Febrile neutropenia (non -chemotherapy) - what would you do?

[leaf345]

Hi guys, I just have a question and I was hoping to get an idea of how others would manage something like this.

I'm a non-EM intern, doing my EM block. I had a discussion with my EM attending (who seems pretty smart) about what to do if an otherwise young healthy patient presented with severe neutropenia and a high fever (>39C). Other than the CBC, everything is fine. Feels well. No concerns, like a high lactate on bloodwork. Obviously since its EM, we don't know if the patient has had chronic neutropenia or what her bone marrow looks like. But no marrow transplant or chemo.

My inkling was to start broad-spectrum Abx and consult IM, but my EM attending disagreed. Another EM attending, who is also young and pretty smart, also disagreed. Close follow up would be fine, if there is no Hx of BM suppression and there was no focus to an infection. I don't mean to second-guess them, but I've been doing some reading since then that seems to suggest that the patient should at least be started on oral broad-spectrum Abx. So now I'm confused, which leads me to why I'm here.

What would you guys do?

---

Members don't see this ad.

 

[deleted547339]

Hi guys, I just have a question and I was hoping to get an idea of how others would manage something like this.

I'm a non-EM intern, doing my EM block. I had a discussion with my EM attending (who seems pretty smart) about what to do if an otherwise young healthy patient presented with severe neutropenia and a high fever (>39C). Other than the CBC, everything is fine. Feels well. No concerns, like a high lactate on bloodwork. Obviously since its EM, we don't know if the patient has had chronic neutropenia or what her bone marrow looks like. But no marrow transplant or chemo.

My inkling was to start broad-spectrum Abx and consult IM, but my EM attending disagreed. Another EM attending, who is also young and pretty smart, also disagreed. Close follow up would be fine, if there is no Hx of BM suppression and there was no focus to an infection. I don't mean to second-guess them, but I've been doing some reading since then that seems to suggest that the patient should at least be started on oral broad-spectrum Abx. So now I'm confused, which leads me to why I'm here.

What would you guys do?

I had something similar recently. I gave abx and admitted. Hospital course was uneventful. WBCs rebounded. No source found. DC'ed home. Never got a BM bx.

 

[Zebra Hunter]

What was the ANC?

 

[leaf345]

Lets say <0.5 but >0.1

 

[Interpolfanclub]

In the community this is an admit. Easy to play fast and loose when you're protected by academics. You send that guy home in my community hospital and he goes toes up and you're in some unpleasant meetings with guys in suits.

Admit, abx, wbc rebound in a few days and bcx neg and they go home. no harm done.

 

[Daiphon]

In the community this is an admit. Easy to play fast and loose when you're protected by academics. You send that guy home in my community hospital and he goes toes up and you're in some unpleasant meetings with guys in suits.

Admit, abx, wbc rebound in a few days and bcx neg and they go home. no harm done.

Even in academia... this is a slam dunk admission for me, as well.

In addition to ABx, strongly consider an HIV panel. Untreated & unknown HIV can cause significant neutropenia; in a young, otherwise previously healthy patient, this is my #1 concern after some sort of marrow disorder.

-d

 

[WilcoWorld]

At my academic center fever + neutropenia = admit. All day, every day. The presence of chemo only determines whether the patient goes to the Heme/Onc service.

If the other cell lines were down I'd ask about cocaine. Saw a few of these 2/2 levamisole adulterants a few years back.

 

[VentJockey]

I'm Internal Medicine PGY2. To me, this is a slam dunk admit to a step down unit. Febrile neutropenia is bad juju. Elevated lactate is also bad juju. The combination makes me super, super nervous. This is someone with the potential to die in a rather rapid and dramatic way. No internist worth their salt would fight you on an admission like this.

Based on your description, in fact, the patient meets criteria for severe sepsis (2/4 SIRS criteria, plus lactic acidosis). Broad spectrum antibiotics are a must. Close observation is a must. Obviously depending upon the situation, I'd be quite eager to get a bone marrow biopsy early in this patient.

But, maybe there is more to this story? I'm having a hard time believing that two attending EM docs were willing to send this case home.

 

[VentJockey]

Just want to point one more thing out. Even if your gut tells you the patient is fine to go home, here are two ways this situation can play out:

1) You admit, the patient does great, and gets discharged the next day or two. Someone maybe complains about a soft admit (though in this situation, I doubt it).

2) You send home, the patient cumps and dies on the kitchen floor. You get a lawsuit and a report to the data bank.

I'd rather deal with someone complaining about a soft admit.

 

[engineeredout]

IM/EM resident here so I play both fields. Either way, my usual practice for febrile neutropenia = broad spec abx, pan culture, admit, with the caviot being that I don't trust the follow up for most of the patients I see in my EDs

That being said, there are definitely treatment algorithms which provide for certain lower risk patients to get an initial dose of IV antibiotics, discharged with PO antibiotics, and close follow up. Basically, relatively young, good follow up, well appearing, normal VS aside from temp can get fluoroquinolones and stay on them until their ANC recovers. These patients would get all the cultures drawn in the ED prior to DC. Most importantly, these patients need to have supurb follow up. Like an entire family of supporters and a nearby oncologist who is willing to see them the following morning and follow up with them constantly. In the absence of that, there is too much to lose to send these patients home.

Children's Hospital of Philadelphia has all their treatment algorithms available online for public viewing. This is for kiddos but the same principles apply:

http://www.chop.edu/clinical-pathway/oncology-patient-fever-clinical-pathway

 

[TooMuchResearch]

I'm Internal Medicine PGY2. To me, this is a slam dunk admit to a step down unit. Febrile neutropenia is bad juju. Elevated lactate is also bad juju. The combination makes me super, super nervous. This is someone with the potential to die in a rather rapid and dramatic way. No internist worth their salt would fight you on an admission like this.

Based on your description, in fact, the patient meets criteria for severe sepsis (2/4 SIRS criteria, plus lactic acidosis). Broad spectrum antibiotics are a must. Close observation is a must. Obviously depending upon the situation, I'd be quite eager to get a bone marrow biopsy early in this patient.

But, maybe there is more to this story? I'm having a hard time believing that two attending EM docs were willing to send this case home.

I read non-elevated lactate.

 

[EctopicFetus]

Lactate high its crazy to send someone home. If the lactate is normal and they are neutropenic thats still an admit. Anyone who dc's this type of patient is ballsy.

 

[WilcoWorld]

IM/EM resident here so I play both fields. Either way, my usual practice for febrile neutropenia = broad spec abx, pan culture, admit, with the caviot being that I don't trust the follow up for most of the patients I see in my EDs

That being said, there are definitely treatment algorithms which provide for certain lower risk patients to get an initial dose of IV antibiotics, discharged with PO antibiotics, and close follow up. Basically, relatively young, good follow up, well appearing, normal VS aside from temp can get fluoroquinolones and stay on them until their ANC recovers. These patients would get all the cultures drawn in the ED prior to DC. Most importantly, these patients need to have supurb follow up. Like an entire family of supporters and a nearby oncologist who is willing to see them the following morning and follow up with them constantly. In the absence of that, there is too much to lose to send these patients home.

Children's Hospital of Philadelphia has all their treatment algorithms available online for public viewing. This is for kiddos but the same principles apply:

http://www.chop.edu/clinical-pathway/oncology-patient-fever-clinical-pathway

This PLUS the patient wants to go home, and I'm on board.

 

[VentJockey]

Oh, my bad. I misread the original post and thought the patient had *elevated* lactate. Sorry about that. Then the patient could indeed go home under the right circumstances.

 

[Apollyon]

Anyone who dc's this type of patient is ballsy.

As I learned in military school, "there's a fine line between balls and stupidity".

 

[Xerxes1729]

I would feel more comfortable with the patient going home if he was on chemo for some malignancy, since there are starting to be guidelines for managing some of these people as outpatients. We'd know why he was neutropenic, I'd be able to talk to his oncologist who would know if this was expected or not, and I know that the oncologist would see him promptly.

With this patient, I would definitely admit. You don't know why he's neutropenic. You don't know what the course of this will be. And if he's young and otherwise healthy, I'm going to bet he doesn't have a PCP or doesn't know who his PCP is. Even if he has one, how is that call going to go? "Hi Dr. Internist. Your patient has a high fever and neutropenia. He looks okay. I'm going to discharge him so you can work this up as an outpatient." One of the PCPs in my area sent in a 25yo asymptomatic guy at 2 AM because a routine CBC came back with a WBC of 12. No way they'd go for this.

 

[Birdstrike]

At my academic center fever + neutropenia = admit.

Agree. Of all the BS that gets admitted frivolously, and you're going to send a patient this sick home? Uh...no. What kind of garbage are people being taught nowadays, at academic centers? Is there something lost in translation here?

Or do they want the gullible amongst you to "save money for the common good for the 'mother system'" at the expense of the patient, and at your own medical-legal risk? Think for yourselves, people, and have some common sense in the face of the utopian dreams of academicians. Asymptomatic neutropenia, with absolutely no fever would be different. Fever/neutropenia? You worked all week and you finally got someone sick. Admit them for God's sake.

 

[RustedFox]

...Admit them, for God's sake.

Just recently, I sent a 30 year old to the cath lab for a STEMI.

A 30 year old.

Every time some IM-mer gives me flak about a soft admit because they're "young and healthy and can follow-up", I refer to that case.

 

[VentJockey]

As I mentioned previously in this thread, I'm a PGY2 IM. I used to be interested in blocking admissions. Then, I had a few close calls, and some friends of mine had some adverse events. The fall out from all of that made me realize something: it doesn't matter if I think admission is indicated or not. Two reasons:

1) An ER doc think they should be admitted. Even if I'm right, and they meet appropriate discharge criteria, the ER doc is still going to write in the chart that they requested admission. If something goes bad, who do you think gets the blame? Me! If I'm the one who blocked the admission, I'm the one who goes right under the bus when things go bad. I'm not interested in going under the bus, thanks very much. So, on this basis alone, any request for admission from the ED gets my immediate consent (there are exceptions, yes, but they are few and far between). I would rather have my attending scold me for admitting something weak than to be on the podium for an M&M.

2) I'm a second year resident. I must admit that I think I'm awesome. And, I am indeed pretty awesome. Also, I'm pretty inexperienced. If an attending wants something admitted, there is most likely a reason. I agree that some are just douches who are dumping liability, but the majority are sincerely concerned about how the patient will do. The right thing to do is defer to the more experienced doc, who knows more than me. There is the potential of iatrogenesis, yes, but an overnight stay in the hospital for an otherwise healthy person is not a recipe for iatrogenesis. If the ER doc is wrong, the patient spent a night in the hospital for no reason. If I'm wrong, the patient may die. So, a night in the hospital it is.

 

[Hair Police]

As I mentioned previously in this thread, I'm a PGY2 IM. I used to be interested in blocking admissions. Then, I had a few close calls, and some friends of mine had some adverse events. The fall out from all of that made me realize something: it doesn't matter if I think admission is indicated or not. Two reasons...

This is what a maturing resident looks like. These are good experiences and thoughts to have early in residency.

One suggestion - you don't have to be a seive. The hospitalists will sometimes have more info on a patient or know of outpatient services I don't. If you know of a mechanism to get someone appropriate follow up instead of staying a night, I'll listen.

 

[Daiphon]

This is what a maturing resident looks like. These are good experiences and thoughts to have early in residency.

One suggestion - you don't have to be a seive. The hospitalists will sometimes have more info on a patient or know of outpatient services I don't. If you know of a mechanism to get someone appropriate follow up instead of staying a night, I'll listen.

Ditto.

I'm very likely to reverse course on a "soft" admit if I can get some help arranging close follow up/procedures, nonclinical services, etc... things I'm not very well versed in.

In fact, I usually start my pitch on these patients with that statement. Makes it more an IM consult than an admission request. Tends to go over better, too.

-d

 

[Arcan57]

Agree. Of all the BS that gets admitted frivolously, and you're going to send a patient this sick home? Uh...no. What kind of garbage are people being taught nowadays, at academic centers? Is there something lost in translation here?

Or do they want the gullible amongst you to "save money for the common good for the 'mother system'" at the expense of the patient, and at your own medical-legal risk? Think for yourselves, people, and have some common sense in the face of the utopian dreams of academicians. Asymptomatic neutropenia, with absolutely no fever would be different. Fever/neutropenia? You worked all week and you finally got someone sick. Admit them for God's sake.

As much as I hate the "you admit patients less sick than this" argument when looked at something retrospectively, in this case I agree completely.

 

[TheBlueBlazer]

I would probably admit, but after talking through a similar patient, albeit a kid, with heme, who actually had a relationship with the kid for something unrelated - they basically looked at the number of atypical lymphs, the immature grans and basically said that this is simply a viral process and the kid was at no risk of serious infection and he could go home. They wrote that on the chart and arranged follow-up in a couple days and mom was a reasonable human being so I was ok with it. I think if those stars don't align, the pt comes in.

 

[Birdstrike]

I would probably admit, but after talking through a similar patient, albeit a kid, with heme, who actually had a relationship with the kid for something unrelated - they basically looked at the number of atypical lymphs, the immature grans and basically said that this is simply a viral process and the kid was at no risk of serious infection and he could go home. They wrote that on the chart and arranged follow-up in a couple days and mom was a reasonable human being so I was ok with it. I think if those stars don't align, the pt comes in.

The kid in the OP had a fever >39, though.

 

[EctopicFetus]

As I learned in military school, "there's a fine line between balls and stupidity".

Yes.. I would consider this stupid as well. The patient's workup tells you something bad is about to happen.

 

[TheBlueBlazer]

The kid in the OP had a fever >39, though.

Mine also had a similar fever but was a toddler. The hematologist's explanation was that this was not an atypical thing. We don't routinely send CBCs on otherwise healthy kids with viral illnesses otherwise we'd find it more. When we draw labs in kids, it's usually because we have a higher index of suspicion. Our kid got labs because of an inexperienced intern. He said that kids that are not ill-appearing should get cultures and close follow-up but don't necessarily need to be admitted. He was +/- about empiric antibiotics. I honestly can't remember whether we gave them or not. Not necessarily trying to argue that this is the way we should be doing things, rather communicating the interaction with them and their reasoning.

The OP's pt is different in that he was not a child. I'd imagine the rules are a bit different

 

[Birdstrike]

Mine also had a similar fever but was a toddler. The hematologist's explanation was that this was not an atypical thing. We don't routinely send CBCs on otherwise healthy kids with viral illnesses otherwise we'd find it more. When we draw labs in kids, it's usually because we have a higher index of suspicion. Our kid got labs because of an inexperienced intern. He said that kids that are not ill-appearing should get cultures and close follow-up but don't necessarily need to be admitted. He was +/- about empiric antibiotics. I honestly can't remember whether we gave them or not. Not necessarily trying to argue that this is the way we should be doing things, rather communicating the interaction with them and their reasoning.

The OP's pt is different in that he was not a child. I'd imagine the rules are a bit different

From Birdstrike's Undeniable Laws of Emergency Medicine:

Undeniable Law #47- Over the phone, a consultant will always say your patient has a 99% chance of being okay.

Undeniable Law #48- In person, a consultant will always say your patient has a 1% chance of dying.



Specialists tend to be very cavalier with things that they deal with every day, especially,

1-Over the phone, and

2-When their signature is not on the chart.

I'm not saying they're wrong, but often they'll talk like this, then once they come see the patient it turns into, "Well....you know...uh....we just...well you know...."

"What'd you do with the kid?"

"Ah..well...it's no big deal....we....you know...we see this all the time in clinic and send these kids home..."

"You saw him right? What did you DO with him?"

"Oh, I admitted him."

"Why?"

"Well, you know. If something had happened...


Uh. Yeah. Thanks @&&!•¥***#=. Most of the time this stuff turns out fine, but...



But...



But....

 

[gutonc]

IM/EM resident here so I play both fields. Either way, my usual practice for febrile neutropenia = broad spec abx, pan culture, admit, with the caviot being that I don't trust the follow up for most of the patients I see in my EDs

That being said, there are definitely treatment algorithms which provide for certain lower risk patients to get an initial dose of IV antibiotics, discharged with PO antibiotics, and close follow up. Basically, relatively young, good follow up, well appearing, normal VS aside from temp can get fluoroquinolones and stay on them until their ANC recovers. These patients would get all the cultures drawn in the ED prior to DC. Most importantly, these patients need to have supurb follow up. Like an entire family of supporters and a nearby oncologist who is willing to see them the following morning and follow up with them constantly. In the absence of that, there is too much to lose to send these patients home.

These guidelines are great for chemo induced febrile neutropenia. But that's not what the case was. It was a de novo, symptomatic neutropenia of unknown etiology. Those guidelines don't apply.

This guy will get admitted 11 times out of 10. (I was going to say 12 times out of 10 but this case screwed that up). Sure, it may be nothing, but it's more likely to be something and he needs to come in to get it figured out.

 

[BAM!]

slam dunk admission.

If they decline admission and are otherwise low risk, I'd enlist them in shared decision making with MASCC Risk Index as a guideline as long as they have good follow up.

 

[shookwell]

I'm 50/50 on this one. I would probably talk to the patient if they are a reasonable person and offer them admission vs. outpatient follow up.

 

[TheBlueBlazer]

From Birdstrike's Undeniable Laws of Emergency Medicine:

Undeniable Law #47- Over the phone, a consultant will always say your patient has a 99% chance of being okay.

Undeniable Law #48- In person, a consultant will always say your patient has a 1% chance of dying.

Specialists tend to be very cavalier with things that they deal with every day, especially,

1-Over the phone, and

2-When their signature is not on the chart.
.

Like I wrote in my post, they put a note on the chart to the same effects as the conversation that we had, which again, is a big part of why I was comfortable doing it and why I wouldn't necessarily recommend folks do the same, but I felt it was worthwhile to share my experience.

 

[Birdstrike]

Like I wrote in my post, they put a note on the chart to the same effects as the conversation that we had, which again, is a big part of why I was comfortable doing it and why I wouldn't necessarily recommend folks do the same, but I felt it was worthwhile to share my experience.

It's most likely fine, but this is where I think the game needs to be played with the proper finesse. It does not go like this:

A-"Heme consulted, saw patient, and then I discharged patient." No, no, no.

It goes like this:

B-"Pt diagnosed with fever/neutropenia. Heme consultant for admission/further work up and treatment. Care transferred to Dr XYZ at 18:22." That defines the point at which you're no longer the patient's doctor anymore, and which MD accepts all responsibility. You're done at that specific time.

Maybe I over think these things, but there's a subtle but important difference, as it see it. "A" shows you consulting someone who says the patient isn't sick, and the patient takes a detour to another service briefly but comes back to you and remains under your care. They've added little to the process and offloaded little liability. You're left holding the hot potato.

"B" documents a clear and defined transfer of care, passing of the baton and transfer of responsibility and liability. Whether they discharge the patient in 5 minutes, 5 hours, five days or five weeks, is irrelevant. They become their patient at 18:22 and you've signed off the case.

It may sound like semantics but I think it's important to make the distinction, between "A" which implies you're not really sure what's going on, if it's serious or not or how concerned you really are. "B" shows you've committed to believing the patient is sick, needs a specialist and has a problem outside the scope of your capabilities. If they turn out not to be that sick, have a benign outcome, and do just fine, then great. If not, it's clear who was most concerned, who acted on the matter and used the abundance of caution.

Bottom line: you were uncomfortable enough about the case to start an SDN thread on it after the fact. Therefore, there's a certain amount of CYA in order. Always hedge your bets. It's always better to look like the cautious generalist and let a specialist be the hero (and accept full responsibility) as opposed to trying to be the cavalier hero yourself and accept an oversized portion of responsibility in something out of your scope of bread and butter expertise that makes you uncomfortable.

That's my overly wordy way of saying, "Do what allows you to sleep at night."

 

[Groove]

I can't remember ever sending a >39C <500 ANC out the door. I also can't imagine a situation where my disposition gestalt would be to do anything other than admit them. I'm sure it could happen, but I doubt it.

Work up, early IV abx (8% mortality/hr for the truly sick ones) and easy, easy admission.

 

[BAM!]

It's most likely fine, but this is where I think the game needs to be played with the proper finesse. It does not go like this:

A-"Heme consulted, saw patient, and then I discharged patient." No, no, no.

It goes like this:

B-"Pt diagnosed with fever/neutropenia. Heme consultant for admission/further work up and treatment. Care transferred to Dr XYZ at 18:22." That defines the point at which you're no longer the patient's doctor anymore, and which MD accepts all responsibility. You're done at that specific time.

Maybe I over think these things, but there's a subtle but important difference, as it see it. "A" shows you consulting someone who says the patient isn't sick, and the patient takes a detour to another service briefly but comes back to you and remains under your care. They've added little to the process and offloaded little liability. You're left holding the hot potato.

"B" documents a clear and defined transfer of care, passing of the baton and transfer of responsibility and liability. Whether they discharge the patient in 5 minutes, 5 hours, five days or five weeks, is irrelevant. They become their patient at 18:22 and you've signed off the case.

It may sound like semantics but I think it's important to make the distinction, between "A" which implies you're not really sure what's going on, if it's serious or not or how concerned you really are. "B" shows you've committed to believing the patient is sick, needs a specialist and has a problem outside the scope of your capabilities. If they turn out not to be that sick, have a benign outcome, and do just fine, then great. If not, it's clear who was most concerned, who acted on the matter and used the abundance of caution.

Bottom line: you were uncomfortable enough about the case to start an SDN thread on it after the fact. Therefore, there's a certain amount of CYA in order. Always hedge your bets. It's always better to look like the cautious generalist and let a specialist be the hero (and accept full responsibility) as opposed to trying to be the cavalier hero yourself and accept an oversized portion of responsibility in something out of your scope of bread and butter expertise that makes you uncomfortable.

That's my overly wordy way of saying, "Do what allows you to sleep at night."

Don't get me wrong, I love using this phrase. But this wording only works if you can actually transfer care another doctor. some places operate this way, others don't. The admitting doctor or consultant may refuse admission (probably not in this case, but may in other scenarios). then what are you going to do? transfer to another facility?

 

[Birdstrike]

Don't get me wrong, I love using this phrase. But this wording only works if you can actually transfer care another doctor. some places operate this way, others don't. The admitting doctor or consultant may refuse admission (probably not in this case, but may in other scenarios). then what are you going to do? transfer to another facility?

The wording works just fine. You consult. They come see the patient. You're done. Chart says, "Patient discharged by Dr Superconsultant X."

But then, the nurse comes to you and says, "He's not admitting your patient."

You say, "He's not admitting his patient, you mean? Okay."

Nurse, "He says the patient can go home."

You, "Okay, do whatever Dr X says to do, with his patient."

Then the nurse might say Dr X is gone, you have to click "discharge" in the EMR, or whatever. Okay, fine. You do it (if it's within reason). You got a consultant to say, in writing, the patient is not sick. Sometimes that's the best you can do.

As far as transferring to another facility or not, it depends on if you generally think Dr X is giving a competent recommendation, or if Dr X is dangerous, incompetent, or negligent. If the latter, which hopefully is a very rare occurrence at your hospital, then you disregard the consult and start over by consulting someone else, or transferring if there's no one else available and following up through the administrative route to correct the consultant behavior. Again, this should be rare, but could happen in your career.

Keep in mind, most of the time consultants are going to be right when dealing with a question outside of your scope. You'll learn which ones to trust or not. Most are trust worthy, but none are 100% right, 100% of the time. These are the messy, gray areas of Emergency Medicine.

Bottom line- Advocate for the patient at all times and do what you think is right, and you'll be fine.

 

[BAM!]

Well... at my hospital, when I call for admission often times they may not see the patient until hours after my shift has ended and they're admitted on the floor. Maybe the next morning etc. This means I have to write some general admission orders to get them on the floor, and the internists do the rest at a later time. I get it, it's not ideal. it goes against the mantras of emergency medicine. But we do it. There is no, "they come see the patient and say they're not admitting the patient."

I do some part time at another place where as soon as I consult, the care gets turned over to the internist. Much different work environment and obviously more EM friendly.

The problem is that in some hospitals, the work flow is different. Is it due to discrepancies in departmental power? Maybe it's because of private insurance systems causing systemic breakdowns as internists try to bite off more than they can chew by running busy daytime clinics while admitting a hospital's-worth load of patients.

I agree with your bottom line -- always do the right thing. Most of the time -- this isn't the issue. When it starts to become one, that's when the salesmanship gets turned on.

 

[LaBusqueda]

Well... at my hospital, when I call for admission often times they may not see the patient until hours after my shift has ended and they're admitted on the floor. Maybe the next morning etc. This means I have to write some general admission orders to get them on the floor, and the internists do the rest at a later time. I get it, it's not ideal. it goes against the mantras of emergency medicine. But we do it. There is no, "they come see the patient and say they're not admitting the patient."

I do some part time at another place where as soon as I consult, the care gets turned over to the internist. Much different work environment and obviously more EM friendly.

The problem is that in some hospitals, the work flow is different. Is it due to discrepancies in departmental power? Maybe it's because of private insurance systems causing systemic breakdowns as internists try to bite off more than they can chew by running busy daytime clinics while admitting a hospital's-worth load of patients.

I agree with your bottom line -- always do the right thing. Most of the time -- this isn't the issue. When it starts to become one, that's when the salesmanship gets turned on.

I worked at a shop like that before, and while not ideal for the pt, admissions are generally easier as the trade off for them armchairing admissions was there was NO pushback! There was also no waking up consults overnight unless their presence/management was needed immediately, just put their name on and rounding specialist will see them on the list.
Going to EPIC ruined our nice little world.