Fever and undifferentiated abdominal pain - WWYD?

[gro2001]

Finally, I’ve noticed a disturbing trend in medicine over the past 20 years - doctors demanding tests, making treatment recommendations, and deciding who needs admission by computer screen without ever having seen or examined the patient. The reasons are legion but I suspect the growing dominance of the EHR on our practice probably is a big factor. Granted, there are emergent, time-sensitive circumstances, but those are the extreme minority. I’m talking about doctors pushing-back on admissions and demanding expensive testing without ever having seen the patient. This is a growing poison in the house of Medicine that needs to be clipped.

I feel this could relatively easily be fixed through a combination of a better EHR and better hospital level policies. In some systems, consults have to be documented through the EHR (ie a consult is an actual order you enter through the system that automatically sends a page to the consultant and gets entered in the medical record). That has the advantage of time stamping the consult and allows for the documentation of the consultation question. What's missing is that this should result in a to-do popping up on the consultant's screen requiring them to write a note about the patient and a hospital policy saying they actually do have to document if consulted. That way, once the page goes out, it's not the ED doc trying to convince them to actually see the patient when giving phone recs. Or if they really do feel comfortable with some recommendation over the phone, then they should have no problem documenting it. And if it's completely a wrong page, it should be quick and easy to type a one line "Orthopedics is not covering spine surgery this week, according to the schedule it's neurosurgery. Consult to neurosurgery placed instead. Signing off."

---

Members don't see this ad.

 

[GeneralVeers]

I feel this could relatively easily be fixed through a combination of a better EHR and better hospital level policies. In some systems, consults have to be documented through the EHR (ie a consult is an actual order you enter through the system that automatically sends a page to the consultant and gets entered in the medical record). That has the advantage of time stamping the consult and allows for the documentation of the consultation question. What's missing is that this should result in a to-do popping up on the consultant's screen requiring them to write a note about the patient and a hospital policy saying they actually do have to document if consulted. That way, once the page goes out, it's not the ED doc trying to convince them to actually see the patient when giving phone recs. Or if they really do feel comfortable with some recommendation over the phone, then they should have no problem documenting it. And if it's completely a wrong page, it should be quick and easy to type a one line "Orthopedics is not covering spine surgery this week, according to the schedule it's neurosurgery. Consult to neurosurgery placed instead. Signing off."

All good suggestions, but to me the bigger problem is that we are using pagers in 2019? Why are we using an unreliable, 40 year old technology in the age of text messages and SnapChat? It shouldn't take 30 min or an hour for a consultant to call me back when communication is instantaneous. I do agree that consults should formally enter their consult in the medical record. Easily done with current technology.

 

[gro2001]

All good suggestions, but to me the bigger problem is that we are using pagers in 2019? Why are we using an unreliable, 40 year old technology in the age of text messages and SnapChat? It shouldn't take 30 min or an hour for a consultant to call me back when communication is instantaneous. I do agree that consults should formally enter their consult in the medical record. Easily done with current technology.

Yeah, the medical communication infrastructure has to change. I was using the paging term loosely. Some sort of instant messaging would probably be better, could be more directly tied into the EHR, easier for everyone to use.

 

[TheSingularity]

The hospital is not interested in forcing the orthopedist on call to actually consult on the ED patient. The hospital is interested in keeping the orthopedist happy so they can continue to generate massive revenue for the hospital.

 

[deleted836128]

The last time I checked she was in the ICU but doing well on antibiotics and hadn’t had to undergo surgery. Such a crazy case and I shudder thinking I could have sent her home. Come to think of it, she was a pretty poor historian and it was like pulling teeth to get her to elaborate on anything. Pretty scary.

 

[Rendar5]

It’s a minefield out there. That’s what good discharge instructions are for, if you were to have sent her home. Without seeing the person, if they didn’t look sick or my clinical gestalt weren’t ringing, I would’ve sent them home and been ok with that decision even in retrospect because they would have had good dc instructions to come back. That said the couple of RPA cases I’ve seen the patient did look like crap and warranted admission solely on gestalt.

 

[thegenius]

The last time I checked she was in the ICU but doing well on antibiotics and hadn’t had to undergo surgery. Such a crazy case and I shudder thinking I could have sent her home. Come to think of it, she was a pretty poor historian and it was like pulling teeth to get her to elaborate on anything. Pretty scary.

Nah...don't shudder about it. If you ask someone several times "do you have any problems in your neck" and they say no everytime...and they end up having pathology in their neck over the next 1-2 days. There is only so much you can do.

That being said, an independent predictor of being admitted in my book is having a crappy memory or just not being able to articulate anything. Regardless of what language you speak.

 

[deleted109597]

munication infrastructure has to change. I was using the paging term loosely. Some sort of instant messaging would probably be better, could be more directly tied into the EHR, easier for everyone to use.

These things exist.
We use Cortext, and it links both to your computer logon and your phone. So you get a notification as long as you have one of them turned on.


And we use Meditech, so it's not like we are that fancy

 

[WilcoWorld]

It’s a minefield out there. That’s what good discharge instructions are for, if you were to have sent her home. Without seeing the person, if they didn’t look sick or my clinical gestalt weren’t ringing, I would’ve sent them home and been ok with that decision even in retrospect because they would have had good dc instructions to come back. That said the couple of RPA cases I’ve seen the patient did look like crap and warranted admission solely on gestalt.

Y'know, now that I think of it, in most of the RPA cases I've seen - the diagnosis was made on a return visit of some sort. It almost seems like that's just the natural history of RPA - get seen once & get diagnosed with a URI...get worse, come back & get diagnosed with an RPA.

 

[Apollyon]

Y'know, now that I think of it, in most of the RPA cases I've seen - the diagnosis was made on a return visit of some sort. It almost seems like that's just the natural history of RPA - get seen once & get diagnosed with a URI...get worse, come back & get diagnosed with an RPA.

Well, since it IS an abscess, that makes sense. As I tell the patients, "An abscess is like smoke from a fire - it takes time to be generated". And, what, clinically, clues me in? Fever, and the trismus. The one I missed as a resident, well, my attending missed too. However, as I, and the dearly absent Roja, have stated, "Good discharge instructions are better than an accurate diagnosis". The guy got good DC instructions, came back, and did fine. (And it's a good story, because it would have been a political bomb if he died.)

 

[WilcoWorld]

"Good discharge instructions are better than an accurate diagnosis".

 

[thegenius]

Y'know, now that I think of it, in most of the RPA cases I've seen - the diagnosis was made on a return visit of some sort. It almost seems like that's just the natural history of RPA - get seen once & get diagnosed with a URI...get worse, come back & get diagnosed with an RPA.

Yea...RPA's are rare. And despite what we learn about how it's a medical emergency and you gotta evaluate the airway THIS SECOND RIGHT NOW DON'T EVER LEAVE THE ROOM AND GET A CRIC KIT NOW...the patients are usually stable and they are not gonna die in 30 minutes, or even in days. They get admitted, occasionally go to the ER and get a drainage, but most of the time they get Abx and they do fine.

I don't think RPA is on the list of diagnoses where you go to bed just fine with no symptoms, and die while you are sleeping a few hours later

 

[thegenius]

Discharging patients is probably the most dangerous thing we do. I agree with that concept!

 

[KGflyboy]

I probably would have admitted that patient. Old people and abdominal pain are a bad combo. Old people, abdominal pain, and fevers are even worse.

Old is a relative term though. 60 doesn't really seem old to me, but it's old enough to make me hesitate.

Many patients with simple infections meet Sepsis criteria. Heck, even a 20 year old with steep throat meets sepsis criteria. I don't admit everyone with sepsis criteria, but I admit the vast majority and need a good reason not to if I discharge them. In this case, I think the outcome would have been the same if you had sent her home with good return precautions. She would have returned and been admitted when she failed to improve or worsened.

But if there was any morbidity in this case, whether preventable or not, it would be blamed on you in court for not admitting her on the first visit if you had sent her home. Heck even if she had normal vitals and labs on the first visit she may have gotten some money out of a lawsuit if she came back with a retropharyngeal abscess. It's always better to make the diagnosis or admission on the first visit then at the bounce back visit from a legal perspective as far as I'm concerned.