How competitive is Radiology?

[Lifeblood_20]

Current M1. So far in med school I have just focused on passing and haven't done any meaningful research or started building up my CV for residency apps. From my still limited exposure, I feel that radiology may be a good fit. I was wondering how competitive it is and if research is very important to matching? If not, is there something else more important? I personally am NOT passionate about research and want to do just enough to match well.
Of course, I am keeping an open mind, but I wanted to have an idea early of how hard I should go for research (or anything else important to matching Rads).

---

Members don't see this ad.

 

[ChordaEpiphany]

My understanding is that radiology is quite competitive at top programs and competitive overall, but that most people do match somewhere. It's not quite as competitive as fields like derm or some subspecialty surgery fields that have a far more limited number of residencies and spots.

It has a 95.5% match rate among US MDs. That's higher than everything aside from IM, FM, neuro, peds, and rad onc. The nightmare scenario isn't so much not matching (though it's definitely possible), but matching in an undesirable location. Pretty much any average MD student will match.

I only expect rads to get less competitive in the next 3-4 years. Right now it's sort of a golden specialty. Easy to match. Very high reimbursement. Possibility for procedural work if you hate reading all the time. Tons and tons of demand with a relatively cornered market. However, there are a lot of ways that gravy train could come to an end (e.g., AI, allowing outsourced reads, etc...). Regardless of whether those things have any basis in reality, the fear of those things happening will be a big point of discussion among potential rads applicants. I think for some med students, who are extremely risk averse, it will be enough to dissuade them and rads will likely get even easier to match.

 

[Dr G Oogle]

Agree with competitiveness assessment above, but need to debunk sons stuff regarding threats. Re: AI, is more of a pipe dream than a threat. You’re war more likely to negate the usefulness of a PCP with AI then a radiologist. You only need to look at EKG and the inability to build an AI to correctly read a 2D output with relatively little variance in outcomes; the tech is no where near to being able to read an fracture X-ray let alone an mri or CT. Re: outsourcing : I currently am not legally allowed to do a telehealth visit with my own patient who’s spending the winter in another state without having a license to practice medicine in that state. It is highly improbable that the US will grant a medical license to a non citizen who practices in a foreign country.

The big threats to radiology are falling reimbursement and oversupply which are problems in any field and rads is not more prone to it than any other field. Rads is more prone to private equity type of stuff, which overall can lead to lower reimbursement and job satisfaction. It’s generally hard to fire a doctor in general, but the threshold to fire a rad is probably lower than say a surgeon who brings in new patients and generates lots of downstream revenue.

None of the above are new, even PE has been applying sticky fingers to rads for decades. Rads competitiveness will likely vary with job cycles (currently hot) but will remain overall stable when looked through a long term lens.

At least do some research to demonstrate interest.

 

[Maximus2020]

Couldn't agree more with Dr. G's post - really phenomenal. The AI fear is overblown and meaningless. US trained rads will always have a leg up on the market.

Medicine and surgery are completely dependent on Radiology so you will always have a job. As to what to do to make yourself competitive, research is helpful in all areas but it's not an end all/be all. No golden horse exists, just the basics - strong Step scores, Honors, AOA, etc. A good LOR from a rad elective taken early on in 3rd or 4th year would prove worthwhile.

Note this is for diagnostic (DR) rads. If you dream of IR, then nailing your surgery rotation and doing an away rotation in IR is your go-to if gunning for the top programs. Two separate application/rank processes.

 

[Angus Avagadro]

Rads is definitely changing. Can't recruit to our area as reimbursement is low. Local residents can got to any state bordering ours and get a significant pay increase. Rads is becoming very sub-specialty oriented. At our local trauma center, it takes 3 radiologists to read images from a trauma case. A chest, a body imager, and a neuro radiologist. MSK, womens imagers and interventionalists usually fill out the department. Often, you are only allowed to read within your subspecialty. This requires more people to be on call for stroke eval, MRA's etc.. AI is not a big factor for me as I dont think AI can evaluate all the tiny lung nodules out there and determine what needs to be worked up. I think the biggest threat is outsourcing to local hospital networks with all the subspecialty coverage or those groups outside the country that will read for much less money. When the culture changes for your group, i.e., bought by a local network or the hospital ourtsources the contract for radiology services to another group, you might end up selling you house and pulling your kids out of school due to restrictive covenants preventing you from working withing a certain radius of any facility the group covers. Medicine is changing rapidly and the linear career, i.e., staying in one job for 30 years is going the way of the dinosaur.

 

[odyssey2]

Rads is definitely changing. Can't recruit to our area as reimbursement is low. Local residents can got to any state bordering ours and get a significant pay increase. Rads is becoming very sub-specialty oriented. At our local trauma center, it takes 3 radiologists to read images from a trauma case. A chest, a body imager, and a neuro radiologist. MSK, womens imagers and interventionalists usually fill out the department. Often, you are only allowed to read within your subspecialty. This requires more people to be on call for stroke eval, MRA's etc.. AI is not a big factor for me as I dont think AI can evaluate all the tiny lung nodules out there and determine what needs to be worked up. I think the biggest threat is outsourcing to local hospital networks with all the subspecialty coverage or those groups outside the country that will read for much less money. When the culture changes for your group, i.e., bought by a local network or the hospital ourtsources the contract for radiology services to another group, you might end up selling you house and pulling your kids out of school due to restrictive covenants preventing you from working withing a certain radius of any facility the group covers. Medicine is changing rapidly and the linear career, i.e., staying in one job for 30 years is going the way of the dinosaur.

Mostly bad changes it sounds like?

 

[Angus Avagadro]

Mostly bad changes it sounds like?

I don't think rads is unique when it comes to changes that are not Dr friendly. Ther is no mid-level encroachment and good opportunities exist if you are flexible about where you want to live. The next 5 yrs are hard to predict, let alone 30 yrs. Tesla started self driving cars in 2014 and they are still not perfected. With uncertainty abounding, I would pursue the specialty you would enjoy the most and add as much flexibility to your choices as an attending as possible.

 

[jambro]

My understanding is that radiology is quite competitive at top programs and competitive overall, but that most people do match somewhere. It's not quite as competitive as fields like derm or some subspecialty surgery fields that have a far more limited number of residencies and spots.

It has a 95.5% match rate among US MDs. That's higher than everything aside from IM, FM, neuro, peds, and rad onc. The nightmare scenario isn't so much not matching (though it's definitely possible), but matching in an undesirable location. Pretty much any average MD student will match.

I only expect rads to get less competitive in the next 3-4 years. Right now it's sort of a golden specialty. Easy to match. Very high reimbursement. Possibility for procedural work if you hate reading all the time. Tons and tons of demand with a relatively cornered market. However, there are a lot of ways that gravy train could come to an end (e.g., AI, allowing outsourced reads, etc...). Regardless of whether those things have any basis in reality, the fear of those things happening will be a big point of discussion among potential rads applicants. I think for some med students, who are extremely risk averse, it will be enough to dissuade them and rads will likely get even easier to match.

Those potential issues have been around forever though so they’re not going to make it more uncompetitive. And in the last few years and probably for the foreseeable future, the rads job market has been fire. Plus one of the few jobs in medicine where some portion can be done from home when we’ve seen how nice Wfh has been for so many people. So why wouldn’t it get more competitive?

 

[Matthew9Thirtyfive]

Agree with competitiveness assessment above, but need to debunk sons stuff regarding threats. Re: AI, is more of a pipe dream than a threat. You’re war more likely to negate the usefulness of a PCP with AI then a radiologist. You only need to look at EKG and the inability to build an AI to correctly read a 2D output with relatively little variance in outcomes; the tech is no where near to being able to read an fracture X-ray let alone an mri or CT. Re: outsourcing : I currently am not legally allowed to do a telehealth visit with my own patient who’s spending the winter in another state without having a license to practice medicine in that state. It is highly improbable that the US will grant a medical license to a non citizen who practices in a foreign country.

The big threats to radiology are falling reimbursement and oversupply which are problems in any field and rads is not more prone to it than any other field. Rads is more prone to private equity type of stuff, which overall can lead to lower reimbursement and job satisfaction. It’s generally hard to fire a doctor in general, but the threshold to fire a rad is probably lower than say a surgeon who brings in new patients and generates lots of downstream revenue.

None of the above are new, even PE has been applying sticky fingers to rads for decades. Rads competitiveness will likely vary with job cycles (currently hot) but will remain overall stable when looked through a long term lens.

At least do some research to demonstrate interest.

To expand on the AI thing, the studies that are often cited showing AI is “better than radiologists” are extremely limited. They are given hundreds and thousands of images from the same machines and techs and then tested on images from the same machines and techs looking for the same single pathologies. They have yet to produce a prospective study where AI is able to reliably read images from a test set that is not effectively identical to the training set. The idea that AI is a threat is laughable.

 

[Lawpy]

To expand on the AI thing, the studies that are often cited showing AI is “better than radiologists” are extremely limited. They are given hundreds and thousands of images from the same machines and techs and then tested on images from the same machines and techs looking for the same single pathologies. They have yet to produce a prospective study where AI is able to reliably read images from a test set that is not effectively identical to the training set. The idea that AI is a threat is laughable.

Eh anything can happen in 20 years and medicine in its current form is barreling towards a disaster at all levels. AI may suck now but in 20 years time, it can be a real threat

 

[Matthew9Thirtyfive]

Eh anything can happen in 20 years and medicine in its current form is barreling towards a disaster at all levels. AI may suck now but in 20 years time, it can be a real threat

Anything is possible. In 20 years, dogs could be voting. They already run for mayor.

 

[redsox93]

My understanding is that radiology is quite competitive at top programs and competitive overall, but that most people do match somewhere. It's not quite as competitive as fields like derm or some subspecialty surgery fields that have a far more limited number of residencies and spots.

It has a 95.5% match rate among US MDs. That's higher than everything aside from IM, FM, neuro, peds, and rad onc. The nightmare scenario isn't so much not matching (though it's definitely possible), but matching in an undesirable location. Pretty much any average MD student will match.

I only expect rads to get less competitive in the next 3-4 years. Right now it's sort of a golden specialty. Easy to match. Very high reimbursement. Possibility for procedural work if you hate reading all the time. Tons and tons of demand with a relatively cornered market. However, there are a lot of ways that gravy train could come to an end (e.g., AI, allowing outsourced reads, etc...). Regardless of whether those things have any basis in reality, the fear of those things happening will be a big point of discussion among potential rads applicants. I think for some med students, who are extremely risk averse, it will be enough to dissuade them and rads will likely get even easier to match.

I think the opposite will happen and it’s going to get much more competitive. This seems to already be the case at least anecdotally this year I guess we’ll have to wait until charting the outcomes comes out. I remember seeing early this cycle that applications were up 30% for rads this year compared to last

 

[MDMechEngr]

one day we all will be able to board an aircraft and fly to our destination without a pilot and copilot flying that plane. Heck we can do this already. I dont know about you, but I still still want a pilot there just incase because **** happens.

 

[ChordaEpiphany]

Agree with competitiveness assessment above, but need to debunk sons stuff regarding threats. Re: AI, is more of a pipe dream than a threat. You’re war more likely to negate the usefulness of a PCP with AI then a radiologist. You only need to look at EKG and the inability to build an AI to correctly read a 2D output with relatively little variance in outcomes; the tech is no where near to being able to read an fracture X-ray let alone an mri or CT. Re: outsourcing : I currently am not legally allowed to do a telehealth visit with my own patient who’s spending the winter in another state without having a license to practice medicine in that state. It is highly improbable that the US will grant a medical license to a non citizen who practices in a foreign country.

The big threats to radiology are falling reimbursement and oversupply which are problems in any field and rads is not more prone to it than any other field. Rads is more prone to private equity type of stuff, which overall can lead to lower reimbursement and job satisfaction. It’s generally hard to fire a doctor in general, but the threshold to fire a rad is probably lower than say a surgeon who brings in new patients and generates lots of downstream revenue.

None of the above are new, even PE has been applying sticky fingers to rads for decades. Rads competitiveness will likely vary with job cycles (currently hot) but will remain overall stable when looked through a long term lens.

At least do some research to demonstrate interest.

To expand on the AI thing, the studies that are often cited showing AI is “better than radiologists” are extremely limited. They are given hundreds and thousands of images from the same machines and techs and then tested on images from the same machines and techs looking for the same single pathologies. They have yet to produce a prospective study where AI is able to reliably read images from a test set that is not effectively identical to the training set. The idea that AI is a threat is laughable.

Agreed the threats are overblown in present day. Like I said, regardless of whether those things have a basis in reality, plenty of people are scared of them. In medicine you gain a hyperspecialized skill and are then subject to whatever changes the market/government throw at you for the next 30 years with almost no alternative pathway should things go sour in your specialty. Med students are understandably wary of the future.

I wouldn't completely disregard the effect of AI-enhanced radiology for an M1 looking 40+ years into the future. It doesn't need to be better than radiologists. All it really needs to do is create more confident/competent radiologists who can read and dictate faster. In 10 years it's hard to imagine anything changing on that front. In 20-30 years, still well within the window of a current M1's career, you could imagine a PE-owned world of radiology where AI makes it safer/more profitable for PE firms to push a smaller number of radiologists to read significantly faster, creating an effective oversupply.

All it really takes to imagine this world is to look back at the past 20 years of tech and 20 years before that and recognize that we're on an exponential curve. That said, medical doctors are facing threats from mid-levels, surgeons are facing threats from interventional medical doctors, and everyone is facing threats from PE, government, COVID, etc...

Agreed with @Lawpy. It's hard to be optimistic about a career in healthcare right now. I don't think rads is an exception to that.