DNPs will eventually have unlimited SOP

[SpecterGT260]

Post-grad, I'm not sure. However, the last rotation I was on I was with a DNP student. She had "clinic" one day a week every week versus the PA students that were there everyday (much to their dismay).

post graduate training is optional for both disciplines. They CAN do some if they want to "specialize". But really after they have graduated they can practice wherever they can get hired.

---

Members don't see this ad.

 

[Lokhtar]

I'm hearing a lot of assertions and anecdotes. If I was in the legislature, I wouldn't be convinced. The plural of anecdote is not data.

If the quality of care is inferior, it must be proven. I think it's a bit crazy to hope for some sort of mass awakening of the public once DNPs have more scope - people (including people here) can't even agree on basic things. People's opinions (including mine and yours) aren't necessarily rational - they are informed heavily by their beliefs, politics, and a whole slew of other factors that determine which facts get in and which don't.

 

[peteB]

I'm hearing a lot of assertions and anecdotes. If I was in the legislature, I wouldn't be convinced. The plural of anecdote is not data.

If the quality of care is inferior, it must be proven. I think it's a bit crazy to hope for some sort of mass awakening of the public once DNPs have more scope - people (including people here) can't even agree on basic things. People's opinions (including mine and yours) aren't necessarily rational - they are informed heavily by their beliefs, politics, and a whole slew of other factors that determine which facts get in and which don't.

The problem is that a lot of people will need to be harmed in order to prove this. Is this something that we want to do?

Also, if the data proves that a DNP is equivalent to an MD as a primary care provider (as many DNPs claim), why don't we just pick one of the two for primary care, instead of having both? Either reduce the requirements of the MD for primary care, or get rid of MD primary care all together and have the DNPs handle it. Why do we need both? It's just confusing to the public

 

[45408]

3. Oh and my favorite. Very healthy pt is admitted to the hospital by a NP for a very small cellulitis. Turns out the NP never even tried to I&D it...which is all it needed. This was benign enough for an urgent care let alone an ER... admitting him to hospital was beyond ridiculous.

You don't I&D cellulitis though. Cellulitis is treated with antibiotics. Abscesses need to be drained because there's no blood supply to bring antibiotics or immune system access to them.

 

[45408]

Yeah there aren't "scores of unemployed accountants" or "huge vacant grocery stores" TODAY, but you are extremely naive if you think that what you see today is going to remain the same forever.

Are you reading your own posts?

software and robotics have actually impacted all of these in major ways. Accounting software nowadays does the majority of work that accountants used to do by hand. Retail has largely moved online over the past decade.

Accounting software does the majority of the work.
Retail has moved largely online.

Yeah there aren't "scores of unemployed accountants" or "huge vacant grocery stores" TODAY.

Is the incongruity just flying right past you?

This idea of medicine being an 'art' is simply romantic rhetoric. I've been in med school 3 years and what I've realized is that people love to make their career out to be bigger than it is.

In med school for three years, which probably means you've been doing clinical medicine for...3 months now?

 

[ArcGurren]

Could you give me an example?

Try managing people with multiple different forms of shock, especially when the management is so different for each. Brief example is a patient of mine who had combined septic shock and cardiogenic shock - we debated a bit about getting cardiology to put in an IABP vs. just treating his sepsis and improving that first b/c we were afraid of getting his lines infected. Ended up having the IABP put in anyway and dousing him with antibiotics. Patient lived and left the ICU about a week later.

So no it's not just an algorithm. You very often have to balance the risks/benefits of something and sometimes just grit your teeth and do it. I don't know what year you're in or what sort of training you've already had, but I'm a fourth year and I'm constantly amazed at the difficult decisions some doctors have to make which definitely can't come out of some textbook.

 

[link2swim06]

You don't I&D cellulitis though. Cellulitis is treated with antibiotics. Abscesses need to be drained because there's no blood supply to bring antibiotics or immune system access to them.

You are right, I meant to say a small cellulitis with an abscess. As a relatively new M3, I am obviously not the expert on abscesses but I have seen far worse I&Ded in the ED and released.

The situation seemed strange that a NP would be able have the pt bypass the ED and admit straight to the internal med service.

 

[ArcGurren]

You are right, I meant to say a small cellulitis with an abscess. As a relatively new M3, I am obviously not the expert on abscesses but I have seen far worse I&Ded in the ED and released.

The situation seemed strange that a NP would be able have the pt bypass the ED and admit straight to the internal med service.

Stupid BS admissions are unfortunately a part of life though in most fields of medicine, whether it's IM, surg, or peds.

 

[peteB]

Are you reading your own posts?


Accounting software does the majority of the work.
Retail has moved largely online.

Yeah there aren't "scores of unemployed accountants" or "huge vacant grocery stores" TODAY.

Is the incongruity just flying right past you?


In med school for three years, which probably means you've been doing clinical medicine for...3 months now?

You took those quotes out of context, but I'm not going to continue with this discussion because it isn't relevant to the topic of the thread.

 

[xenotype]

I'm hearing a lot of assertions and anecdotes. If I was in the legislature, I wouldn't be convinced. The plural of anecdote is not data.

If the quality of care is inferior, it must be proven. I think it's a bit crazy to hope for some sort of mass awakening of the public once DNPs have more scope - people (including people here) can't even agree on basic things. People's opinions (including mine and yours) aren't necessarily rational - they are informed heavily by their beliefs, politics, and a whole slew of other factors that determine which facts get in and which don't.

This is the problem. When the AAFP comes out with a press release that essentially says "blah blah blah but they have less training" it is meaningless to legislators. The midlevels have a few crappy overcited studies that really don't prove anything, but even a crappy study is better than a mountain of anecdotal evidence.

I think two approaches should be used. The Flexner report of a century ago scalded the entire medical profession for not having enough education and not incorporating basic science. Why are we in the same boat today? Read the Flexnor report wiki page and add in DNP here and there. Astonishing, isn't it?

Secondly, studies with real patients are very difficult to do. The difference between a midlevel and a physician is hard to measure. Why not use test scores to show the difference? Come up with a short test that stresses clinical skills, fill it with cases that are easy to screw up, and measure all the inputs of the NPs and MDs taking the test. Then you get hard, quantifiable data on needless tests and missed diagnoses.

There should also be a retrospective way to analyze referral rates between NPs and MDs.

 

[peteB]

Try managing people with multiple different forms of shock, especially when the management is so different for each. Brief example is a patient of mine who had combined septic shock and cardiogenic shock - we debated a bit about getting cardiology to put in an IABP vs. just treating his sepsis and improving that first b/c we were afraid of getting his lines infected. Ended up having the IABP put in anyway and dousing him with antibiotics. Patient lived and left the ICU about a week later.

So no it's not just an algorithm. You very often have to balance the risks/benefits of something and sometimes just grit your teeth and do it. I don't know what year you're in or what sort of training you've already had, but I'm a fourth year and I'm constantly amazed at the difficult decisions some doctors have to make which definitely can't come out of some textbook.

So you had choices, you weighed the risks and benefits, and you made a choice. This is a cost-benefit analysis that happened in your head. Cost-benefit analysis is not an "art". They're based on logic and evidence and quantifiable measures- a.k.a. science. Art is defined as the "expression of human creative skill and imagination". There is very little creative skill or imagination involved in selecting between predefined treatment options by analyzing the risks and benefits involved. And picking something because you have can't decide what to do is called guessing, not art.

The difficulty of a decision also does not make it any more "artistic" than a simple decision. The same mental framework is applied, which you outlined yourself.

 

[SpecterGT260]

So you had choices, you weighed the risks and benefits, and you made a choice. This is a cost-benefit analysis that happened in your head. Cost-benefit analysis is not an "art". They're based on logic and evidence and quantifiable measures- a.k.a. science. Art is defined as the "expression of human creative skill and imagination". There is very little creative skill or imagination involved in selecting between predefined treatment options by analyzing the risks and benefits involved. And picking something because you have can't decide what to do is called guessing, not art.

The difficulty of a decision also does not make it any more "artistic" than a simple decision. The same mental framework is applied, which you outlined yourself.

I fear you have somehow missed the point.....

 

[peteB]

I fear you have somehow missed the point.....

If you think I've missed the point, please elaborate on what I've missed. Medicine is an applied science, it's not an art. It is about following guidelines and standards of care. It's not about individual expression and imagination or creativity. Sometimes it's hard to figure out what to do when the case is complex, but that just makes it a complex problem, like a difficult math problem. The approach to solving these problems is scientific, not artistic.

 

[Evil_Abed]

So you had choices, you weighed the risks and benefits, and you made a choice. This is a cost-benefit analysis that happened in your head. Cost-benefit analysis is not an "art". They're based on logic and evidence and quantifiable measures- a.k.a. science. Art is defined as the "expression of human creative skill and imagination". There is very little creative skill or imagination involved in selecting between predefined treatment options by analyzing the risks and benefits involved. And picking something because you have can't decide what to do is called guessing, not art.

The difficulty of a decision also does not make it any more "artistic" than a simple decision. The same mental framework is applied, which you outlined yourself.

For any number of complex cases, an algorithm can be defined to handle that particular case. We just need to look at what exactly the case was and what was done step by step. After we map out millions of such cases programmatically, we can handle 99.99% of new cases by finding an algorithm that matches that case as closely as possible. A computer can easily store, access, and evaluate the usefulness of billions of algorithms within a couple of seconds. After all, there is only a fixed number of things you can do for a patient, based on the options you have available, and even if that number is in the trillions, it is still within the reach of a CPU's processing power. Then we add in the "AI" portion, which can reason its way around modifying the algorithms to fit with the already minor differences and boom- you can handle 100% of the cases.

Think about it, how many times are you going to treat a case that has never ever occurred in that same presentation anywhere in the entire world before? You obviously can't store all of the occurrences in your head, but a computer can.

Are you seriously a medical student? If so, you should know how much of a gray area exists in the practice of medicine. There isn't a study that answers every single medical question. Many of the studies currently in existence have significant enough flaws for you to be wary of them. How's your computer going to decide which are well-done studies that it can extrapolate from and which are not? How's your computer going to decide which symptoms the patient presents with are more important to focus on? Who's going to do the physical exam and interpret and input the results accurately into this computer? How is the computer going to deal with cases that don't present in a textbook manner? How is the computer going to console a patient who just received a terminal diagnosis? I have a hard time believing that AI is currently that far advanced. As others have mentioned, even a "simple" EKG reading by a computer is not trusted.

Plus, I thought you said you weren't going to post in this thread anymore? Why do you keep derailing this thread? Please don't even respond to this and just start another thread in allo or wherever to discuss your AI issues. Seriously.

 

[SpecterGT260]

If you think I've missed the point, please elaborate on what I've missed.

If you want to be so literal with the definition of "art", you are technically correct. Medical practice is not about creativity. No... nobody is using body fluids to create abstract finger paints or syncing the equipment to play clinical dubstep.

However, when someone says clinical medicine is "an art" they are referring to the intangibles that come along with experience and the ways that that experience impacts practice.

 

[ArcGurren]

If you want to be so literal with the definition of "art", you are technically correct. Medical practice is not about creativity. No... nobody is using body fluids to create abstract finger paints or syncing the equipment to play clinical dubstep.

However, when someone says clinical medicine is "an art" they are referring to the intangibles that come along with experience and the ways that that experience impacts practice.

Methinks this guy's manner of thinking is very literal in general and speaks of huge amounts of inexperience. He also talks about studies as if they have absolute literal yes/no answers or findings to any clinical question, which is FAR from the truth. Even the best clinical studies I can think of (the Framingham study comes to mind) doesn't take into account so, so many different variables. Multiple studies are contradictory on top of that

 

[peteB]

Are you seriously a medical student? If so, you should know how much of a gray area exists in the practice of medicine. There isn't a study that answers every single medical question. Many of the studies currently in existence have significant enough flaws for you to be wary of them. How's your computer going to decide which are well-done studies that it can extrapolate from and which are not? How's your computer going to decide which symptoms the patient presents with are more important to focus on? Who's going to do the physical exam and interpret and input the results accurately into this computer? How is the computer going to deal with cases that don't present in a textbook manner? How is the computer going to console a patient who just received a terminal diagnosis? I have a hard time believing that AI is currently that far advanced. As others have mentioned, even a "simple" EKG reading by a computer is not trusted.
.

It's going to do all of those things the same way a human does them. Cost-benefit analysis, probability analysis, looking at how other similar cases have been handled, and in the lack of all else, guessing. As for studies, we have defined frameworks for analyzing them in an objective manner. There's nothing magical in the way human beings handle these issues. The EKG reading isn't trusted because the algorithms aren't yet fine tuned enough, not because a computer will never be able to read an EKG. It's the same reason why an M1 can't read an EKG very well, he/she is not trained enough. Fortunately the technology behind those kinds of things have recently started to make leaps and bounds.

I actually deleted my previous post, by the way, before I realized you had responded to it, because I'm done with this argument. You're actually the one who derailed the thread further by responding to that post.

 

[peteB]

Methinks this guy's manner of thinking is very literal in general and speaks of huge amounts of inexperience. He also talks about studies as if they have absolute literal yes/no answers or findings to any clinical question, which is FAR from the truth. Even the best clinical studies I can think of (the Framingham study comes to mind) doesn't take into account so, so many different variables. Multiple studies are contradictory on top of that

So basically what you're saying is that clinical decisions are just guesswork based on some shaky data and personal anecdotes? I'd rather have a doctor who works off of whatever shred of real evidence he has rather than trying to be an "artist" and pull ideas from his imagination. The studies are not perfect, obviously, and many are contradictory, but they offer scientifically tested information that should be the basis of the risk/benefit analysis used to determine treatments.

I never said studies have yes/no answers. All I said, which you agreed with, is that medical decisions are based on risk/benefit analysis, and there is no art to that. Unless you consider the assigning of values to risks and benefits art, in which case everything in the world is art. I call that science.

 

[ArcGurren]

So basically what you're saying is that clinical decisions are just guesswork based on some shaky data and personal anecdotes?

Some studies are stronger than others. Most studies don't describe everything. And sometimes studies contradict each other. You really have an issue with extremes if you broaden that to "shaky data and personal anecdotes". What part of medical school are you in right now?

 

[peteB]

Some studies are stronger than others. Most studies don't describe everything. And sometimes studies contradict each other. You really have an issue with extremes if you broaden that to "shaky data and personal anecdotes". What part of medical school are you in right now?

I'm halfway through 3rd year.

I don't know what you're trying to say. I said medicine is not an art, it is a science. There is no study in all of the world of science that is perfect and describes everything and isn't contradicted. That's how science is. You have to reason and fill in the gaps using your own brain. You have to assign values for the costs and benefits for decisions and then make them. But that's not art. That's still science. You're still using scientific principles to evaluate those studies and figure out what you're going to do.

The definition of art involves imagination and creativity, both of which don't apply to using observations and scientific data and understanding of biological mechanisms to decide which treatment strategy to use. Are musicians and painters using cost-benefit analysis based on observations and data from studies to create their works??

 

[ArcGurren]

I'm halfway through 3rd year.

I don't know what you're trying to say. I said medicine is not an art, it is a science. There is no study in all of the world of science that is perfect and describes everything and isn't contradicted. That's how science is. You have to reason and fill in the gaps using your own brain. But that's not art. That's still science. You're still using scientific principles to evaluate those studies and figure out what you're going to do.

Oh ffs you're one of those people who can't be swayed from using LITERAL definitions of something. Nobody is saying that we make stuff up out of thin air

 

[SpecterGT260]

The definition of art involves imagination and creativity, both of which don't apply to using observations and scientific data and understanding of biological mechanisms to decide which treatment strategy to use. Are musicians and painters using cost-benefit analysis based on observations and data from studies to create their works??

Are you saying this cannot happen?

Art has a pretty broad definition. There are certainly aspects of clinical practice that are more art than science. If you can't name any, I suspect you are having a difficult 3rd year

 

[peteB]

Oh ffs you're one of those people who can't be swayed from using LITERAL definitions of something. Nobody is saying that we make stuff up out of thin air

Exactly my point. That's what artists do. We don't do that. That's why we are not artists.

I think literal definitions are important. How can you use a word if everyone has a different definition of it? Language would be far better if we tried as much as we could to stick to concrete definitions of words. There would be a lot fewer miscommunications. Just ask a lawyer.

 

[peteB]

Are you saying this cannot happen?

Art has a pretty broad definition. There are certainly aspects of clinical practice that are more art than science. If you can't name any, I suspect you are having a difficult 3rd year

I don't think this is an argument that will ever end because it depends on how you define art. I don't define it as broadly as you do, clearly.

 

[SpecterGT260]

Exactly my point.

I think literal definitions are important. How can you use a word if everyone has a different definition of it? Language would be far better if we tried as much as we could to stick to concrete definitions of words. There would be a lot fewer miscommunications. Just ask a lawyer.

this is probably true

however you are really only demonstrating either an inability to follow context or a panache for picking a fight over a non issue knowing full well what the poster intended. Neither of these are good things

 

[xenotype]

Can we get back on topic to DNPs/midlevels?

 

[ArcGurren]

Exactly my point. That's what artists do. We don't do that. That's why we are not artists.

I think literal definitions are important. How can you use a word if everyone has a different definition of it? Language would be far better if we tried as much as we could to stick to concrete definitions of words. There would be a lot fewer miscommunications. Just ask a lawyer.

You know what's more annoying than not using literal definitions? Losing your **** over it like you have been

Anyway forget it. I can only tolerate this stuff for so long.

 

[peteB]

this is probably true

however you are really only demonstrating either an inability to follow context or a panache for picking a fight over a non issue knowing full well what the poster intended. Neither of these are good things

Actually the poster was using the definition of art too broadly to argue that a machine could not handle the 'artistic' portions of clinical practice. My argument was intended to show that his definition of art extends into a region that machines are capable of handling. But I don't want to continue this anymore.

 

[peteB]

You know what's more annoying than not using literal definitions? Losing your **** over it like you have been

Losing my ****? I'm just having a discussion on a forum. How's that losing my ****?

 

[SpecterGT260]

I don't think this is an argument that will ever end because it depends on how you define art. Can you name some of these aspects of clinical medicine you consider to be 'art'?

how about coaxing a non-compliant patient into following through with treatment? Interviewing an emotional or hostile patient? Or such cliches as breaking bad news? Are these science? because they are definitely a part of clinical practice.

And.... if we go look at the webster definition of art:

art/ärt/
Noun:
The expression or application of human creative skill and imagination, typically in a visual form such as painting or sculpture,...: "the art of the Renaissance"
Works produced by such skill and imagination.

Id say that nearly everything we use in the clinic, from diagnostic tools to medications, are the product of "creative skill and imagination". "Typically" does not limit the term only to things produced only for sensory appeal, even though this is how the term is generally used. If we want to be strictly literal as you suggest is an imperative, you just plain cannot NOT call medicine an art Doing so would artificially restrict the definition of the term to your own common usage and make you a hypocrite

 

[peteB]

Anyway forget it. I can only tolerate this stuff for so long.

Yeah most people who are losing arguments feel that way.

(that was a joke, don't jump on my ass, please)

 

[peteB]

how about coaxing a non-compliant patient into following through with treatment? Interviewing an emotional or hostile patient? Or such cliches as breaking bad news? Are these science? because they are definitely a part of clinical practice.

And.... if we go look at the webster definition of art:


Id say that nearly everything we use in the clinic, from diagnostic tools to medications, are the product of "creative skill and imagination".

All the questions at the beginning of your response are not specific to the practice of medicine by physicians. I am talking about the aspects of clinical practice that are specific to physicians. Dealing with difficult situations, breaking bad news, and handling difficult people are problems that just about all humans face in their lives. Many fields have to deal with problems like these more often than physicians do. They're not unique to medicine.

As for the second part- yes, creating new diagnostic tools and medications is. But the average clinician isn't doing that. He/she is simply applying things that have already been created. And as far as I know, most of those inventions are done by researchers and engineers, not clinical doctors. How much of clinical practice involves developing new diagnostic tools or medications? Usually zero.

 

[SpecterGT260]

Can we get back on topic to DNPs/midlevels?

If you have something interesting to contribute on that subject go for it, scooter

Actually the poster was using the definition of art too broadly to argue that a machine could not handle the 'artistic' portions of clinical practice. My argument was intended to show that his definition of art extends into a region that machines are capable of handling. But I don't want to continue this anymore.

A machine will only ever be capable of protocol driven medicine. Sure, enough inputs will allow a machine a probability chance at a correct diagnosis with that chance increasing with increasing inputs of information.

How do you think this would work IRL? You will still need someone to do the examinations. Personally, I think working up a patient in this way (feeding the machine information, it spits out instructions, gathering requested info, refeeding, respitting, regathering, until we hit some arbitrary level of confidence in the diagnosis) will be cumbersome and very inefficient. There is also the fact that the machine is only as good as the database it pulls from and such things are not fool proof. I don't think database driven medicine will be happening any time soon, and certainly not to an extent that PCPs risk losing work.

 

[SpecterGT260]

All the questions at the beginning of your response are not specific to the practice of medicine by physicians. I am talking about the aspects of clinical practice that are specific to physicians. Dealing with difficult situations, breaking bad news, and handling difficult people are problems that just about all humans face in their lives. Many fields have to deal with problems like these more often than physicians do. They're not unique to medicine.

As for the second part- yes, creating new diagnostic tools and medications is. But the average clinician isn't doing that. He/she is simply applying things that have already been created. And as far as I know, most of those inventions are done by researchers and engineers, not clinical doctors. How much of clinical practice involves developing new diagnostic tools or medications? Usually zero.

see, you are artificially restricting the definition again. "Application of".

Per the literal definition which you, again, suggest is of utmost importance: any time you press a stethoscope up to a surface you are applying previously established creative works the definition doesnt say it has to be novel

And yes, the issues I brought up are not unique to medicine. But again... if we are being purely literal, your statement did not restrict to things only unique to physicians. You said medicine isn't an art, it is a science. Clearly there are non-scientific aspects to medicine.

 

[xenotype]

If you have something interesting to contribute on that subject go for it, scooter

I did. It was ignored over this stupid rise of the machines thread derail.

This is the problem. When the AAFP comes out with a press release that essentially says "blah blah blah but they have less training" it is meaningless to legislators. The midlevels have a few crappy overcited studies that really don't prove anything, but even a crappy study is better than a mountain of anecdotal evidence.

I think two approaches should be used. The Flexner report of a century ago scalded the entire medical profession for not having enough education and not incorporating basic science. Why are we in the same boat today? Read the Flexnor report wiki page and add in DNP here and there. Astonishing, isn't it?

Secondly, studies with real patients are very difficult to do. The difference between a midlevel and a physician is hard to measure. Why not use test scores to show the difference? Come up with a short test that stresses clinical skills, fill it with cases that are easy to screw up, and measure all the inputs of the NPs and MDs taking the test. Then you get hard, quantifiable data on needless tests and missed diagnoses.

There should also be a retrospective way to analyze referral rates between NPs and MDs.

 

[peteB]

If you have something interesting to contribute on that subject go for it, scooter




A machine will only ever be capable of protocol driven medicine. Sure, enough inputs will allow a machine a probability chance at a correct diagnosis with that chance increasing with increasing inputs of information.

How do you think this would work IRL? You will still need someone to do the examinations. Personally, I think working up a patient in this way (feeding the machine information, it spits out instructions, gathering requested info, refeeding, respitting, regathering, until we hit some arbitrary level of confidence in the diagnosis) will be cumbersome and very inefficient. There is also the fact that the machine is only as good as the database it pulls from and such things are not fool proof. I don't think database driven medicine will be happening any time soon, and certainly not to an extent that PCPs risk losing work.

I agree that it will be inefficient, but it'll be cheaper and easier to make and provision than more physicians. And in the more distant future, it will be tuned to become far more efficient with natural language capabilities and advanced sensors that can replace a lot of the human-in-the-middle activity that needs to happen.

No, PCPs are not going to risk losing work for a very long time. They'll probably lose their salaries bit by bit, but they'll still have jobs.

 

[mmmcdowe]

"The curricula for both degrees also vary. Just as an example, medical students learn anatomy, biochemistry, physiology, pharmacology, psychology, microbiology, and pathology, among numerous other courses focusing on treatment and prevention of a wide array of diseases.
​

Some of the courses offered by various DNP programs (determined by each educational institution) include Evidence-Based Practice and Nursing Systems, Health Policy Development & Implementation, Ethics and Public Policy in Healthcare Delivery, and Global Health & Social Justice."


from - http://studentdoctor.net/2011/04/sdn-reports-the-dnp-degree/

​

The problem with this data is that it is a double edged sword. Without outcomes data justifying our side of it, this data actually supports the NPs.

 

[peteB]

see, you are artificially restricting the definition again. "Application of".

Per the literal definition which you, again, suggest is of utmost importance: any time you press a stethoscope up to a surface you are applying previously established creative works the definition doesnt say it has to be novel

And yes, the issues I brought up are not unique to medicine. But again... if we are being purely literal, your statement did not restrict to things only unique to physicians. You said medicine isn't an art, it is a science. Clearly there are non-scientific aspects to medicine.

OK, fine.

 

[peteB]

The problem with this data is that it is a double edged sword. Without outcomes data justifying our side of it, this data actually supports the NPs.

How does it support the NPs without any outcome data? Technically, it is meaningless without data. It doesn't support or oppose them.

But common knowledge would suggest that more training and experience is better.

 

[SpecterGT260]

This is the problem. When the AAFP comes out with a press release that essentially says "blah blah blah but they have less training" it is meaningless to legislators. The midlevels have a few crappy overcited studies that really don't prove anything, but even a crappy study is better than a mountain of anecdotal evidence.

I think two approaches should be used. The Flexner report of a century ago scalded the entire medical profession for not having enough education and not incorporating basic science. Why are we in the same boat today? Read the Flexnor report wiki page and add in DNP here and there. Astonishing, isn't it?

Secondly, studies with real patients are very difficult to do. The difference between a midlevel and a physician is hard to measure. Why not use test scores to show the difference? Come up with a short test that stresses clinical skills, fill it with cases that are easy to screw up, and measure all the inputs of the NPs and MDs taking the test. Then you get hard, quantifiable data on needless tests and missed diagnoses.

There should also be a retrospective way to analyze referral rates between NPs and MDs.

Mathematically, if you could correlate step scores with outcomes you could also correlate that back to DNPs. Last figure I saw had fewer than half of them able to pass USMLE.

Here is what the question boils down to: how much patient death is acceptable? This applies both to DNP and to Dr. Terminator.

If we go strictly protocol and stats based, the stats say that some given % of the time we will get it wrong. Is it ok to outright kill some small % of our patients? The right move, in my mind, is to remind people that they wan't the best when it comes to their health, not the cheapest. We need to combat the notion that people are entitled to steak at the quarter pounder price.

 

[xenotype]

Mathematically, if you could correlate step scores with outcomes you could also correlate that back to DNPs. Last figure I saw had fewer than half of them able to pass USMLE.

Here is what the question boils down to: how much patient death is acceptable? This applies both to DNP and to Dr. Terminator.

If we go strictly protocol and stats based, the stats say that some given % of the time we will get it wrong. Is it ok to outright kill some small % of our patients? The right move, in my mind, is to remind people that they wan't the best when it comes to their health, not the cheapest. We need to combat the notion that people are entitled to steak at the quarter pounder price.

A counter argument is that when you measure outcomes for surgeons, everyone will want the #1 surgeon. They cannot possibly have the #1 surgeon, and some will be forced to go to the worst surgeon. Economic scarcity has to play in at some point.

 

[SpecterGT260]

I agree that it will be inefficient, but it'll be cheaper and easier to make and provision than more physicians. And in the more distant future, it will be tuned to become far more efficient with natural language capabilities and advanced sensors that can replace a lot of the human-in-the-middle activity that needs to happen.

No, PCPs are not going to risk losing work for a very long time. They'll probably lose their salaries bit by bit, but they'll still have jobs.

I disagree. you are now talking about theoretical sensors that only exist in sci-fi. The only realistic form of this is just a database with inputs and it spits out a DDx with % probability. The rest is pure speculation on things that could maybe exist one day maybe possibly. Since the world is set to explode in exactly 2 months and a day anyways.... I don't think it will be an issue. I would also guess that such an intricate machine would be oober expensive and require special training to keep calibrated and working correctly.

OK, fine.

That sounded bitter

How does it support the NPs without any outcome data? Technically, it is meaningless without data. It doesn't support or oppose them.

But common knowledge would suggest that more training and experience is better.

because NPs are cheaper. So if outcomes are the same, its better to go with the cheaper option.

Yes, common knowledge and sense would tell you that more training = better. The NP papers discussing outcomes include only bread and butter cases. In such an analysis it is impossible to tease out whether one is better than the other.

The way I have described it in other threads is like this:

Lets say we have a doctor and a DNP each seeing patients. The doctor has X number of negative outcomes. Some due to error, some due to unavoidability of the outcome. Let's say this rate is 1% +/- 3% (the doc works with high risk cases).

Now, our NP.... bitter over god knows what, decides that for every 30th patient, s/he is going to walk into the room and shoot the patient in the face with a bazooka.

These two people have equal outcomes per this analysis with absolutely no regards for reasons for the outcomes. If you look at the total patient populaiton, especially in primary care, most people are either not going to die or experience a negative outcome regardless of what happens. If these people are in the analysis, it doesnt matter what is looked at, you will never show a statistically significant difference between the groups. It is mathematically impossible. The appropriate study would look at high risk cases, because nobody is asking if they are proficient enough to treat someone who probably only needs to go home and rest and take a broad spectrum antibiotic just for funsies.

 

[SpecterGT260]

A counter argument is that when you measure outcomes for surgeons, everyone will want the #1 surgeon. They cannot possibly have the #1 surgeon, and some will be forced to go to the worst surgeon. Economic scarcity has to play in at some point.

that is exactly what this discussion is. That is not a counter argument.

As a community, we are talking about where to set the bar for "minimally competent". The "worst surgeon" still surpasses this bar.

However, mechanistically, if we are talking about gaps in education, the surgeon should still have a % chance to identify something complex correctly where as someone with limited education will have 0%chance based on limited exposure.

That is overly simplistic, but it is really what is at the heart of the matter.

 

[SpecterGT260]

Here is maybe a better way to think about the studies than the one I gave a minute ago:

Lets say we have a patient population with ACCX (Acute complicated case X)

Now, lets also assume that physicians have a 99% accuracy rating with identifying and treating. Therefore only 1/100 patients with this disease will suffer a bad outcome.

Now, lets also assume that DNPs choose to shoot every patient with these symptoms in the face with bazookas. Every one of them dies (except for those with SNPs which convey bazooka resistance).

Now, in a patient outcome analysis, if ACCX is only encountered 1% of the time in the clinic, the outcomes between DNPs and doctors is still statistically insignificant, because we are looking at two %ages in the 99% range. Even though doctors are 99.9% vs 99%, if there is any variability here, the standard errors will likely overlap (nobody go all math geek on that one... If the math doesn't work just sub in numbers that do....) and the AANP can publish a paper saying they are just as good as doctors, even though this patient population will always die if treated by a DNP. This is why the studies are so inappropriate for the conclusions they try to push.

 

[xenotype]

Here is maybe a better way to think about the studies than the one I gave a minute ago:

Lets say we have a patient population with ACCX (Acute complicated case X)

Now, lets also assume that physicians have a 99% accuracy rating with identifying and treating. Therefore only 1/100 patients with this disease will suffer a bad outcome.

Now, lets also assume that DNPs choose to shoot every patient with these symptoms in the face with bazookas. Every one of them dies (except for those with SNPs which convey bazooka resistance).

Now, in a patient outcome analysis, if ACCX is only encountered 1% of the time in the clinic, the outcomes between DNPs and doctors is still statistically insignificant, because we are looking at two %ages in the 99% range. Even though doctors are 99.9% vs 99%, if there is any variability here, the standard errors will likely overlap (nobody go all math geek on that one... If the math doesn't work just sub in numbers that do....) and the AANP can publish a paper saying they are just as good as doctors, even though this patient population will always die if treated by a DNP. This is why the studies are so inappropriate for the conclusions they try to push.

Excellent points. So the focus really needs to be on what minimum competency is considered to be.

 

[45408]

For any number of complex cases, an algorithm can be defined to handle that particular case. We just need to look at what exactly the case was and what was done step by step. After we map out millions of such cases programmatically, we can handle 99.99% of new cases by finding an algorithm that matches that case as closely as possible. A computer can easily store, access, and evaluate the usefulness of billions of algorithms within a couple of seconds. After all, there is only a fixed number of things you can do for a patient, based on the options you have available, and even if that number is in the trillions, it is still within the reach of a CPU's processing power. Then we add in the "AI" portion, which can reason its way around modifying the algorithms to fit with the already minor differences and boom- you can handle 100% of the cases.

Think about it, how many times are you going to treat a case that has never ever occurred in that same presentation anywhere in the entire world before? You obviously can't store all of the occurrences in your head, but a computer can.

Prepare to be surprised. It really doesn't take much. In the past month, I operated on one kid that has a disease that's only been documented in 200 people, and last week I operated on a kid with a malignancy diagnosed 20 times/year in North America. Throw in one complicating condition, and you're suddenly actually in virgin territory. I'm not even at a large pediatric center by any means. Even if we had every case documented, an n=200 doesn't exactly give you the power to make an algorithm that will be unilaterally applicable.

Your algorithms will still only give you the probability of something happening. Most elderly people tolerate post-operative narcotics just fine, and some of them go wildly delirious, with no detectable difference in their pre-operative condition. The computer is only going to be able to give you a probability of a bad outcome based on a course of action.

 

[cbrons]

The problem with this data is that it is a double edged sword. Without outcomes data justifying our side of it, this data actually supports the NPs.

How does it support the NPs without any outcome data? Technically, it is meaningless without data. It doesn't support or oppose them.

But common knowledge would suggest that more training and experience is better.

If the gaps in training between the 2 make absolutely no difference in practice, then it is time for the fluff, fat, and the garbage to be trimmed from the curricula at medical schools. If someone who takes sociology classes coupled with a few clinical classes and on-the-job training can become as proficient and knowledgeable as an attending, then we have a lot of stuff we need to cut out of basic sciences and probably clinical curriculum.

Mathematically, if you could correlate step scores with outcomes you could also correlate that back to DNPs. Last figure I saw had fewer than half of them able to pass USMLE.

.

As far as I understand it, less than half of them were able to pass the USMLE Step III (not all Steps). They just took Step III, and not even the Step III we take, but a watered down version thereof. I haven't taken Step 3 and I could be wrong on this, but I hear it is more or less a scam when you take into consideration how easy it is relative to how much it costs (that is, for American medical school graduates).

And if half of them still can't pass a watered down version of the easiest section of the exam that is supposed to determine whether or not new physicians are capable enough to practice medicine autonomously then either a.) the exam is worthless and expensive scam or b.) NPs are far too dangerous to be practicing on their own, and in fact they should receive more (not less) supervision.

 

[Bacchus]

If you want to be so literal with the definition of "art", you are technically correct. Medical practice is not about creativity. No... nobody is using body fluids to create abstract finger paints or syncing the equipment to play clinical dubstep.

However, when someone says clinical medicine is "an art" they are referring to the intangibles that come along with experience and the ways that that experience impacts practice.

 

[docnotsopc]

In med school for three years, which probably means you've been doing clinical medicine for...3 months now?

I fear you have somehow missed the point.....

We have a huge shortage of primary care health care professionals in the US. That shortage needs to be addressed and if that means giving DNPs the right to act as a primary health care provider.....than so what?!?! I know it wont affect me as a physician down the road.

PetB, your ego is getting in the way of this. People are entitled to health care and the gap in primary care needs to limited with or without DNPs. If they are medically trained, who cares?! Its FM where the training is minimal anyway

 

[peteB]

Your algorithms will still only give you the probability of something happening. Most elderly people tolerate post-operative narcotics just fine, and some of them go wildly delirious, with no detectable difference in their pre-operative condition. The computer is only going to be able to give you a probability of a bad outcome based on a course of action.

Humans work with probabilities too. No one is ever sure exactly what is going what to happen. What we do is look at the situation and figure out what is likely to happen in our head (the probabilities), and then we act on the most likely scenarios. The way we figure out probabilities in our head is usually based on our personal experiences or data we've seen- the same way a computer would figure it out.

Look, I really really want to end this argument so here's my final statement (for real this time). No one is going to replace humans completely any time soon. But computers can and will eventually be able to do a lot of the work humans do. So in the end we will need a lot fewer humans.

You were mocking me before, telling me that you don't see any grocery stores lying vacant..but if you actually look at the data of how many stores have closed you'll find that quite a few businesses have gone bankrupt because they can't compete with online retailers.

For example: http://www.reuters.com/article/2011/07/18/us-borders-liquidation-idUSTRE76H0BH20110718

"Borders was unable to overcome competition from larger rival Barnes & Noble Inc and from Amazon.com Inc, which began to dominate book retail when the industry shifted largely online. Borders, for which online sales represented only a small fraction of revenue, never caught up to its rivals' e-reader sales, namely Amazon's Kindle and Barnes & Noble's Nook."

Target is also seriously concerned. They just sent a letter out to their investors about how they're going to fight the practice of 'showrooming' - people coming into their stores to look at items and then buying them cheaper online: http://moneyland.time.com/2012/01/2...oom-for-the-stuff-you-buy-for-less-at-amazon/

As for another example, let's look at the postal service: "The Internet has arguably hurt the postal service more than any other business. The drastic reduction in mail delivery has cost thousands of jobs in delivery and caused many post office closures around the country. Now, mail sorting jobs are set to disappear as a new automated system is implemented. Between 2008 and 2018, more than 54,500 jobs, or about 30 percent of current positions, are expected to vanish."

"Word processing, voicemail and the internet all make it easier for skilled professionals to do clerical work themselves. In the U.S. alone, word processors and typists are slated to lose 13,200 jobs by 2020 and data entry clerks are expected to lose 15,900 according to the BLS."

Manufacturing assembly jobs: In the Unied Kingdom specifically, 400,000 manual jobs are forecast to disappear before 2020.

I could go on and on, but I'll stop there.

As you can see these are real consequences happening today.

So the bottom line is, computer technology is going to change the game of medicine. When will it happen? I don't know. But it will gradually happen.

The end.