Moral Injury in Physicians (NY Times)

As the focus on revenue and the adoption of business metrics has grown more pervasive, young people embarking on careers in medicine are beginning to wonder if they are the beneficiaries of capitalism or just another exploited class. In 2021, the average medical student graduated with more than $200,000 in debt. In the past, one privilege conferred on physicians who made these sacrifices was the freedom to control their working conditions in independent practices. But today, 70 percent of doctors work as salaried employees of large hospital systems or corporate entities, taking orders from administrators and executives who do not always share their values or priorities.

Just another reason why physician parents are discouraging their kids from going into medicine. I tell my kids and any young people that nowadays, the only reason to go into medicine is if you feel that you have a calling for it, like a calling to go into the priesthood. And this is why all of us adults, especially the docs, are begging kids on here to see reason and keep their undergrad expenditure as low as possible if they’re planning on med school.


I have this discussion with everyone of my patients who is an autonomous provider of any kind, physicians, dentists, optometrists, veterinarians, etc. The consensus is that it costs too much and that as you also suggest, one need be VERY careful about debt. One of my patients made his last loan payment in his 60s. Another said had he calculated the debt load in advance, he’d have been an electrician.


I went to the same general practitioner for more than 30 years. Originally he was independent. At some point he switched to be part of a big corporate practice. He said that what made him switch was when 3 1/2 full time employees were not enough to keep up with the paperwork for just him.

My daughter who is studying to be a DVM has also told us a few times that most of the students in her program are taking on way too much debt, and that most do not want to talk about it. The few who are taking on little or no debt also do not want to talk about it, partly because they do not want to make their friends feel bad about it. She has also thanked me for not letting her take on any debt at all for her bachelor’s degree (which if possible is indeed something that several of us frequently recommend here on CC). DVMs also are increasingly likely to be working for a corporate practice.

And we have seen some ridiculous stalling from insurance companies, such as stalling a PET scan for a cancer patient, and then approving the PET scan but not approving the injection that you need in order have the PET scan. I would expect this to be nearly as frustrating for the doctors as it is for the patients, and the doctors get to see it all the time.


There is no perfect system, but the boards and stocks that control every aspect of our for-profit healthcare are the actual “death panels”. Literally.


As a physician now under control of a financial firm, what is good for the patients does not necessarily match with what is good for metrics. Even though it was (and still is) my “calling” ever since I was a child, it gets more and more difficult to fight the daily fight to be an advocate for my patients at the expense of profit. Yes the patient needs this test I recommended, yes I need more office staff to keep the office running smoothly, yes being asked to work Saturdays and longer hours is causing burnout, yes I am still repaying student loans 23 years later, and yes I know that if I take more time with each patient the potential profit (that I no longer see because I’m salaried and there have been years of excuses as to why there is no bonus or distribution) to admin goes down.

This following is from a white paper written by a colleague (David West, MD). It might explain the reasons that patients are seeing a delay on their radiology exam reports. It also touches on some of the reasons behind the moral crisis of one speciality (radiology):

Radiology is in the middle innings of a transformational change. This change has been slowly occurring,
almost imperceptibly for two decades, made possible by technological advancements in data storage,
manipulation, and transmission. The pace of change was markedly accelerated by the Covid pandemic
when there was a realization by even the most recalcitrant radiologists that the majority of their job
could be performed with greater productivity and less stress at a place and time of their choosing with
the benefit of reduced virus exposure.
Burnout in radiologists is among the highest in all of medicine with 45% burnout in 2018, pre-pandemic.
In 2021 radiology had the 5th highest burnout of 23 specialties. As the population ages over the next
decade, imaging volume is predicted to grow annually at 5-7%, and the workforce of active radiologists
is predicted to shrink.
This paper will review the current relevant data trends, discuss major factors influencing the current
transformation, and finally provide potential outcomes of the transformation.
Demographic trends of radiologists in the United States:
1990: 21,900 full and part-time physicians, 47% less than 45 years old (1993)
2022: 20,970 full and part-time physicians, 18% less than 45 years old, 53% over 55 years old
In q1 2019 there were 1000 radiology job postings; in q1 2022 there were 3000
Projected shortage of radiologists according to Jackson+Coker in 2025 in the “TENS of THOUSANDS”
Imaging volume trends of radiologists in the United States:
2008 14,900 studies per FTE radiologist
2018 26,457 studies per FTE radiologist
1998 26 million CT exams performed
2019 91.4 million CT exams performed
1998 to 2019 - An increase of 65.4 million CT exams, 3,100 more CTs PER RADIOLOGIST PER YEAR
Imaging volume trends in Riverside Radiology:
2019 to 2020 -10.6% (first decrease in decades, secondary to covid)
2020 to 2021 +20.1%

Major drivers of the transformation:

Factors influencing the transformation can be divided into three categories. The first is the rapidly
accelerating commoditization of radiology. The second is the overwhelming increase in imaging
volumes with a simultaneous dramatic increase in the number of images per study. The third is a
generational change in younger radiologists as well as those towards the end of their career.
The first major driver of the transformation is the commoditization of radiology made possible by the
digitization of all things. There are a number of fully developed platforms which allow the storage and
bidirectional data transfer required for any facility to send studies from any modality to any workstation
anywhere in the world in real time. Advantages of these platforms include real time load balancing
across all available radiologists on that enterprise platform. The platforms also facilitate subspecialty
radiology interpretations regardless of geographic location or size of the facility which performed the
imaging exam.
The downside is the work can now be divided literally by the exam, providing a constant and never-
ending steam of exams directly to the radiologist who is now paid piecemeal for each exam they
interpret. In other words, we are no longer paid by the day to cover a particular facility; we are paid by
the exam or by productivity which is measured for every single study that is interpreted. If you Interpret
more studies or interpret studies which have inconsistencies in wRVU, you get paid more. If you get
bogged down with complex cases with literally thousands of images, you get paid less.
In the past, radiologist workflow was variable, busy times punctuated by slow times with only critical
exams after hours and on the weekends. This allowed radiologists to find a work life balance measured
over the course of the week. In the current environment, imaging is performed without regard for the
radiologist. There is a fire hose of exams to be read every minute of every day, 60 minutes an hour, 24
hours a day, 7 days a week, 365 days a year.
The commoditization of radiology has also dramatically affected the ability to recruit for on-site
radiology positions. There are inherent inefficiencies for on-site radiologists where we essentially
function as an employed physician, subject to constant and often necessary interruptions from ancillary
staff, referring physicians, and administration to provide the best possible level of care. After careful
discussions with many of the onsite radiologist across our enterprise, we estimate there is a 20-30%
reduction in productivity related to being on-site. This does not account for non-productivity factors of
markedly increased stress related to interruptions, interactions, and juggling multiple problems
simultaneously. It also does not take into account the inconvenience of travel to and from the site and
the requirement to live in a less or even undesirable geographic location.
Contrast this with teleradiology where a radiologist is presented with one case at a time, with minimal
to no interruptions. This allows near total focus on the study and can be done in the comfort of home.
There is little incentive to work 2nd or 3rd shift as there is ample available daytime imaging work. Given
the current imaging volumes and workforce mismatch, as a radiology provider, there is little leverage to require participation in 2nd or 3rd shift work for teleradiogists. Salary requirements to entice 3rd shift
participation can exceed double 1
st shift. In the past, money could be shifted from profitable daytime
imaging to pay for unprofitable night time imaging and for shift differentials. In the commoditized
world, if we reduce the pay per RVU for day time imaging to supplement night time imaging, we are no
longer competitive in the marketplace for day time imaging. Unlike the employed physician model, we
do not have the ability to fund physician losses from hospital operations.
Increase in workload, imaging volumes, and total number of images:
The second factor driving the transformation is the overwhelming increase in workload reflected in
imaging volumes (i.e. the total number of exams performed) and a staggering increase in image count
(i.e. the number of images which must be viewed for each exam).
In 2000, the average annual workload per FTE was 5000 RVUs averaged for academic and diagnostic
radiologists. In 2021 for our practice, it is over 11,000 RVUs, a 220% increase.
Despite policy and clinical efforts to reduce imaging, from 2000 to 2016 there was 209% growth in CT
imaging and 224% growth in MRI imaging with no change in the number of radiologists to provide the
interpretations. The imaging growth is only one part of the equation. Increasing volumes have been
almost entirely in cross sectional imaging modalities with huge increases in total image count due to
multiplanar reformations, thinner slices, and complex multiphase protocols.
The actual number of images to be reviewed by a radiologist has dramatically increased in the same
period. As data storage and transmission costs have declined exponentially, the number of images has
increased exponentially. In 2005, an average CT exam of the abdomen and pelvis contained 80 images.
In 2022, an average CT exam of the abdomen and pelvis contains 400 images. At some facilities the
same CT exam can contain well over 2000 images. In 2005, a screening mammogram had on average 5
images. In 2022, the average 3d tomographic mammogram has 300 images with some exams over 600
I could not find much data on the increase in the number of images over time, so I reviewed my
personal data. In 2012, an average day consisted of approximately 120 exams with a total image count
of 7000. In a 10 hour work day with 9 hours of interpretation, I had up to 4.62 seconds to review each
image. In 2022, an average day consisted of approximately 150 exams with a total image count of
33,000. In a 10 hour work day with 9 hours of interpretation, I have 0.98 seconds to review each image.
This does not account for the time to dictate, review, and sign off reports. In one exceptionally busy day
in 2022, reading at a facility which has little consideration for total number of images per study, where
both CT and ultrasound exams are routinely over 2000 images each, I reviewed well over 100,000
images, or 0.324 seconds for each image.
The expectation for interpretation is perfection. Missing one finding on a single image of the 100,000
reviewed that day can have disastrous professional and reputational consequences.

Generational change:
The third factor driving the transformation is the generational change in attitudes of newly trained
radiologists who are not interested in the hamster wheel of productivity. They demand a work life
balance over all else, with that balance decidedly slanted towards life over work as compared to the last
generation of radiologists. In this cohort, we see several trends.
The first trend is finding a work life balance by maximizing their income for the work they do but
minimizing the total amount of work performed by decreasing the total number of shifts or by reducing
total daily productivity. Rather than a goal to complete all the work on the worklist, as past generations
have done, the goal is an individually set productivity level. Once met, they are done for the day.
The second trend comes in the form of geographic location. This cohort understands they can command
a similar salary to on-site radiologists, but live wherever they like. The newest generation fully
understands the stress and reduced productivity of working on-site rather than remote radiology, and
they have opted almost wholesale for remote radiology. Less than 5% of our recruits opt for on-site
The third trend is a move away from the do-it-all “generalist” radiologist to super sub specialization.
There is an increasing demand to interpret only exams for which they have sub specialized training.
This newest generation is fully aware of the technology available to them and understands, in light of
the commoditization of the specialty, that their labor is fully mobile. They can be signed up with
multiple radiology service providers simultaneously. As the shortage worsens, physicians can
increasingly dictate the number of shifts, times of day, and holiday and vacation schedule to the
provider. Colloquially we call this flexibility. As the radiology service provider, we have little leverage
over either the radiologist or our competitors other than meeting teleradiologist lifestyle or income
Charting the Future – Potential solutions and pathways
There are a number of pathways to solve the current shortage. The solution will be a combination of
these and other pathways as no pathway alone provides a solution to fully address the demographic,
workload, and generational challenges. Regardless of the pathway, the simple imbalance of supply and
demand will result in increased cost to the facilities collecting the technical component for the imaging
exams as competition for a limited workforce accelerates.
The first potential pathway is the fragmentation of diagnostic imaging interpretation. In this pathway,
referring clinicians will use the shortage as a wedge to pick off profitable segments of the imaging
volume by promising improved turnaround times and/or quality. This pathway in the end will prove to
be costly to the facilities performing the exams, as the profitability of radiology relies on day time
interpretations from third party payers and Medicare, allowing the absorption of the unprofitable
exams. As this profitable business is redirected, the facilities will increasingly lose affordable access to
radiology interpretation as the shortage intensifies. Access to radiology to meet governmental and
facility requirements will increasingly become more expensive.

The second potential pathway involves the capture of desired radiology services through payment of
subsidies. This favors large and more profitable facilities who will be able to provide necessary and
desired radiology services regardless of the cost structure. As the industry moves towards a subsidy
model, these partners will be able to share in the increased cost to provide mission-critical and mission-
important services. This will unfortunately be at the expense of smaller facilities with low to no margins
that will have to make changes to radiology service; including choosing on-site vs off-site radiology as
well as having to take into account the profitability of subsets of radiology services to fit their budgets.
The third potential pathway is for the ordering providers and facilities to understand, recognize, and
work towards a more sustainable path. This is a win for healthcare as a whole and for the patients with
reduced cost as well as reduced exposure to radiation and or inconvenience. The end result is to reduce
unnecessary imaging exams as well as the number or images per exam. Unnecessary exams include
contemporaneous replication of exams due to insufficient systems to obtain previous outside or even
own healthcare system imaging results, ordering of exams which do not follow published and vetted
guidelines, as well as unnecessary imaging ordered by advanced practice providers before patients are
triaged or seen by the physician due to time or personnel constraints. It also involves reduction in total
number of images by working within established protocols developed by the radiologist. While this
dovetails nicely with the very slow transition to value-based care, the barrier is our current fee for
service model.
The fourth potential pathway is a reduction or reset of service expectations. Increases in turnaround
times allows for a smoothing of the peaks and valleys of the imaging volumes, reduces the number of
required radiologists to handle the peaks, thereby increasing scheduling flexibility. As the shortage
continues, turnaround times have and will undoubtedly continue to increase. I do not know when
equilibrium will be reached, but I predict 1-2 hour turnaround time for emergent studies, 4-8 hours for
inpatient exams, and 72-96 hours for outpatient exams. There will be an increase in outliers secondary
to lack of bandwidth to handle large imaging volume surges.
The fifth potential pathway is a tiered system of interpretations based on payment for speed. Imaging
traffic could be segregated to dedicated lanes of radiologists, with a surcharge paid to put imaging
exams into a “faster” lane. Alternatively, imaging traffic could be managed by “surge” pricing, a.k.a.
Uber. In this model, the radiologists would be paid increasingly more as demand increases and less as
demand decreases. Correspondingly, the cost to facilities would vary with demand in real time.
The sixth potential pathway is the use of AI (Artificial Intelligence) to augment and or increase
radiologist productivity. The currently available data shows AI decreases, rather than increases
productivity. Anecdotally, I have found current implementations of AI to be of limited value in terms of
productivity. However, our current implementation of AI to elevate critical exams for intracranial
hemorrhage and pulmonary embolism in a very busy workload environment has proven very successful,
but technical and institutional barriers have made implementation slower than expected.
The seventh potential pathway is alternative staffing with advanced practice non-physician radiology
providers including nurse practitioners, physician assistants, and radiology assistants. This remains a
controversial subject, supported by some, vehemently opposed by others. Currently this requires
nimble navigation of changing federal, state, and payer rules and regulations.

The eighth and final potential pathway is a segregation of on-site and teleradiology service providers,
with companies and radiologists gravitating to one side or the other as the economics demand. Perhaps
the outcome is teleradiology provided by increasingly larger radiology service providers and on-site
radiology provided by employed physicians with the facilities bearing the full cost of employment as well
as the responsibilities inherent in physician management.
In conclusion, we believe the current path of radiology is unsustainable from a workforce perspective.
The demographics of the workforce coupled with the mountain of never-ending data is overwhelming.
If a solution is not found, burnout will increase resulting in a downward spiral affecting all of healthcare.
While there remains successful pockets of traditional radiology, as their existing workforce ages, or they
try to expand, or even just keep up with imaging growth, they will be forced to deal with the same
commoditization and workforce trends currently affecting most practices.
We are hopeful that defining the problem, trends, and potential outcomes will jump-start a
collaborative effort to maintain access to Radiology for the population as a whole while providing the
best service levels possible for our partners. Proactive problem solving will allow us to take care of all of
our mutual patients efficiently, effectively, and empathetically.


Summary: Wall Street may now own your physician’s office and the emergency room, just like how it owns many of the shrinking number of insurance companies, drug companies, pharmacies, etc. as well as being your landlord.

Perhaps that is why so many Harvard students go to Wall Street where they can collect part of the money flowing into there, rather than being part of everyone else (including physicians and other health care workers) that Wall Street squeezes money out of.


My H quit medicine in his late forties and is now a much happier high school teacher. He was an outstanding pediatrician taking care of the most needing patients in city clinics, but the combination of insurance nightmares, terrible hours, and overall stress took a deep toll on him. He definitely discouraged our kids from doing medicine.

My only regret is that it took till they were almost grown for us to figure this out. Financially, we’d have probably been better off even with the lower salary because of the years of no salary (med school) or very low (residency) plus huge loans. had the kids before/during med school, so their early childhood was a struggle of no money and often no dad.

This system is really broken.


So do we think it’s finally time to have single-payer healthcare, so that for-profit companies don’t run that industry? And do doctors think it’s finally time??


Oh no, socialism!

Well, exactly my point. The doctors used to hate the idea because they thought all control would be taken away from them. Guess what they’ve got now – no control! It’s only a matter of time before medical care consumers come to the same conclusion. I don’t think we’re there yet; maybe another 10 or 20 years. But it’ll happen eventually.


Single payer healthcare is not a panacea, just ask those in England who rely on NHS.

There is no perfect healthcare system in any country, people have to pick their poison so to speak. It will be interesting to see if the coming generations have the political will to make a significant change when they are in power.

1 Like

H was and remains a proponent of single payer, universal insurance. England’s might not be perfect, but it is not the only model. And I still respect what it has provided for many, many years. A friend who moved here from UK a few years ago is aghast every time she hears about how things work here. She sounds like the NHS was pretty okay in her eyes.

Just think of all the money that goes to private enrichment of insurance companies, the money spent on advertising, the time and money spent in doctor’s offices and hospitals parsing through the many different insurance plans, getting the codes right, re-billing to fight the payment rejections. And the time we all spend getting insurance to pay for the things we pay it to cover.

Read the analyses of how they deliberately turn down procedures and payments just because some people will give up and not fight back.

It’s hard to imagine a system more broken than this one.


Let’s not let the perfect be the enemy of the good.

1 Like

The alternative to corporate medicine is government medicine. I don’t trust the government for at least a hundred reasons not even related to medicine. Do I trust corporate medicine?…NOPE. I try to eat healthy and exercise. If I get sick or injured, I do my research on doctors I can trust.

Currently it’s not corporate medicine. It’s insurance company medicine. They make the decisions.

1 Like

It does seem that the US health care system contains many of the worse aspects of other systems: most expensive, extremely complex in terms of dealing with insurance, many people go without or go into debt or bankruptcy due to medical bills, health outcomes overall are mediocre compared to other rich countries, most patients face limited choices for providers and insurance options and have little knowledge of or access to information to make the best decisions on both medical care options and insurance options, some health care workers are burdened by very high educational debt due to the cost of necessary education, employers disincentivized from keeping jobs in the US due to the cost burden of medical insurance.


All of what you post is true. But somehow- whether it’s MD Anderson, Cleveland Clinic, Mass General, Sloan Kettering, Cedars Sinai, St. Jude, Yale New Haven (I could go on but you all know the geography of the US) these hospitals are filled with billionaires from other countries who hear their diagnosis and get on the first plane to come to the US to meet with whichever specialist will see them first.

So yes- many problems. But you could write a book about why people who could go anywhere for treatment choose the US.


Yes, the elite providers and the plutocrat patients (plus occasionally a hardship story that makes for great charity PR for the providers) can meet in the US.

But how much of that is accessible to ordinary people in the US or elsewhere? The US may be a great place for plutocrats to get medical care, but most people in the US are not plutocrats.

Meanwhile, there is medical tourism from the US by somewhat more ordinary (probably merely upper middle class) patients who may find procedures in other countries to be much less expensive than in the US. There are also occasional cases where a particular condition is much more common and better known in other countries, leading to medical tourism there by those with that condition.


Two places @blossom mentioned (Cleveland Clinic and Mass General) were readily accessible by members of our family, without requiring any connections whatsoever. We also know a family whose child had leukemia and was treated at Children’s Hospital in Boston, one of the nation’s best pediatric hospitals, and she is now 30+ and healthy.

These hospitals are among the best that exist, and they do serve the general public.