The Thomas Sowell quote, “On closer scrutiny, it turns out that many of today's problems are a result of yesterday's solutions” has been ringing in my head a lot lately.
On Epic so many of the items on the patient interface are "self reported" questions. I worked in public health for 40 years at the birth and death level. I KNOW that the questions on both documents are entered incorrectly on the document. This is where the beginning of the problems start, I'm not sure where this fits with topic, but felt the need to say it. I feel strongly when the era of "hospitalists" began was the beginning of our decline.
Thanks so much for sharing your thoughts -- and even more for taking the time to really grapple with the challenges, rewards, and trade-offs embodied in a difficult -- and important! -- profession.
As a lawyer (inactive) and one of the 5% that make up 50% of US medical spend, it's been, uh, interesting to see up-close the vast differences between the two professions. Particularly the very different approaches taken to the profound agency challenges presented in each career path.
1. “If the map doesn’t follow the terrain, follow the terrain.” This was perfectly illustrated in the case of the PE. Sometime the test is wrong. Don’t be afraid to start over again when things aren’t making sense. Listen to the patient, they will usually give you what to need to come to the diagnosis. I can tell many such stories like this.
2. 95 hours a week? That would have been a light week for us and we felt lucky because some of our attendings had been on call every other day for 3 years. But, we had a bunker mentality. The patients came first.
3. I learned more at night than any other time, because patients all seemed to be their worst after 10pm.
4. Thomas Sowell is a brilliant man. I love hearing or reading whatever he has to say.
I was with you all the way until "local produce" in the last paragraph. Why toss in that one word, "local," which has a host of political notions behind it, but nothing medical? Maybe "fresh produce" is all you need to say. But then I'd realize that unprocessed frozen and even canned produce are also better than processed foods for people who can't afford fresh produce, and so maybe we're just left with "produce." :)
On work hours, as a physician who studied medicine in Australia, the Australians thought that our work hours were insane. That’s today, after 80-hour limits. 60 is on the high end for surgery folks, I was told (maybe some people work more). And this probably holds true for a lot of the world; American labor practices are an outlier among first world countries.
The work hour change is one of many that push medicine away from being a vocation (or a lifestyle or an identity) and more towards just being a job. And of the patients who I’ve taken care of at the end of life, not one of them has told me “I wish I’d spent more time at work”.
Personally, I think expanding the medical profession to include people who couldn’t previously handle the crazy work schedules is a benefit. I see no reason why medicine cannot be practiced on a full time or even part time basis for some people. I am certainly not a 24/7 on call physician and have no desire to ever be that. I specifically chose a training path that let me spend a relatively small amount of time actually practicing medicine, and I put the rest of my time to good use. But I can see the tradeoffs, both in terms of physician quality and of the increasingly hostile relationship we have with our jobs now that we see them as jobs.
I know I’m harping on this, but ozempic may be weight loss drug but it is not a weight maintenance drug. Addictive foods are not the problem. Addiction is the problem. A bad metaphor would be to saying being LGBTQ is a choice. Being overweight or obese is not a choice. It’s an addiction for most of us. The solution is not a pill or shot or telling me to diet and exercise. The solution is treating my addiction. I never ask a smoker if they know smoking is bad. That’s a slap in the face. They know it and telling someone to smoke less or eat less just makes me want to do more. Anyway loved the writing and tytytytyty!
Adam, I will have to respectfully disagree with both of your points, although each, of course, has an element of obvious truth.
With regard to training, I had a similar schedule to yours: 24 on/12 off/12 on/12 off -- rinse and repeat. Ours was the very first cohort that actually got paid to be interns ($6,000/year!!) so it was expected that we would work hard -- but we were working no harder than the many cohorts before us that were unpaid (but often got room/board).
But we learned. One does not learn about DKA walking out halfway through the struggle...now house staff look at their watches and just leave. Eventually they will have a patient that would have needed them to understand the whole process -- and they will not.
Similarly, the replacement of patient-centric physicians with shift-working hospitalists has likely done more to deprecate patient care than almost any other thing (other than EMRs) of my acquaintance. Some things just do not lend themselves to shift work -- good patient care at the physician level (which requires an integrative understanding of each individual's health and disease) is one of them. I watch hospitalists change shifts and each new shift worker decides to take a completely different tack with the same patient -- because they represent eight hours on a shift -- not that patient's physician. It makes a major, uncorrectable difference.
As far as EMRs, I was one of the pioneers in that space (still working on cognitive AI approaches) and note that they have, by and large, been an abject failure -- albeit an impossibly expensive one. As O’Reilly et al (as highlighted by Soumerai/Koppel in WSJ) pointed out in their analysis of 36,000 studies: “No evidence of cost reduction or quality enhancement with EHRs." The current approach is built on “unsubstantiated promises.”
EMRs (particularly Epic) were largely designed as adjuncts to billing systems to support the requirement for "claims attachments" when plans refused to pay without "clinical data". They are largely visit centric (not patient centric) because of this orientation. There is no understanding of any data -- just compilations of often conflicting/duplicate/unnecessary stuff completely obscuring what is important. Mountains of pdfs of visits from other times/institutions are largely unread. A solid case can be made for the advantages of the paper chart when one analyzes quality vs quantity - but irrespective, the EMR's aggregation of massive quantities of data that no one wants and no one needs to satisfy the lawyers in exchange for burying what is important is a giant deal. And, of course, unless one is at Kaiser South or North (but not both, because they do not share a record, either) every patient has multiple EMRs out there with multiple PHRs -- all ensuring that a complete picture of what is going on with a patient cannot be derived.
Incidentally, the deep learning/LLM class of probabilistic AI tools are foundationally unable to solve these issues as has been repeatedly demonstrated since Shortliffe/Feigenbaum/Weed began that journey decades ago. An answer that is wrong 10% of the time is unacceptable in health care -- and even with more iterations and more training, that is about the theoretical maximum. Around 77% is the best such engines can do at the moment (setting aside limited computer vision cases). AI is basically just another term to throw around to suck money from people for the next 10 years of their EMRs not delivering substantial patient care value.
So yes, you are correct. We have massively failed on both axes. The question is who of significance cares? (I am discounting all of us peons.) And what is anyone going to do about it?
A similar “excess safety” phenomenon occurs specifically in surgery. A surgeon friend graduated from Fellow to Attending a few years ago, and went to a new institution. He discovered that as “the one in charge” he had to make a lot of decisions that previously had been made for him. So he had no experience with those issues! How is that possible? When something novel/risky came up during his training his own instructors told him what to do! Now he is the one in charge, for the first time.
What has changed in surgery is that the Attending is never absent. Usually in the OR, but if not then accessible within 2 minutes. In the “old days” they were often home in bed, apparently, so the terrified residents had to cope. When coping they had to confront the obscure and bizarre on their own. (And got feedback the next morning about whether they had done the right thing.)
Of course my friend has coping strategies, but it was a shock.
I appreciate your comments about residency training (I trained in the early 80s). And I especially appreciate your succinct synopsis of the diet/exercise disaster that we have created. I remain bewildered at how we (the general public) are so unconscious of this. In my years of practice, most of the times I broached the subject of nutrition and exercise my patients would suddenly remember that they had somewhere else they needed to be, and thus would end our visit.
As I near retirement I will have spent close to 2 years working in New Zealand’s publicly funded health care system, primarily in their breast screening service. An amazing opportunity to see the successes of such a system, as well as the problems. I certainly have a more nuanced view of our health care system (and our economics in general). Although, I am retiring to play with my grandchildren who, sadly, don’t really care about the nuances of differently funded health care systems!
So thank you for another wonderful post. I’m trying to get all of my colleagues, American and otherwise, to follow you.
The Thomas Sowell quote is so applilcable to medicine! Really appreciate reading your thoughtful take on a problem that requires stepping back and looking at the big picture.
"There are people in every time and every land who want to stop history in its tracks. They fear the future, mistrust the present, and invoke the security of a comfortable past which, in fact, never existed." RFK. Senator and AG
Halstead's cocaine addiction and Osler's racism and ageism deserved a footnote.
Is this really EMR problem? Failure to note in report inadequacy of test to detect is error. Technician and procedure that resulted in too long between dye and scan is error. Hospital put your carrier on notice!
(The salubrious effects of no-fault workers compensation to make workplaces safer too long and off topic to flesh out here.)
Ozempic as "cure" to symptom rather than root cause is well taken. Food is culture. If you don't know how and what your grandparents and great grandparents ate, you are alienated. Alienation has negative medical, political and social consequences.
Medicine as fraternity/sorority complete with hazing (or even a guild or "familial" surrogate as envision by Hippocrates) cannot cure alienation either. It could be a granfalloon masquerading as a karass. See Vonnegut.
The philosophy and history of medicine cannot be separated from a coherent and complex anthropology.
On Epic so many of the items on the patient interface are "self reported" questions. I worked in public health for 40 years at the birth and death level. I KNOW that the questions on both documents are entered incorrectly on the document. This is where the beginning of the problems start, I'm not sure where this fits with topic, but felt the need to say it. I feel strongly when the era of "hospitalists" began was the beginning of our decline.
Thanks so much for sharing your thoughts -- and even more for taking the time to really grapple with the challenges, rewards, and trade-offs embodied in a difficult -- and important! -- profession.
As a lawyer (inactive) and one of the 5% that make up 50% of US medical spend, it's been, uh, interesting to see up-close the vast differences between the two professions. Particularly the very different approaches taken to the profound agency challenges presented in each career path.
Adding to point one… I had a sense of meaning during training that may be lost to residents today. Meaning matters
Excellent article but my favorite part was fn 3. LOL.
A couple of comments:
1. “If the map doesn’t follow the terrain, follow the terrain.” This was perfectly illustrated in the case of the PE. Sometime the test is wrong. Don’t be afraid to start over again when things aren’t making sense. Listen to the patient, they will usually give you what to need to come to the diagnosis. I can tell many such stories like this.
2. 95 hours a week? That would have been a light week for us and we felt lucky because some of our attendings had been on call every other day for 3 years. But, we had a bunker mentality. The patients came first.
3. I learned more at night than any other time, because patients all seemed to be their worst after 10pm.
4. Thomas Sowell is a brilliant man. I love hearing or reading whatever he has to say.
I was with you all the way until "local produce" in the last paragraph. Why toss in that one word, "local," which has a host of political notions behind it, but nothing medical? Maybe "fresh produce" is all you need to say. But then I'd realize that unprocessed frozen and even canned produce are also better than processed foods for people who can't afford fresh produce, and so maybe we're just left with "produce." :)
On work hours, as a physician who studied medicine in Australia, the Australians thought that our work hours were insane. That’s today, after 80-hour limits. 60 is on the high end for surgery folks, I was told (maybe some people work more). And this probably holds true for a lot of the world; American labor practices are an outlier among first world countries.
The work hour change is one of many that push medicine away from being a vocation (or a lifestyle or an identity) and more towards just being a job. And of the patients who I’ve taken care of at the end of life, not one of them has told me “I wish I’d spent more time at work”.
Personally, I think expanding the medical profession to include people who couldn’t previously handle the crazy work schedules is a benefit. I see no reason why medicine cannot be practiced on a full time or even part time basis for some people. I am certainly not a 24/7 on call physician and have no desire to ever be that. I specifically chose a training path that let me spend a relatively small amount of time actually practicing medicine, and I put the rest of my time to good use. But I can see the tradeoffs, both in terms of physician quality and of the increasingly hostile relationship we have with our jobs now that we see them as jobs.
I ALWAYS enjoy your reflection posts. Provoking and thoughtful in a concise post. So many things to unpack today.
Thank you.
I know I’m harping on this, but ozempic may be weight loss drug but it is not a weight maintenance drug. Addictive foods are not the problem. Addiction is the problem. A bad metaphor would be to saying being LGBTQ is a choice. Being overweight or obese is not a choice. It’s an addiction for most of us. The solution is not a pill or shot or telling me to diet and exercise. The solution is treating my addiction. I never ask a smoker if they know smoking is bad. That’s a slap in the face. They know it and telling someone to smoke less or eat less just makes me want to do more. Anyway loved the writing and tytytytyty!
Regarding EMRs, it’d be nice if clinical trial sites were actually interested in accessing them in order to reduce the patient paperwork burden: https://bessstillman.substack.com/p/please-be-dying-but-not-too-quickly , but they do not seem to be.
Adam, I will have to respectfully disagree with both of your points, although each, of course, has an element of obvious truth.
With regard to training, I had a similar schedule to yours: 24 on/12 off/12 on/12 off -- rinse and repeat. Ours was the very first cohort that actually got paid to be interns ($6,000/year!!) so it was expected that we would work hard -- but we were working no harder than the many cohorts before us that were unpaid (but often got room/board).
But we learned. One does not learn about DKA walking out halfway through the struggle...now house staff look at their watches and just leave. Eventually they will have a patient that would have needed them to understand the whole process -- and they will not.
Similarly, the replacement of patient-centric physicians with shift-working hospitalists has likely done more to deprecate patient care than almost any other thing (other than EMRs) of my acquaintance. Some things just do not lend themselves to shift work -- good patient care at the physician level (which requires an integrative understanding of each individual's health and disease) is one of them. I watch hospitalists change shifts and each new shift worker decides to take a completely different tack with the same patient -- because they represent eight hours on a shift -- not that patient's physician. It makes a major, uncorrectable difference.
As far as EMRs, I was one of the pioneers in that space (still working on cognitive AI approaches) and note that they have, by and large, been an abject failure -- albeit an impossibly expensive one. As O’Reilly et al (as highlighted by Soumerai/Koppel in WSJ) pointed out in their analysis of 36,000 studies: “No evidence of cost reduction or quality enhancement with EHRs." The current approach is built on “unsubstantiated promises.”
EMRs (particularly Epic) were largely designed as adjuncts to billing systems to support the requirement for "claims attachments" when plans refused to pay without "clinical data". They are largely visit centric (not patient centric) because of this orientation. There is no understanding of any data -- just compilations of often conflicting/duplicate/unnecessary stuff completely obscuring what is important. Mountains of pdfs of visits from other times/institutions are largely unread. A solid case can be made for the advantages of the paper chart when one analyzes quality vs quantity - but irrespective, the EMR's aggregation of massive quantities of data that no one wants and no one needs to satisfy the lawyers in exchange for burying what is important is a giant deal. And, of course, unless one is at Kaiser South or North (but not both, because they do not share a record, either) every patient has multiple EMRs out there with multiple PHRs -- all ensuring that a complete picture of what is going on with a patient cannot be derived.
Incidentally, the deep learning/LLM class of probabilistic AI tools are foundationally unable to solve these issues as has been repeatedly demonstrated since Shortliffe/Feigenbaum/Weed began that journey decades ago. An answer that is wrong 10% of the time is unacceptable in health care -- and even with more iterations and more training, that is about the theoretical maximum. Around 77% is the best such engines can do at the moment (setting aside limited computer vision cases). AI is basically just another term to throw around to suck money from people for the next 10 years of their EMRs not delivering substantial patient care value.
So yes, you are correct. We have massively failed on both axes. The question is who of significance cares? (I am discounting all of us peons.) And what is anyone going to do about it?
Always appreciate your views on things.
A similar “excess safety” phenomenon occurs specifically in surgery. A surgeon friend graduated from Fellow to Attending a few years ago, and went to a new institution. He discovered that as “the one in charge” he had to make a lot of decisions that previously had been made for him. So he had no experience with those issues! How is that possible? When something novel/risky came up during his training his own instructors told him what to do! Now he is the one in charge, for the first time.
What has changed in surgery is that the Attending is never absent. Usually in the OR, but if not then accessible within 2 minutes. In the “old days” they were often home in bed, apparently, so the terrified residents had to cope. When coping they had to confront the obscure and bizarre on their own. (And got feedback the next morning about whether they had done the right thing.)
Of course my friend has coping strategies, but it was a shock.
I appreciate your comments about residency training (I trained in the early 80s). And I especially appreciate your succinct synopsis of the diet/exercise disaster that we have created. I remain bewildered at how we (the general public) are so unconscious of this. In my years of practice, most of the times I broached the subject of nutrition and exercise my patients would suddenly remember that they had somewhere else they needed to be, and thus would end our visit.
As I near retirement I will have spent close to 2 years working in New Zealand’s publicly funded health care system, primarily in their breast screening service. An amazing opportunity to see the successes of such a system, as well as the problems. I certainly have a more nuanced view of our health care system (and our economics in general). Although, I am retiring to play with my grandchildren who, sadly, don’t really care about the nuances of differently funded health care systems!
So thank you for another wonderful post. I’m trying to get all of my colleagues, American and otherwise, to follow you.
The Thomas Sowell quote is so applilcable to medicine! Really appreciate reading your thoughtful take on a problem that requires stepping back and looking at the big picture.
Thought essay as usual. Thank you.
"There are people in every time and every land who want to stop history in its tracks. They fear the future, mistrust the present, and invoke the security of a comfortable past which, in fact, never existed." RFK. Senator and AG
Halstead's cocaine addiction and Osler's racism and ageism deserved a footnote.
Is this really EMR problem? Failure to note in report inadequacy of test to detect is error. Technician and procedure that resulted in too long between dye and scan is error. Hospital put your carrier on notice!
(The salubrious effects of no-fault workers compensation to make workplaces safer too long and off topic to flesh out here.)
Ozempic as "cure" to symptom rather than root cause is well taken. Food is culture. If you don't know how and what your grandparents and great grandparents ate, you are alienated. Alienation has negative medical, political and social consequences.
Medicine as fraternity/sorority complete with hazing (or even a guild or "familial" surrogate as envision by Hippocrates) cannot cure alienation either. It could be a granfalloon masquerading as a karass. See Vonnegut.
The philosophy and history of medicine cannot be separated from a coherent and complex anthropology.