in the first stage. Or the second stage gets pushed too late to create onboarding problems. It just seems like it won't work.

But, how about this:

Some number of medical students end up staying at their home program for training. Those programs know those students well -- they have been working with them for years. Interviews would be easy to arrange, and honestly not all that important since they have been "interviewing" for at least their clinical training years.

So, each program could choose ONE medical school that they consider their "prime" school. This might be obvious for university programs, might be less obvious for community programs but could be the school that sends them the most rotators -- and in fact this might drive medical schools to partner with more community programs. This might include carib schools also, if programs where they rotate were to choose them. Then, each of those programs could give some limited number of spots to applicants from that school, early. I could imagine getting my school's applications in Sept (Oct 1 for MSPE), interview them all in early October, and then offer them spots by Nov 1. This would take some percentage of applicants out of the pool right away, there would be no second guessing of where students should "use" their early application (since it's their home program or nothing). It would be completely optional, applicants could simply apply for the match.

This works (reasonably) well for larger programs. If a Derm program only has 2 spots, and they give 1 away each year to internal candidates, that could be a problem. It depends on whether there are lots of Derm programs with only 2 spots. Or, this might be limited to programs with some minimum number of spots (perhaps 4, so no more than 25% of the spots can be given away early).

Hmmm. This is interesting. What's also interesting is that it could be done as a test in a single field, since it doesn't change anything else in the match for anyone.

(Note that this is actually possible in the current match if people trust one another. I can tell my internal candidates that I will rank them at the top of my list so they are methematucally guaranteed to match. But this plan would remove the trust issue, and open this pathway to more programs).

 

[mvenus929]

I was thinking about the timing of a two staged match, and it's always problematic. Either the first stage gets pushed really early and it's almost impossible to interview people, or it's late enough that even people ultimately matching in the first stage end up going on a bunch of interviews in case they don't get a spot in the first stage. Or the second stage gets pushed too late to create onboarding problems. It just seems like it won't work.

But, how about this:

Some number of medical students end up staying at their home program for training. Those programs know those students well -- they have been working with them for years. Interviews would be easy to arrange, and honestly not all that important since they have been "interviewing" for at least their clinical training years.

So, each program could choose ONE medical school that they consider their "prime" school. This might be obvious for university programs, might be less obvious for community programs but could be the school that sends them the most rotators -- and in fact this might drive medical schools to partner with more community programs. This might include carib schools also, if programs where they rotate were to choose them. Then, each of those programs could give some limited number of spots to applicants from that school, early. I could imagine getting my school's applications in Sept (Oct 1 for MSPE), interview them all in early October, and then offer them spots by Nov 1. This would take some percentage of applicants out of the pool right away, there would be no second guessing of where students should "use" their early application (since it's their home program or nothing). It would be completely optional, applicants could simply apply for the match.

This works (reasonably) well for larger programs. If a Derm program only has 2 spots, and they give 1 away each year to internal candidates, that could be a problem. It depends on whether there are lots of Derm programs with only 2 spots. Or, this might be limited to programs with some minimum number of spots (perhaps 4, so no more than 25% of the spots can be given away early).

Hmmm. This is interesting. What's also interesting is that it could be done as a test in a single field, since it doesn't change anything else in the match for anyone.

(Note that this is actually possible in the current match if people trust one another. I can tell my internal candidates that I will rank them at the top of my list so they are methematucally guaranteed to match. But this plan would remove the trust issue, and open this pathway to more programs).

Would you, as a PD, then feel like a backup if an applicant chose to go through the match rather than match with you? I loved all the faculty at my Med school for my specialty. And I would have gotten great training there. But, I wanted a bigger program for many reasons, and wanted a place without fellows. I matched at the program I ranked one slot above my home program.

 

[NotAProgDirector]

Would you, as a PD, then feel like a backup if an applicant chose to go through the match rather than match with you? I loved all the faculty at my Med school for my specialty. And I would have gotten great training there. But, I wanted a bigger program for many reasons, and wanted a place without fellows. I matched at the program I ranked one slot above my home program.

I wouldn't, not for an internal candidate. I don't expect most of my internal candidates to do this, happy to consider them in the full match.

 

[DoctwoB]

What if both applications and interview offers were ranked or tiered? Say that you could mark 10 applications as tier 1, 10 as tier 2, and so on down. Programs could offer 10 interviews as tier 1 and so on. These rankings are visible to both applicants and programs. No restrictions on numbers of applications or interviews, just more transparency. Applicants can “inform” programs that they are their top priority and more likely to want to go there. Programs will be more likely to target their interviews towards applicants who are using higher priority applications on them, e.g a mid tier program would rather interview a mid tier applicant who sends them a top 10 application then a top applicant who sends them a top 50 application. Likewise applicants could focus on interviews from programs that “want” them more. This way there are no hard and fast limits on applying but more transparency about the diminishing returns of over applying and over interviewing.

 

[Raryn]

What if both applications and interview offers were ranked or tiered? Say that you could mark 10 applications as tier 1, 10 as tier 2, and so on down. Programs could offer 10 interviews as tier 1 and so on. These rankings are visible to both applicants and programs. No restrictions on numbers of applications or interviews, just more transparency. Applicants can “inform” programs that they are their top priority and more likely to want to go there. Programs will be more likely to target their interviews towards applicants who are using higher priority applications on them, e.g a mid tier program would rather interview a mid tier applicant who sends them a top 10 application then a top applicant who sends them a top 50 application. Likewise applicants could focus on interviews from programs that “want” them more. This way there are no hard and fast limits on applying but more transparency about the diminishing returns of over applying and over interviewing.

People barely understand the match as it stands and play mind games with themselves about how it works - this system sounds like a nightmare to navigate.

 

[Stagg737]

The last option I've discussed is some sort of early decision process. One application to a program you're very interested in as a binding application -- if they take you, you're matched. Programs would only be able to fill some maximum of their spots this way (perhaps 30%). This would remove some of the most competitive applicants from the pool up front, saving all of those interview slots.

How about just limiting the number of interviews you can accept?

You would need a completely centralized interview invitation and scheduling system but I envision it like this:
- Students apply to as many programs as they want, but there is an earlier deadline (say September 30)
- On a specific date (say October 15), programs offer interviews through the system
- Within 2-3 days, students select up to 7 invitations to accept
- There will be a second round of interviews offered (for example, by October 25)
- Students get an additional 3 interview slots and can now schedule up to 10 interviews total
- After this point, programs can offer interviews at any time, but students cannot attend more than 10 interviews

Imo a combination of the above seems like a solid option. AMCAS used to do an early admission where individuals could apply to a single school as an early admission process and if accepted had to commit to the program and would be withdrawn from the rest of applications. If they got it then they were done and the rest of the application process would continue. Med schools would know how many more positions they had open and applicants would know if they had to keep applying or if they could sit back. This would also fix the problem @Gastrapathy mentioned of elite programs opting out. If they want to completely fill in the "early decision" match, then they could. After that it would allow both programs and applicants to know what was still available. Elite applicants and those committed to a single program would be matched and everyone else could have a limited number of interviews for the remaining positions. This would remove applicants like myself, who knew my #1 before even interviewing and ended up there from taking up interview spaces at other programs that could have gone to other candidates.

I also think this solves the problem of individuals who go on 20+ interviews wasting people's time and resources. There may be the unicorn out there who unexpectedly finds their dream program in the program they initially thought was last out of 15+ programs and matches there, but for almost everyone I've talked to interview burnout hits at or before program #10. So most of the time the later interviews for those with many are a waste anyway.

I get it but fundamentally the match is asking the most and least competitive programs to act against their best interest. Making that worse could make them decide to change behavior. Why would being all out be bad for a top 20 program?

The biggest thing protecting the match is inertia. It really sucked for the GI applicants in the early 2000s. You had to take that first exploding offer. Imagine if UCSF or Brigham came calling and offered slots to their top applicants with 2 weeks to decide. They would fill. Then next year 3-4 more programs join them. Then next year, the match is dead.

The issue of a program not participating in the match is that if an applicant participates in the match they can't apply to outside programs from what I understand. So if elite programs said they weren't participating, then only the elite applicants were were risk-takers would apply and the program would likely miss out on some outstanding candidates. With the current match rules in place, a program not participating hurts both the applicants (as they'd have to take a huge gamble by removing themselves from the match to apply to that elite program) and the programs (as they'd lose out on interviewing some elite caliber candidates). I understand the point you're making, but with the current rules there's a very high risk to both applicants and programs to leaving that I think most would find not worth it (which can also be fixed by having the early decision process in place).

There should not be limitations on applicants. I understand it's a rat race and neurotic people want to do 10-15 interviews in IM to make sure they match, but there is not a great solution for that.

There should be teaching on the minimum number of interviews necessary to guarantee application success (that number is like 5-6 for a USMG in IM).

The issue is not the number of applications, but rather the amount of interviews. Limiting interviews is not a great solution either, IMO.

Having an early decision style of interview would probably be the best way to manage most of the sheer numbers.

Agree that it's interviews and not applications that should be limited. I applied to 75 programs and interviewed at ~10. If I'd only been allowed to apply to 20 I probably would have only had about 4, maaaaaybe 5 interviews. Additionally, 3 of my top 4 in my rank list I wouldn't have interviewed at. I would be okay with limiting the number applications only if the limit were very high (around 100 programs). I think the better option for limiting a shot-gun application would be to allow for the first x number of programs to be extremely cheap (first 20-25), the next large amount to be relatively cheap (first 100 or so), and then make it prohibitively expensive to apply beyond that.

I also think having programs explicitly outline hard and suggested criteria for applicants and interviewees in FREIDA or some other source (minimum board scores, class rank or AOA status, research, commitment to a certain aspect of care or setting, etc) and adhering to it would help some.


The match certainly has it's problems, but overall I think it's a pretty decent system given the alternatives. I may be biased since it worked out for me, but I haven't really heard any other realistic systems that I think are that much better.

 

[NotAProgDirector]

What if both applications and interview offers were ranked or tiered? Say that you could mark 10 applications as tier 1, 10 as tier 2, and so on down. Programs could offer 10 interviews as tier 1 and so on. These rankings are visible to both applicants and programs. No restrictions on numbers of applications or interviews, just more transparency. Applicants can “inform” programs that they are their top priority and more likely to want to go there. Programs will be more likely to target their interviews towards applicants who are using higher priority applications on them, e.g a mid tier program would rather interview a mid tier applicant who sends them a top 10 application then a top applicant who sends them a top 50 application. Likewise applicants could focus on interviews from programs that “want” them more. This way there are no hard and fast limits on applying but more transparency about the diminishing returns of over applying and over interviewing.

Interesting you'd mention this, since it was just published this month as an idea with a mathematical model of how it would work out: Article

There are also three editorials about what could be done: Editorial 1, Editorial 2, and Editorial 3

The idea was to allow residents to notify programs if they were "preferred", which each applicant was allowed to do for 10 programs. You could apply to as many programs as you wanted, the rest would not be "preferred". The model assumed that if programs had to review less applications, they would do a more holistic review. What the paper suggests is that allowing this was overall a good -- the only people "hurt" by the model were the top students, they got less interview offers BUT they were already getting so many interview offers that "less" was meaningless, since they couldn't go on all of them anyway.

The editorials mostly support this notion, although also mention the idea of a match for interview offers, limits on the number of interviews, and some other ideas.

My main quibble with the article is how they handled the "preference". The way their model works, each student gets a score from each program and that determines whom they interview and rank. When a student picked a program as a preference, they added some points to their score -- per their description, about as much as 10 points on the USMLE. But what happens if programs decide to ONLY interview applicants who indicate a preference? Then, all we've done is create a system with a limit of 10 applications. (This is mentioned in one of the editorials)

 

[deleted480308]

Interesting you'd mention this, since it was just published this month as an idea with a mathematical model of how it would work out: Article

There are also three editorials about what could be done: Editorial 1, Editorial 2, and Editorial 3

The idea was to allow residents to notify programs if they were "preferred", which each applicant was allowed to do for 10 programs. You could apply to as many programs as you wanted, the rest would not be "preferred". The model assumed that if programs had to review less applications, they would do a more holistic review. What the paper suggests is that allowing this was overall a good -- the only people "hurt" by the model were the top students, they got less interview offers BUT they were already getting so many interview offers that "less" was meaningless, since they couldn't go on all of them anyway.

The editorials mostly support this notion, although also mention the idea of a match for interview offers, limits on the number of interviews, and some other ideas.

My main quibble with the article is how they handled the "preference". The way their model works, each student gets a score from each program and that determines whom they interview and rank. When a student picked a program as a preference, they added some points to their score -- per their description, about as much as 10 points on the USMLE. But what happens if programs decide to ONLY interview applicants who indicate a preference? Then, all we've done is create a system with a limit of 10 applications. (This is mentioned in one of the editorials)

but why would a program use an interview slot with someone they know doesn't even have them in the top 10. I think for all but the most uncompetitive places it would become an effective 10 program limit

 

[Dave1980]

OP, the title of your post sounds like it came from buzzfeed

 

[NotAProgDirector]

but why would a program use an interview slot with someone they know doesn't even have them in the top 10. I think for all but the most uncompetitive places it would become an effective 10 program limit

Yes, that was what I was attempting to say, apparently poorly. If you make the "preference" weight 100% (i.e. only invite people who applied with a preference), then it becomes essentially a 10 application limit. And would likely completely change the results of the study. Assuming that indicating a top 10 preference would equate to a 10 point increase on the USMLE seems ridic small to me.