What is the NNT for an EM physician to have a better outcome...

[Boatswain2PA]

---

Members do not see this ad.

Comparing physicians and PAs as solo providers in a rural... : JAAPA

f the host institution were analyzed. Although minor variances in metrics were appreciated, transfers, 72-hour returns, and death rates were all similar among provider types. Results from this work do not demonstrate a clinically meaningful difference among ED metrics between physicians and PAs...

journals.lww.com

Just came off shift where I worked alongside an EM doc a few years out of 3 year residency. Last patient with pneumonia failing bipap with hypercapbeic failure. Elderly, obese, COPD on 5 L at home, multiple neck surgeries on what little neck there was, kyphotic, malampatti 3. Looked like a terrible intubation. Did DSI with ketamine and when we laid pt back the neck was stiff as a board with neck locked in flexion. Just a terrible situation.

Doc got that tube so smoothly it was just amazing. Made it look easy. Got it in about 3 seconds.

I'm confident I could have gotten it too. Would have taken longer. I would have "ticked" down the tongue instead of just inserting to just the right depth and position....but I'm confident I would have gotten it.

Would the outcome have been different had I been single coverage? I dunno.

In my experience there is an obvious difference in how EM trained docs think and practice. But how much of a difference does this make with actual outcomes? There HAS to be a point of diminishing returns with training (are 4 year EM residencies better than 3? What's the NNT?) Are EM docs over trained for what they do?

What is the NNT to show an improved outcome for an EM trained doc over a non-em trained doc, or over an experienced EM PA, or a fresh PA, or a nurse....

 

[TheComebacKid]

You can try to get all academic about it if you want. But it's very simple. More training means you are more comfortable and better in these high stress situations. The question of 3 vs 4 year residencies is one that clearly has some component of diminishing returns. But EM docs vs FPs, or midlevels, it's really no contest. The performance of a senior resident versus an intern on airway makes it very clear that the additional training makes a difference.

No disrespect to you, I think you could have gotten this airway too. The vast majority of "difficult airways" aren't truly difficult. They need some minor adjustments, maybe a couple of extra seconds here or there. But if it comes down to a challenging airway like this one, seconds matter and you want the most experienced person for the job.

I don't think "overtrained" is a bad thing. If 3 year EM residencies train "overtrained" physicians, we should all be good with that.

 

[dadaddadaBATMAN]

Comparing physicians and PAs as solo providers in a rural... : JAAPA

f the host institution were analyzed. Although minor variances in metrics were appreciated, transfers, 72-hour returns, and death rates were all similar among provider types. Results from this work do not demonstrate a clinically meaningful difference among ED metrics between physicians and PAs...

journals.lww.com

Just came off shift where I worked alongside an EM doc a few years out of 3 year residency. Last patient with pneumonia failing bipap with hypercapbeic failure. Elderly, obese, COPD on 5 L at home, multiple neck surgeries on what little neck there was, kyphotic, malampatti 3. Looked like a terrible intubation. Did DSI with ketamine and when we laid pt back the neck was stiff as a board with neck locked in flexion. Just a terrible situation.

Doc got that tube so smoothly it was just amazing. Made it look easy. Got it in about 3 seconds.

I'm confident I could have gotten it too. Would have taken longer. I would have "ticked" down the tongue instead of just inserting to just the right depth and position....but I'm confident I would have gotten it.

Would the outcome have been different had I been single coverage? I dunno.

In my experience there is an obvious difference in how EM trained docs think and practice. But how much of a difference does this make with actual outcomes? There HAS to be a point of diminishing returns with training (are 4 year EM residencies better than 3? What's the NNT?) Are EM docs over trained for what they do?

What is the NNT to show an improved outcome for an EM trained doc over a non-em trained doc, or over an experienced EM PA, or a fresh PA, or a nurse....

It depends what pathology you’re referring to.

the case you describe isn’t the one I really worry about, because the management is straightforward and procedures are not what I worry about.

I worry about subtle presentations of sick patients that are missed.

the pas I work with are highly skilled and intelligent people. I see about 20 people in an 8 hr shift, usually 10-12 on my own and 8-10 with one of our pas, +/- 5. I would say on those 10 pa patients I make a substantive change in management like where they are dispositioned or what the principal diagnosis is about once a shift.

not every one of these changes necessarily avoids a negative outcome, but a fair number of them do. So id put anecdotal nnt at about 10-20.

the changes in management are way more common in Peds.

 

[Boatswain2PA]

More training means you are more comfortable and better in these high stress situations. The question of 3 vs 4 year residencies is one that clearly has some component of diminishing returns. But EM docs vs FPs, or midlevels, it's really no contest.

Absolutely. I wish I had a 3-4 year EM residency.

But what is the sweet spot in that diminishing return curve. This study shows it might not be as far down the educational pathway a we thought.

id put anecdotal nnt at about 10-20.

Yet this study suggests it is >25000.

 

[DeadCactus]

Transfers, 72 hour returns, and death rates are pretty shallow endpoints.

It's a single facility. The PA's are unusually experienced. Maybe I missed it but I don't see anything about how patients are assigned to PA vs physician and what the nature of collaboration between PA and physician is which would greatly confound things. Overall acuity appears to be low even for the low overall acuity of EM in general.

So sure, the NNT is > 26,000 in a low acuity rural ED where unusually experienced PA's see patients with some unknown level of physician oversight and support if our endpoints are only transfer rate, 72 hour return, and death rate. I could also just sit at the door and shoot people with paintballs as they approach. No transfers, no one's coming back, and at 0.1% mortality I'm probably still doing alright on that NNT.

Edit: To be clear, I have a generally positive attitude toward PA's. This study is reasonable for concluding that an appropriately supervised PA can safely see patients in a low-acuity rural ED. I think most would find that a reasonable statement. I don't find this study adequate to declare emergency physicians overtrained even if limiting the statement to rural hospitals.

 

[dadaddadaBATMAN]

Absolutely. I wish I had a 3-4 year EM residency.

But what is the sweet spot in that diminishing return curve. This study shows it might not be as far down the educational pathway a we thought.

Yet this study suggests it is >25000.

Just because it is a study does not make it good data.

the ed in your “study” had under 28,000 visits in three years.

About 9000 a year.
The admission rate was 5% in this ed
There were a total of 21 deaths in 3 years

…I could put a blind wombat in that ed for a shift a day and it would take several months for anyone to notice a difference in mortality, and even then it probably would not be statistically significant.

more people died in our er in two months than died in a 3 year period of that study.

in short, your study took place in an urgent care and looked for outcomes like deaths and bounce backs. It’s hard to find a statistical difference in that population

the average admit rate where I work is 45%. 10% icu. 10% transfer.

additional edit: based on the mortality, even the admits in that population were probably pretty soft. And the average age of 33/35 years supports this. 450 admits a year. We admit that many about every 10 days.

and it’s not like we’re some super bad ass trauma center either. I work in a slightly slummy suburb in an area with a lot of substance abuse

 

[Boatswain2PA]

Maybe I missed it but I don't see anything about how patients are assigned to PA vs physician

Since it's a solo coverage ED I assume randomly selected to whomever is on duty at the time.

About 9000 a year.
The admission rate was 5% in this ed
There were a total of 21 deaths in 3 years

…I could put a blind wombat in that ed for a shift a day and it would take several months for anyone to notice a difference in mortality, and even then it probably would not be statistically significant.

more people died in our er in two months than died in a 3 year period of that study.

in short, your study took place in an urgent care and looked for outcomes like deaths and bounce backs. It’s hard to find a statistical difference in that population

the average admit rate where I work is 45%. 10% icu. 10% transfer.

additional edit: based on the mortality, even the admits in that population were probably pretty soft. And the average age of 33/35 years supports this. 450 admits a year. We admit that many about every 10 days.

Concur this appears to be a low acuity department.

Also, at 1 pph annually there is more time to slow down, think through things, and look things up.

 

[Matthew9Thirtyfive]

Yet this study suggests it is >25000.

Consider the author and their objective with this study, then scrutinize the data.

 

[turkeyjerky]

Having worked in ERs like the one described, i can say confidently that this ‘study’ has no external validity to the majority of emergency medicine.

Consider the author and their objective with this study, then scrutinize the data.

Yeah, OP has a well know history of these type of posts and (maybe?) just doxxed himself

 

[Groove]

The author of the study is a PA with no MD/DO co-authors. He’s also an EM provider…so did he perform this study on himself at the ED where he works I wonder? No biases at all, I’m sure. This is such a dumb study, it’s not even worth the neurons needed to criticize it.

 

[dadaddadaBATMAN]

The author of the study is a PA with no MD/DO co-authors. He’s also an EM provider…so did he perform this study on himself at the ED where he works I wonder? No biases at all, I’m sure. This is such a dumb study, it’s not even worth the neurons needed to criticize it.

If it paid reasonably well I’d be very tempted to work at this place with 24 visits a day (I’m guessing it does not) and 5% admissions, because man that sounds like easy money.

unfortunately I think the skill atrophy would leave me as cachectic as a 95 yo mi maw with cirrhosis after a 6 month stay in the icu. After a while it would probably be hard to tell the difference between me and the other gomers. That may be another factor in this study, though I struggle to dignify it with that level of analysis

 

[WilcoWorld]

It feels like we're badgering @Boatswain2PA here, when I see the original post as a humble and genuine question. I think this is the kind of discourse we should be open to.

Then again, maybe I'm off on my perceptions

 

[DeadCactus]

Since it's a solo coverage ED I assume randomly selected to whomever is on duty at the time.

That assumes the PAs/NP and physicians were randomly scheduled to the shifts they fill, maybe that's true but rather atypical. Why is Table 1 not a comparison of patient demographics and clinical characteristics as is pretty standard in research articles?

I don't disagree with the idea that there are settings and situations where a board certified emergency physician is an inefficient use of resources. I disagree with trying to extrapolate a NNT >26,000 from this study.

 

[dadaddadaBATMAN]

It feels like we're badgering @Boatswain2PA here, when I see the original post as a humble and genuine question. I think this is the kind of discourse we should be open to.

Then again, maybe I'm off on my perceptions

I have already long surpassed my daily time allotment for this site, but I treated it this way with my first response, and only went into detailed criticism of the study when I was informed by the person who posed the question that the nnt is > 25,000.

i find it irritating and disingenuous to ask a question when you claim to know the answer. I find it especially irritating to ask a question like this when you have a clear agenda.

if you wanna make the claim that two patients a year in my ed have their management changed meaningfully by having a physician in charge, you better have rigorous data, and this ain’t it. I wrote out some examples from the last month but I don’t really wanna dox myself. I can think of 5 out of 100-150 pa patients I changed outcome on substantively in the last month, and several others that I changed dispo but not outcome.

 

[Boatswain2PA]

I was informed by the person who posed the question that the nnt is > 25,000.

I didn't say the NNT was >25k. DYou gave your opinion that the NNT was 10-20. I said this study suggested it was >25k.

One thing to consider with your examples is how much of that is simply differences in practice patterns that would also have no real difference in outcomes. I've worked with EPs who want lactate on almost everyone, and with EPs who only want lactate on the clearly septic. Probably no measurable difference in outcomes, just practice patterns.

Conversely you are likely in a very busy shop where your much higher training makes a bigger difference.

 

[CoolDoc1729]

Just because it is a study does not make it good data.

the ed in your “study” had under 28,000 visits in three years.

About 9000 a year.
The admission rate was 5% in this ed
There were a total of 21 deaths in 3 years

…I could put a blind wombat in that ed for a shift a day and it would take several months for anyone to notice a difference in mortality, and even then it probably would not be statistically significant.

more people died in our er in two months than died in a 3 year period of that study.

in short, your study took place in an urgent care and looked for outcomes like deaths and bounce backs. It’s hard to find a statistical difference in that population

the average admit rate where I work is 45%. 10% icu. 10% transfer.

additional edit: based on the mortality, even the admits in that population were probably pretty soft. And the average age of 33/35 years supports this. 450 admits a year. We admit that many about every 10 days.

and it’s not like we’re some super bad ass trauma center either. I work in a slightly slummy suburb in an area with a lot of substance abuse

Are you suggesting that PAs are over trained to work in the ER compared to blind wombats? What’s that NNT?
That comment made me literally LOL, thank you 🙏

 

[Matthew9Thirtyfive]

Why is Table 1 not a comparison of patient demographics and clinical characteristics as is pretty standard in research articles?

Can’t imagine why they didn’t want to include that information.

 

[Groove]

It feels like we're badgering @Boatswain2PA here, when I see the original post as a humble and genuine question. I think this is the kind of discourse we should be open to.

Then again, maybe I'm off on my perceptions

Posting a study on MLP and physician parity in the ED on a physician dominated EM forum? It smells too much like a humblebrag question in the guise of "academic discussion". Maybe I'm extra crispy this week from all my hours and lack of sleep...

 

[thegenius]

I don't know of any data and I'm not going to search for it. I think outcomes of EM trained vs NP/PA will show very little difference for low acuity stuff (Emergency Severity Index ESI 3,4,5), probable statistical difference for ESI 2, and definite statistical difference for ESI 1.

If all we saw were ESI 1 and 2, there would be people dying left and right in ER's across the nation if they were run by NP's and PAs as compared to ER trained docs.

Unfortunately, Emergency Medicine is just primary care medicine for the lazy 50m people out there....so most of what we see is ESI 3 and 4. That's why administrators want NP's because they are cheaper.

 

[Porfirio]

Comparing physicians and PAs as solo providers in a rural... : JAAPA

f the host institution were analyzed. Although minor variances in metrics were appreciated, transfers, 72-hour returns, and death rates were all similar among provider types. Results from this work do not demonstrate a clinically meaningful difference among ED metrics between physicians and PAs...

journals.lww.com

Just came off shift where I worked alongside an EM doc a few years out of 3 year residency. Last patient with pneumonia failing bipap with hypercapbeic failure. Elderly, obese, COPD on 5 L at home, multiple neck surgeries on what little neck there was, kyphotic, malampatti 3. Looked like a terrible intubation. Did DSI with ketamine and when we laid pt back the neck was stiff as a board with neck locked in flexion. Just a terrible situation.

Doc got that tube so smoothly it was just amazing. Made it look easy. Got it in about 3 seconds.

I'm confident I could have gotten it too. Would have taken longer. I would have "ticked" down the tongue instead of just inserting to just the right depth and position....but I'm confident I would have gotten it.

Would the outcome have been different had I been single coverage? I dunno.

In my experience there is an obvious difference in how EM trained docs think and practice. But how much of a difference does this make with actual outcomes? There HAS to be a point of diminishing returns with training (are 4 year EM residencies better than 3? What's the NNT?) Are EM docs over trained for what they do?

What is the NNT to show an improved outcome for an EM trained doc over a non-em trained doc, or over an experienced EM PA, or a fresh PA, or a nurse....

One-way ANOVA - Its preference to multiple t-tests and the assumptions needed to run this test | Laerd Statistics

Using the one-way ANOVA as a means to control the increase in Type 1 errors with multiple t-tests and understanding the assumptions underlying the test.

statistics.laerd.com

Big part of the problem with this study.

 

[alpinism]

I don't know of any data and I'm not going to search for it. I think outcomes of EM trained vs NP/PA will show very little difference for low acuity stuff (Emergency Severity Index ESI 3,4,5), probable statistical difference for ESI 2, and definite statistical difference for ESI 1.

If all we saw were ESI 1 and 2, there would be people dying left and right in ER's across the nation if they were run by NP's and PAs as compared to ER trained docs.

Unfortunately, Emergency Medicine is just primary care medicine for the lazy 50m people out there....so most of what we see is ESI 3 and 4. That's why administrators want NP's because they are cheaper.

Yep this is exactly my experience.

Not sure how anyone can be surprised that a large group of patients with no life threatening emergencies whatsoever have the same outcomes regardless of being seen by PAs or MDs. I can almost guarantee you that if you did the same trial with EMTs or RNs you'd find the same results. Hell you could probably assign one of your frequent flier drunks to see all ESI 4s and 5s and they'd provide decent care. Its not that hard when most patents need nothing more than to be sent home and back to bed. What's funny is the whole NPs in EM situation is just because hospitals have finally realized this and since they can't legally hire a drunk off the streets they've settled for hiring nurse practitioners instead who can write scripts for antibiotics for viruses to satisfy all their drug seeking customers.

 

[The Knife & Gun Club]

Comparing physicians and PAs as solo providers in a rural... : JAAPA

f the host institution were analyzed. Although minor variances in metrics were appreciated, transfers, 72-hour returns, and death rates were all similar among provider types. Results from this work do not demonstrate a clinically meaningful difference among ED metrics between physicians and PAs...

journals.lww.com

Just came off shift where I worked alongside an EM doc a few years out of 3 year residency. Last patient with pneumonia failing bipap with hypercapbeic failure. Elderly, obese, COPD on 5 L at home, multiple neck surgeries on what little neck there was, kyphotic, malampatti 3. Looked like a terrible intubation. Did DSI with ketamine and when we laid pt back the neck was stiff as a board with neck locked in flexion. Just a terrible situation.

Doc got that tube so smoothly it was just amazing. Made it look easy. Got it in about 3 seconds.

I'm confident I could have gotten it too. Would have taken longer. I would have "ticked" down the tongue instead of just inserting to just the right depth and position....but I'm confident I would have gotten it.

Would the outcome have been different had I been single coverage? I dunno.

In my experience there is an obvious difference in how EM trained docs think and practice. But how much of a difference does this make with actual outcomes? There HAS to be a point of diminishing returns with training (are 4 year EM residencies better than 3? What's the NNT?) Are EM docs over trained for what they do?

What is the NNT to show an improved outcome for an EM trained doc over a non-em trained doc, or over an experienced EM PA, or a fresh PA, or a nurse....

The bigger problem is the variation in training. At our community site I work with some *dirty* PAs who can go toe to toe with attendings on the vast majority of cases. They’re stellar.

But! That’s because our facility is an SDG, they can chose their PAs, train and retain them. Our other sites with a “pick your own NP” adventure are disasters with all sorts of mismanagement.

So you can’t just split the cohort into APP vs MD. There’s a lot of variation. And the variation is part of the problem. If you can’t guarantee a well trained provider, then you can’t be blanket granted independence.

Based on no experience hiring/overseeing PAs, I’d say the NNT for an inexperienced or green provider is probably around 1/100. For a seasoned, trained APP, maybe 1/1000?

 

[TheComebacKid]

I don't know of any data and I'm not going to search for it. I think outcomes of EM trained vs NP/PA will show very little difference for low acuity stuff (Emergency Severity Index ESI 3,4,5), probable statistical difference for ESI 2, and definite statistical difference for ESI 1.

If all we saw were ESI 1 and 2, there would be people dying left and right in ER's across the nation if they were run by NP's and PAs as compared to ER trained docs.

Unfortunately, Emergency Medicine is just primary care medicine for the lazy 50m people out there....so most of what we see is ESI 3 and 4. That's why administrators want NP's because they are cheaper.

I've always been worried more about the ESI 3, 4, 5 patients with midlevels. Most of the time if for whatever reason they are seeing a sicker patient, they will admit the patient, or there will be a consultant involved and an actually practicing physician will be involved in the care. I worry about the ESI level 3 patient that is 55 years old with multiple cardiovascular risk factors that gets diagnosed with GERD. Or the ESI level 4 vaginal bleeding that has a ruptured ectopic.

I think with enough reps you can train a monkey to manage an airway, put in a line, and get an ICU doc on the phone to accept your patient. Teasing out the patients who get triaged as "low acuity" but are actually hiding badness is much more challenging. That is where extra training and experience really makes a difference.

 

[thegenius]

Absolutely. I wish I had a 3-4 year EM residency.

But what is the sweet spot in that diminishing return curve. This study shows it might not be as far down the educational pathway a we thought.

Yet this study suggests it is >25000.

For what it's worth, an NNT of 25,000 seems absurd and prima facie should make you question the methods of the paper. I didn't read it, but that outcome just doesn't make sense.

 

[thegenius]

Transfers, 72 hour returns, and death rates are pretty shallow endpoints.

It's a single facility. The PA's are unusually experienced. Maybe I missed it but I don't see anything about how patients are assigned to PA vs physician and what the nature of collaboration between PA and physician is which would greatly confound things. Overall acuity appears to be low even for the low overall acuity of EM in general.

So sure, the NNT is > 26,000 in a low acuity rural ED where unusually experienced PA's see patients with some unknown level of physician oversight and support if our endpoints are only transfer rate, 72 hour return, and death rate. I could also just sit at the door and shoot people with paintballs as they approach. No transfers, no one's coming back, and at 0.1% mortality I'm probably still doing alright on that NNT.

Edit: To be clear, I have a generally positive attitude toward PA's. This study is reasonable for concluding that an appropriately supervised PA can safely see patients in a low-acuity rural ED. I think most would find that a reasonable statement. I don't find this study adequate to declare emergency physicians overtrained even if limiting the statement to rural hospitals.

I think ER physicians are appropriately trained for the original intention of Emergency Departments, but these days for a variety of reasons EDs across the nation basically provides free anything care for all those without insurance, Medicaid, and Medicare.

 

[thegenius]

It's going to be unethical to ever design a true study like what is suggested because it would be malpractice to allow PAs and NPs to manage anyone who walks into the ED without supervision and all ESI's. It won't pass the IRB.

 

[thegenius]

I've always been worried more about the ESI 3, 4, 5 patients with midlevels. Most of the time if for whatever reason they are seeing a sicker patient, they will admit the patient, or there will be a consultant involved and an actually practicing physician will be involved in the care. I worry about the ESI level 3 patient that is 55 years old with multiple cardiovascular risk factors that gets diagnosed with GERD. Or the ESI level 4 vaginal bleeding that has a ruptured ectopic.

I think with enough reps you can train a monkey to manage an airway, put in a line, and get an ICU doc on the phone to accept your patient. Teasing out the patients who get triaged as "low acuity" but are actually hiding badness is much more challenging. That is where extra training and experience really makes a difference.

I know what you are getting at, and statistically there is so little true pathology with the ESI 3-5 (less so 3) that it would require thousands of patients to see a difference. I find the ESI's where I work, or whomever defines them, are pretty good at it.

Another problem with this study is defining "bad medicine". It's hard to do. You can't just look at death. You have to look at shiiit decision making. How do you measure that.

I would be very scared to have an average PA/NP take care of this patient:
62 yo man with kidney transplant, HTN, afib on AC comes in with shortness of breath and vague upper abdominal pain with HR 146, BP 90/40, SpO2 82% RA, Temp 99.2, and RR 38 and he looks terrible on exam.

Where ER docs excel is just flat out managing this patient with little to no data. We know exactly what to do and in what order. We don't need labs. We basically need just a few pieces of data and we can resuscitate this guy to health. The only other doctor who I think could handle this kind of thing is probably a critical care doc. Certainly not Internal Medicine, Cardiology, Pulmonology, Surgery, Nephrology, and CERTAINLY not a PA/NP. Everyone would be ordering 45 labs, ABGs, CT's of the entire body, and then not know how to interpret the results and will have done little in the first 2 hours to resuscitate and stabilize this guy.

 

[Vermicious Knid]

anyone else wondering why this medical disaster of a patient even got intubated in the first place? Or maybe that is just my non USA bias?

 

[GeneralVeers]

anyone else wondering why this medical disaster of a patient even got intubated in the first place? Or maybe that is just my non USA bias?

Cuz in 'Merica we have to "do everything" even if it is unethical, doesn't help the patient and costs $millions

 

[e30ftw]

I don't think it's possible to really answer the original question since it would be unethical. The way you'd design the study is find a high acuity single coverage ED, staff it even days with ABEM physicians, staff it odd days with APPs (solo coverage), then look for trends in sentinel events, high-risk bounce backs, etc. No one is going to let you do that with good reason.

Does anyone practice where ED APPs see the same patients as ABEM docs? We have great APPs but none of them have any interest in critical patients, many don't like seeing high-risk chest pain or a simple CVA. None of them do critical care procedures; none of them intubate, place CVLs, do LPs or ortho reductions other than simple shoulder. This is throughout a large healthcare system.

Just curious since the only time I see concern about APP autonomy or EPs being "overtrained" is on SDN. In the real world every crashing patient, every time, gets seen by a physician, or taken over by the physician if an APP saw them first. This just seems like an answer to a question that doesn't really exist..

 

[Matthew9Thirtyfive]

Just curious since the only time I see concern about APP autonomy or EPs being "overtrained" is on SDN. In the real world every crashing patient, every time, gets seen by a physician, or taken over by the physician if an APP saw them first. This just seems like an answer to a question that doesn't really exist..

Except when they don’t.

 

[CajunMedic]

Does anyone practice where ED APPs see the same patients as ABEM docs? We have great APPs but none of them have any interest in critical patients, many don't like seeing high-risk chest pain or a simple CVA. None of them do critical care procedures; none of them intubate, place CVLs, do LPs or ortho reductions other than simple shoulder. This is throughout a large healthcare system.

They do where I'm a resident. The more aggressive ones tend to work shifts with attendings that don't have residents. There was a controversy during my 2nd year when the dictate came down that they had to all be credentialed for intubation and CVL's. A couple would perform procedures when there were residents on shift who were supposed to have priority for all procedures. There are legitimately shifts where the acuity is such that the residents are handling multiple critical patients and the APP winds up with one as well

 

[robellis]

The bigger problem is the variation in training. At our community site I work with some *dirty* PAs who can go toe to toe with attendings on the vast majority of cases. They’re stellar.

But! That’s because our facility is an SDG, they can chose their PAs, train and retain them. Our other sites with a “pick your own NP” adventure are disasters with all sorts of mismanagement.

So you can’t just split the cohort into APP vs MD. There’s a lot of variation. And the variation is part of the problem. If you can’t guarantee a well trained provider, then you can’t be blanket granted independence.

Based on no experience hiring/overseeing PAs, I’d say the NNT for an inexperienced or green provider is probably around 1/100. For a seasoned, trained APP, maybe 1/1000?

You’re pretty confident in your assessments for an intern

 

[The Knife & Gun Club]

You’re pretty confident in your assessments for an intern

I mean, that’s just like, your option man…

Really, I like to think Im pretty measured when posting on here and usually qualify all my posts with something like “hey, disclaimer, I’m a resident” like I did in my last line here.

I enjoy engaging in hearty discussion about medicine and EM stuff and would love to hear your thoughts if you think something I said is outta whack.

 

[Boatswain2PA]

Posting a study on MLP and physician parity in the ED on a physician dominated EM forum? It smells too much like a humblebrag question in the guise of "academic discussion".

I understand how it could come across this way, but that is not my intention. Who staffs what, where, at what cost to the payers and the patients is an ongoing discussion. This is one part of it.

For what it's worth, an NNT of 25,000 seems absurd

Perhaps, although neither this paper nor myself made that proposition. The paper said in that ONE ED with >25K visits during the study, there were no difference in measured outcomes.

I think ER physicians are appropriately trained for the original intention of Emergency Departments, but these days for a variety of reasons EDs across the nation basically provides free anything care for all those without insurance, Medicaid, and Medicare.

I think they are best trained. "Appropriate" is too subjective I think.

It's going to be unethical to ever design a true study like what is suggested because it would be malpractice to allow PAs and NPs to manage anyone who walks into the ED without supervision and all ESI's. It won't pass the IRB.

Unless there are enough retrospective observational studies like this to open that door. As the EP job market gets flooded and EPs move to the secondary and maybe tertiary care centers I could see there being more availability for such studies.

And in many small EDs PA & NPs do manage anyone who walks into the ED (with little to no real supervision). These EDs are dramatically different from the EDs that EPs work in. So much so that perhaps we shouldn't really even call them EDs. Comparing outcomes between these types of EDs would be useless as confounding factors like transportation times, skill of transportation crew, etc would be impossible to correct for.

I would be very scared to have an average PA/NP take care of this patient:
62 yo man with kidney transplant, HTN, afib on AC comes in with shortness of breath and vague upper abdominal pain with HR 146, BP 90/40, SpO2 82% RA, Temp 99.2, and RR 38 and he looks terrible on exam.

What percentage, do you think, of the 140 million annual ED visits are visits like this? How many of the 5500 EDs in the US have to have a BC EP in house 24/7 to manage this percentage of patients?

Does anyone practice where ED APPs see the same patients as ABEM docs?

This study was done in one such hospital.

Just curious since the only time I see concern about APP autonomy or EPs being "overtrained" is on SDN. In the real world every crashing patient, every time, gets seen by a physician, or taken over by the physician if an APP saw them first. This just seems like an answer to a question that doesn't really exist..

In big EDs, where most EPs are trained and work, this is true. In many areas of the country there are no big hospitals like this, just a local hospital with an ED that might have a FP doc in the ED, but just as likely will have a PA, or a new-grad NP. As the EP market continues getting oversaturated the EPs will have to move away from the big city EDs into the smaller EDs. Meanwhile NP programs (and to a lesser extent PA programs) are cranking out NPs by the thousands, so beancounter CEO have to choose between a $60/hr NP, a $90/hr PA, or a $200/hr EP to staff these EDs. I think this question really DOES exist outside of secondary/tertiary care centers.

 

[sylvanthus]

The very fact this discussion is even occurring should tell people all they need to know about the state of emergency medicine. Pin this uptop and save med students from the folly of entering a dying specialty.

 

[Groove]

The very fact this discussion is even occurring should tell people all they need to know about the state of emergency medicine. Pin this uptop and save med students from the folly of entering a dying specialty.

Agreed... I had a med student rotating with me not long ago who was so gung ho interested in EM and anytime I would stop to caution him about the state of the specialty and future challenges he/she might encounter, it's like their eyes glazed over and they got this silly grin on their face and said something to the effect "oh gosh...I'm sure it will be fine! These things always work out! I mean...we wouldn't have residency spots open if the jobs weren't there...right?!". I just inwardly rolled my eyes.

 

[turkeyjerky]

I understand how it could come across this way, but that is not my intention. Who staffs what, where, at what cost to the payers and the patients is an ongoing discussion. This is one part of it.

I agree with the poster above who estimated an NNT of between 10 and 20 for a meaningful change in management. (Obviously, since most treatments aren't perfect, and most natural outcomes are necessarily bad, this would result in a higher NNT in terms of outcomes.) As most patients who come to the ED don't actually need medical care, it becomes nearly impossible to drive this number much lower.

A second, different but related, question arises as to the effect of this staffing decision on resource utilization and department flow. I'd put this NNT much lower, probably on the order of 4-5. All these unnecessary CTs, cbc/bmps, etc end up having a real effect on dept flow (at least in real EDs--the effect is typically lower in CAHs where >5 pts in the ER at once is rare).

If we limited this discussion to patients' with actual time-dependant medical conditions, I would put the estimate at between 2 and 3. While this estimate is anecdotal, based on my experience, I think it's at least as scientific as the study referenced. Even if the OP is competent at managing the sick or subtle patient, most midlevels clearly are not.

 

[turkeyjerky]

Agreed... I had a med student rotating with me not long ago who was so gung ho interested in EM and anytime I would stop to caution him about the state of the specialty and future challenges he/she might encounter, it's like their eyes glazed over and they got this silly grin on their face and said something to the effect "oh gosh...I'm sure it will be fine! These things always work out! I mean...we wouldn't have residency spots open if the jobs weren't there...right?!". I just inwardly rolled my eyes.

Yeah, and it doesn't help that the average PD is telling them all that this is a momentary blip and will pass. It doesn't matter if students are 'in love' with Emergency Medicine. The way things are going, they're just as likely to get stuck working in an urgent care as they are to get to work in an ED.

 

[Boatswain2PA]

All these unnecessary CTs, cbc/bmps,

In this study there were no differences between tests ordered by EPs and the PAs.

 

[turkeyjerky]

In this study there were no differences between tests ordered by EPs and the PAs.

Yeah. In a tiny ER in f’ing Sedona.

 

[skougess]

Yeah. In a tiny ER in f’ing Sedona.

I’ve always said NPs/PAs greatly benefit from the fact that humans are just so damn resilient. You could have a pre-med manage an ED and 90% of those patients would be just fine. Even worse in peds. That’s the problem with studies on this topic.

 

[Boatswain2PA]

This is arguably a better study than the one that showed MLPs did much more testing than physicians. It is only PAs and docs, and it is exact same patient population. One downside is it didn't differentiate between EP and other specialty docs.

FWIW - I think supervised MLPs do over test. I know I do, and one driver of that is because I have a lower risk acceptance when I know someone else is on the hook for my decisions (if that makes sense).

 

[aim-agm]

@Boatswain2PA to chime in with a psych perspective, the question regarding NNT is a potentially interesting one and did indeed lead to a useful discussion. There is little to remark on.

On the other hand, the inclusion of the study is curious. It is obviously being used to support your near-explicit opinion that there is less difference between PAs and EPs than EPs here believe. Holding this opinion is unremarkable given your degree and the incentives and history it implies. What is remarkable is that the study is obviously garbage but you included it to support your position. There are several potential explanations:

-The most benign is that the finding was ego syntonic so you didn't think too much about it and accepted the conclusion. This happens to all of us from time to time. Your resistance to accepting that the study was garbage is concerning, and in this explanation would suggest the need to defend your belief is getting in the way of making it truer (i.e. improvement).

- The other likely explanation is that you were aware the study was garbage but included it anyway in bad faith (e.g. trolling). Your resistance would then reflect commitment to this purpose.

-The other reasonable explanation is that you did not recognize that you did not recognize the study was garbage because you couldn't (e.g. inadequate intellectual capacity or training) or did not invest even minimal time to analyze it (i.e. laziness). Your resistance is less congruent with this explanation.

Recommendations to SDN ED:
Dx: Adjustment disorder, r/o with disturbance of conduct
Recs:
-Not imminent danger to self, likely not imminent danger to others as long as they are ESI 3-5. Does not require inpatient psychiatric hospitalization or psych 1:1
-Outpatient/outforum follow up and resources
-Advise against further engagement while pattern of behavior continues

 

[robellis]

@Boatswain2PA to chime in with a psych perspective, the question regarding NNT is a potentially interesting one and did indeed lead to a useful discussion. There is little to remark on.

On the other hand, the inclusion of the study is curious. It is obviously being used to support your near-explicit opinion that there is less difference between PAs and EPs than EPs here believe. Holding this opinion is unremarkable given your degree and the incentives and history it implies. What is remarkable is that the study is obviously garbage but you included it to support your position. There are several potential explanations:

-The most benign is that the finding was ego syntonic so you didn't think too much about it and accepted the conclusion. This happens to all of us from time to time. Your resistance to accepting that the study was garbage is concerning, and in this explanation would suggest the need to defend your belief is getting in the way of making it truer (i.e. improvement).

- The other likely explanation is that you were aware the study was garbage but included it anyway in bad faith (e.g. trolling). Your resistance would then reflect commitment to this purpose.

-The other reasonable explanation is that you did not recognize that you did not recognize the study was garbage because you couldn't (e.g. inadequate intellectual capacity or training) or did not invest even minimal time to analyze it (i.e. laziness). Your resistance is less congruent with this explanation.

Recommendations to SDN ED:
Dx: Adjustment disorder, r/o with disturbance of conduct
Recs:
-Not imminent danger to self, likely not imminent danger to others as long as they are ESI 3-5. Does not require inpatient psychiatric hospitalization or psych 1:1
-Outpatient/outforum follow up and resources
-Advise against further engagement while pattern of behavior continues

LOL

 

[Mount Asclepius]

What is the NNT for an EM physician to have a better outcome…

1

 

[deleted6669]

This is arguably a better study than the one that showed MLPs did much more testing than physicians. It is only PAs and docs, and it is exact same patient population. One downside is it didn't differentiate between EP and other specialty docs.

FWIW - I think supervised MLPs do over test. I know I do, and one driver of that is because I have a lower risk acceptance when I know someone else is on the hook for my decisions (if that makes sense).

I would have liked them to break out the data for EM boarded physician vs FP boarded physician vs EMPA. I have no doubt the EM physicians would have the best scores, but I am willing to bet that experienced EMPAs would have had better scores than FP boarded physicians. EM IS a different mindset and it is hard to go from working in clinic 20 days/month with one mindset to an EM mindset for 2-3 shifts/month. I work at several rural facilities that are recruiting EM docs> EMPAs> FP MDs. No facilities I work at use NPs for solo coverage.

 

[deleted6669]

Does anyone practice where ED APPs see the same patients as ABEM docs? We have great APPs but none of them have any interest in critical patients, many don't like seeing high-risk chest pain or a simple CVA. None of them do critical care procedures; none of them intubate, place CVLs, do LPs or ortho reductions other than simple shoulder. This is throughout a large healthcare system.

There are something like 250 rural EDs nationwide with solo coverage provided either by a doc or a PA. I work at several of these. I see every pt and do every procedure during my shifts, just like the docs do on theirs. I am the only clinician in the hospital most of the day (aside from the 4 hrs/day the fp hospitalists round on inpatients) , so I am also responsible for the disasters on the floor. Not many PAs practice in this fashion obviously. The last time I checked an EMPA census it was around 2.5% of EMPAs. Most of us were previously paramedics and have > 15 years of progressive experience working at higher volume facilities. I will be the first to agree that the gold standard staffing in the ED is an EM residency trained and boarded EM physician. I would argue that the second choice is not an FP doc working occasional EM shifts, but an experienced EMPA. EM docs are not my competition, FP docs are and I can tell you that most small, rural facilities would rather have me than an FP doc half my age with 5% of my EM experience. . If any of you EM guys/gals want to come out to the rural environment and work 24 hr shifts 2+ hrs from a major metro area seeing 12-24 pts/shift for about 1/2 what you currently make you are welcome to join me.

 

[skougess]

There are something like 250 rural EDs nationwide with solo coverage provided either by a doc or a PA. I work at several of these. I see every pt and do every procedure during my shifts, just like the docs do on theirs. I am the only clinician in the hospital most of the day (aside from the 4 hrs/day the fp hospitalists round on inpatients) , so I am also responsible for the disasters on the floor. Not many PAs practice in this fashion obviously. The last time I checked an EMPA census it was around 2.5% of EMPAs. Most of us were previously paramedics and have > 15 years of progressive experience working at higher volume facilities. I will be the first to agree that the gold standard staffing in the ED is an EM residency trained and boarded EM physician. I would argue that the second choice is not an FP doc working occasional EM shifts, but an experienced EMPA. EM docs are not my competition, FP docs are and I can tell you that most small, rural facilities would rather have me than an FP doc half my age with 5% of my EM experience. . If any of you EM guys/gals want to come out to the rural environment and work 24 hr shifts 2+ hrs from a major metro area seeing 12-24 pts/shift for about 1/2 what you currently make you are welcome to join me.

To your first point: We know, and it’s terrifying. I’ve told my rural family members to never stop at a critical access ED under any circumstances. And yes that includes when they feel they have a life-threatening emergency. I tell them to take their chances with the drive. This whole “it’s better than nothing, right?” Is patently false. Mismanagement is worse than nothing.

To your second point: Don’t worry, we’ll be knocking at your ED doors with resumes VERY soon.

 

[Boatswain2PA]

Anyone wonder why we see the same psych patients all the time? We admit them to psych, who keeps them inpatient for exactly the number of days their insurance says they qualify for, then are discharged....so we can see them in the ED again.

Worse yet, psych is almost totally owned by a bunch of NPs with 500 hours of clinical training.