As a 3rd year FM resident who recently started moonlighting in a rural ED, this NYT article got MY pulse up to 120 when I read it (considering the presentation and how many times already I’ve dismissed “sepsis alerts” in patients who aren’t septic, and yet still don’t have a “more likely diagnosis” (one of our alert dismissal reasons)).
Anyway, I also had a Sick patient on one of my first shifts that I DID admit for observation and unfortunately STILL ended up dying <48h later - even though I (appropriately) transferred him to a higher level of care within hours of his admission.
The uncertainty of it all is terrifying, and I have no idea how full-time ED docs in tertiary care centers with their constant flow of both Sick and worried well patients manage. Maybe I’m just not cut out for the emergency room… it’s just a shame that rural communities are so desperate for help/relief they even hire ME! 👀
And thank you for adding “uncertainty management” to my list of pithy phrases that characterize my job as clinician. “Alarm fatigue” was already in there!
With hindsight lenses it's much easier to say what should and shouldn't have happened during that ER visit. The information we gather after the fact is very different than what the medical team finds in that moment, in that setting, in those exact circumstances. No doubt that these post-mortem case dissections can teach us so much but if they blame the system, much less the doctor, it's already a blame game - a right and a wrong. A should-have happened but didn't-happen.
Errors in medicine often equal deaths and errors are inevitable. You and me are going to lose loved ones to medical errors because clinical medicine permeates our day to day lives. But we can improve by 1% each time by recognizing one hindsight error without judgement, without blame. This will make for a better health system overall. But it can't happen when we sue or punish systems or the medical team. Thank you for a great write-up.
That is a terrific conclusion and the “eyeball” test is so important. But I would add that it is the combination of two things: sick + uncertainty. That’s a particularly bad combination and I often find our residents are good at determining sick/not sick. They are not good at uncertainty, in part because they “think” they should know more and are fearful of saying they don’t know. But no matter the situation or where you are in your training, you need to listen to the part of the brain that is nagging at you when the patient’s picture doesn’t fit right.
A wonderful essay. In Peds, when I trained, a bounce back within 24 hrs used to be an automatic admission, because something was wrong — we just didn’t know what it was, yet. Since I haven’t worked in the ER for a long time, I don’t know if that rule still holds.
It’s tempting to try to craft a new rule so that a bad outcome doesn’t happen again. But rules don’t replace clinical judgement. And it is the need for clinical judgement in an environment that recognizes its value and supports it, that is what this essay so eloquently argues for. In the ER, office, or inpatient unit, sick vs not sick is the first and most important decision we have to make.
Recently I heard a similar story from a friend who took in their sick father. The ER staff wanted to send him home, but the family refused after recognizing the error. They insisted he be admitted and stayed overnight until the admission took place. The man was eventually admitted and almost immediately moved to intensive care, where he remained for the next 11 days. (He lived, but would not have had they taken him home.)
What — or who — is creating this protocol that allows very sick people to be sent home? Insurance companies? Medicare? The hospital itself? Can you help us understand?
"We have to be willing to say, “I don’t know what this is. But I know they don’t belong at home.”
And then the "Care Manager" calls her boss who calls your boss, and eventually the CFO gets involved to explain the meaning of "margin" (profit) to you.
Cory, knowing my patient is Sick and being unable to get help is The Worst.
This is the first I've heard of this particular case, but I think every primary care physician knows the case in their bones. There is something wrong with a patient, nothing to point at, and the ED cannot justify hospitalizing them, the hospitalist is too busy to even come down and look, and the primary care doctor can do nothing much beyond wringing their hands harder.
Awful case, just awful. But the Sick versus not sick distinction is the bread and butter of emergency medicine. It may sound like a revelation to others when eloquently described, but for EPs it is the first and most crucial question on every patient. And I guarantee it was for poor Sam as well. But sometimes Sick does not look Sick, and the presumption that Sam did is strongly challenged by the choices made by competent, smart physicians at a busy academic center. Before condemning them, perhaps consider that you might have done exactly the same, even in the face of a bounce back with ever so slightly tilted lab results (and many normal ones, like lactate). Let us not pretend that we can tell Sick from not sick through the retrospectoscope.
This is why AI is dangerous if overused in place of human judgment. The most important thing a doctor does is actually look at the patient and say "hes sick". An AI tool can't do that. That's experience, wisdom, the human brain and the art of medicine. Labs and data tell you one thing- looking and talking to the patient tells you so much more.
It was a heartbreaking story, remarkably well told by the lay press. IN reading the case my concern for hospitalization was triggered by a healthy young male with unexplained dyspnea and unexplained tachycardia (pulse was 126) who felt “sick”, 2 ER visits in 24 hours. More than any one diagnostic test, it was the clinical presentation that should have warranted admission. Would that the “system” made this easier.
“Unfortunately, when algorithms and payers become gatekeepers, they sometimes replace judgment rather than inform it. This is not a failure of individual clinicians — it is a failure of culture. One we all own—and one we must collectively repair.”
My second comment is EVERY Hospital CEO in the country and Hospital Lawyer is responsible for this above narrative. Our fault as physicians is that we drink the kool-aid. It has ruined Emergency Medicine in this country.
In medical education, the very first test of whether a resident is ready for independent practice hinges on their ability to triage “Sick vs Not Sick.”
This statement is HUGE. Upon seeing the patient without one single lab or imaging study, the physician should know the answer to this question based on history-exam-clinical acumen. I don’t know anything about the case you referenced, but I do know this information to be true of every Emergency Medicine Physician practicing in this country.
“It is not the job of the patient to prove they are sick. It is my job to prove they are WELL.” This is my lesson to the residents I work with.
I had the same thought. When I was on my ICU rotation, my resident told me that one of the most important things I could learn is “knowing when a patient is sick”. This resident didn’t mean “sick” as in “not feeling well”, but rather “sick” as in “he’s about to tank, so we need to do something right now”.
As a 3rd year FM resident who recently started moonlighting in a rural ED, this NYT article got MY pulse up to 120 when I read it (considering the presentation and how many times already I’ve dismissed “sepsis alerts” in patients who aren’t septic, and yet still don’t have a “more likely diagnosis” (one of our alert dismissal reasons)).
Anyway, I also had a Sick patient on one of my first shifts that I DID admit for observation and unfortunately STILL ended up dying <48h later - even though I (appropriately) transferred him to a higher level of care within hours of his admission.
The uncertainty of it all is terrifying, and I have no idea how full-time ED docs in tertiary care centers with their constant flow of both Sick and worried well patients manage. Maybe I’m just not cut out for the emergency room… it’s just a shame that rural communities are so desperate for help/relief they even hire ME! 👀
Great essay.
And thank you for adding “uncertainty management” to my list of pithy phrases that characterize my job as clinician. “Alarm fatigue” was already in there!
With hindsight lenses it's much easier to say what should and shouldn't have happened during that ER visit. The information we gather after the fact is very different than what the medical team finds in that moment, in that setting, in those exact circumstances. No doubt that these post-mortem case dissections can teach us so much but if they blame the system, much less the doctor, it's already a blame game - a right and a wrong. A should-have happened but didn't-happen.
Errors in medicine often equal deaths and errors are inevitable. You and me are going to lose loved ones to medical errors because clinical medicine permeates our day to day lives. But we can improve by 1% each time by recognizing one hindsight error without judgement, without blame. This will make for a better health system overall. But it can't happen when we sue or punish systems or the medical team. Thank you for a great write-up.
That is a terrific conclusion and the “eyeball” test is so important. But I would add that it is the combination of two things: sick + uncertainty. That’s a particularly bad combination and I often find our residents are good at determining sick/not sick. They are not good at uncertainty, in part because they “think” they should know more and are fearful of saying they don’t know. But no matter the situation or where you are in your training, you need to listen to the part of the brain that is nagging at you when the patient’s picture doesn’t fit right.
A wonderful essay. In Peds, when I trained, a bounce back within 24 hrs used to be an automatic admission, because something was wrong — we just didn’t know what it was, yet. Since I haven’t worked in the ER for a long time, I don’t know if that rule still holds.
It’s tempting to try to craft a new rule so that a bad outcome doesn’t happen again. But rules don’t replace clinical judgement. And it is the need for clinical judgement in an environment that recognizes its value and supports it, that is what this essay so eloquently argues for. In the ER, office, or inpatient unit, sick vs not sick is the first and most important decision we have to make.
Recently I heard a similar story from a friend who took in their sick father. The ER staff wanted to send him home, but the family refused after recognizing the error. They insisted he be admitted and stayed overnight until the admission took place. The man was eventually admitted and almost immediately moved to intensive care, where he remained for the next 11 days. (He lived, but would not have had they taken him home.)
What — or who — is creating this protocol that allows very sick people to be sent home? Insurance companies? Medicare? The hospital itself? Can you help us understand?
"We have to be willing to say, “I don’t know what this is. But I know they don’t belong at home.”
And then the "Care Manager" calls her boss who calls your boss, and eventually the CFO gets involved to explain the meaning of "margin" (profit) to you.
Cory, knowing my patient is Sick and being unable to get help is The Worst.
This is the first I've heard of this particular case, but I think every primary care physician knows the case in their bones. There is something wrong with a patient, nothing to point at, and the ED cannot justify hospitalizing them, the hospitalist is too busy to even come down and look, and the primary care doctor can do nothing much beyond wringing their hands harder.
Awful case, just awful. But the Sick versus not sick distinction is the bread and butter of emergency medicine. It may sound like a revelation to others when eloquently described, but for EPs it is the first and most crucial question on every patient. And I guarantee it was for poor Sam as well. But sometimes Sick does not look Sick, and the presumption that Sam did is strongly challenged by the choices made by competent, smart physicians at a busy academic center. Before condemning them, perhaps consider that you might have done exactly the same, even in the face of a bounce back with ever so slightly tilted lab results (and many normal ones, like lactate). Let us not pretend that we can tell Sick from not sick through the retrospectoscope.
Amen!
Complicating everything: Sick in a young man with tremendous physiological reserve can look very different from Sick in a 45 year old.
I have tremendous empathy for everyone in this case.
This is why AI is dangerous if overused in place of human judgment. The most important thing a doctor does is actually look at the patient and say "hes sick". An AI tool can't do that. That's experience, wisdom, the human brain and the art of medicine. Labs and data tell you one thing- looking and talking to the patient tells you so much more.
What would have been the ideal approach, that would have saved this young man?
Admission to the hospital for observation.
It was a heartbreaking story, remarkably well told by the lay press. IN reading the case my concern for hospitalization was triggered by a healthy young male with unexplained dyspnea and unexplained tachycardia (pulse was 126) who felt “sick”, 2 ER visits in 24 hours. More than any one diagnostic test, it was the clinical presentation that should have warranted admission. Would that the “system” made this easier.
Superb essay; brilliantly written. Yes, it’s a culture problem and it’s a medical education crisis.
Excellent post, a must-read reminder for anyone taking care of patients.
“Unfortunately, when algorithms and payers become gatekeepers, they sometimes replace judgment rather than inform it. This is not a failure of individual clinicians — it is a failure of culture. One we all own—and one we must collectively repair.”
My second comment is EVERY Hospital CEO in the country and Hospital Lawyer is responsible for this above narrative. Our fault as physicians is that we drink the kool-aid. It has ruined Emergency Medicine in this country.
Cory great article.
In medical education, the very first test of whether a resident is ready for independent practice hinges on their ability to triage “Sick vs Not Sick.”
This statement is HUGE. Upon seeing the patient without one single lab or imaging study, the physician should know the answer to this question based on history-exam-clinical acumen. I don’t know anything about the case you referenced, but I do know this information to be true of every Emergency Medicine Physician practicing in this country.
“It is not the job of the patient to prove they are sick. It is my job to prove they are WELL.” This is my lesson to the residents I work with.
I had the same thought. When I was on my ICU rotation, my resident told me that one of the most important things I could learn is “knowing when a patient is sick”. This resident didn’t mean “sick” as in “not feeling well”, but rather “sick” as in “he’s about to tank, so we need to do something right now”.