Outcomes studies in states that "opt out?"

cchoukal · Mar 25, 2009

At my hospital, we have to complete an on-line fraud awareness program every year. I guess a few years ago, we got nipped for Medicare/Medicaid billing issues and this BS was part of the settlement. There was some anesthesiology-specific stuff in there, and one thing that caught my eye was the following:

"If a state decides to drop physician supervision for its Medicare-Medicaid population, HHS's Agency for Healthcare Research and Quality will conduct a prospective patient safety study to assess patient outcomes relating to the work of unsupervised anesthesia nurses in that state versus states where there is continued supervision by a physician."

Several states have opted out of physician supervision. Is this what they mean? Have these "prospective patient safety" studies been done?

Anyone know anything about this?

pgg · Mar 25, 2009

cchoukal said:
Anyone know anything about this?

No, but I'd be interested to see how they handle selection bias. Because you know the 20 yo knee scope will go to CRNAs while the septic 85 yo ICU-bound ex-lap / bowel resection will go to the anesthesiologists ...

Who the hell is going to give informed consent to get randomized to nurse vs doctor care?

I hereby make the bold prediction that these studies, if they're done at all, will be done poorly, will fail to show a difference, and will be seized by the AANA as proof of clinical equality regardless of what the actual data shows or fails to show.

guitarguy09 · Mar 25, 2009

pgg said:
No, but I'd be interested to see how they handle selection bias. Because you know the 20 yo knee scope will go to CRNAs while the septic 85 yo ICU-bound ex-lap / bowel resection will go to the anesthesiologists ...

Who the hell is going to give informed consent to get randomized to nurse vs doctor care?

I hereby make the bold prediction that these studies, if they're done at all, will be done poorly, will fail to show a difference, and will be seized by the AANA as proof of clinical equality regardless of what the actual data shows or fails to show.

👍

foxtrot · Mar 25, 2009

pgg said:
No, but I'd be interested to see how they handle selection bias. Because you know the 20 yo knee scope will go to CRNAs while the septic 85 yo ICU-bound ex-lap / bowel resection will go to the anesthesiologists ...

Who the hell is going to give informed consent to get randomized to nurse vs doctor care?

I hereby make the bold prediction that these studies, if they're done at all, will be done poorly, will fail to show a difference, and will be seized by the AANA as proof of clinical equality regardless of what the actual data shows or fails to show.

Yes, I definitely think we should start doing these types of studies. Can you imagine asking a patient to give informed consent to be randomized to an unsupervised nurse vs physician's care for their anesthestic? Man, that would go over really well with patients (not). 🙂

coprolalia · Mar 27, 2009

You'd have to do this retrospectively, and there are a couple of problems with such a proposed "outcome" study...

First, you have to have a fair case-match. In order to do that (and fit the data into a database to analyze), the biggest data challenge, among many, is that you have to agree on the acuity of the patient. A simple way to do this would be to look at comparable cases with the ASA classification, but there is too much subjectivity. A better way would be to look at individual co-morbidities and code them into the database using ICD-9CM or ICD-10 codes. However, this doesn't always catch the "severity" at presentation to the OR. If you can't compare apples to apples, it becomes very difficult to detect differences. And, this is probably the biggest difference among quality of care received.

Second, you'd have to consider "soft" (i.e., subjective) endpoints. For example, how much "pain" did the patient have in the PACU (which can be altered by several different factors such as whether or not the surgeon injected local, etc.)? What was the "cost" of the anesthetic (i.e., did the patient spend 1 hour in the PACU or 3, was a propofol infusion running the entire time or did they use only volatile, was there any PONV or were all of the anti-emetics at our disposal given, did other unforeseen treatments have to be utilized, etc.)? Did each of the procedures get accomplished on one attempt? Did the patient take two minutes or twenty minutes to wake up at the end of the case? Did the patient move the entire time? Did they bronchospasm? Did they have a prolonged intra-op hypotension? What was their 30-day outcome? You get the drift...

Lastly, you would have to derive very, VERY large numbers because, on the whole, most anesthetics given are very safe in the vast majority of people, especially in the outpatient setting where CRNA care tends to dominate.

Overall, I just don't think the current level of data we are gathering is either rich or granular enough to easily use as well as deduce subtle differences in anesthesia care - which I believe exist - between providers. And, fact is, most CRNAs who "go it alone" are not giving anesthesia in particular sick patients unsupervised.

So, nice idea, but just not practical until we get detailed, easy to use data from universally utilized, for example, electronic record keeping systems, and make a high commitment to following our patients closely in the immediate as well as prolonged peri-operative period... a herculean task that would provide almost insurmountable challenges in conducting such a study.

I will give you an anecdote, though...

I did an outpatient case today. The patient was a follow-up case, and had had the same procedure done this past November. He and his wife commented to me pre-op that it "took a long time to wake-up" after the past procedure. The procedure today, the exact same repeat one on the exact same body part for almost the exact same length of time, I performed the anesthetic for, whereas one of our more seasoned CRNAs had done in November. Today, less than an hour after the procedure was over, the patient was sitting up in the PACU, smiling, drinking a soda, eating crackers, and telling me how much better he felt compared to last time. That's a God's honest true story.

Now, I just don't know how you meaningfully quantify that.

-copro

proman · Mar 27, 2009

coprolalia said:
You'd have to do this retrospectively, and there are a couple of problems with such a proposed "outcome" study...

I did an outpatient case today. The patient was a follow-up case, and had had the same procedure done this past November. He and his wife commented to me pre-op that it "took a long time to wake-up" after the past procedure. The procedure today, the exact same repeat one on the exact same body part for almost the exact same length of time, I performed the anesthetic for, whereas one of our more seasoned CRNAs had done in November. Today, less than an hour after the procedure was over, the patient was sitting up in the PACU, smiling, drinking a soda, eating crackers, and telling me how much better he felt compared to last time. That's a God's honest true story.

Now, I just don't know how you meaningfully quantify that.

-copro

I've had patients say the same thing. What do you think you did differently (no Benadryl??)? I've mostly blown off statements like that.

As for numbers, anesthesia information management systems (AIMS) encompassing tens of thousands of anesthetics across multiple hospital systems is the way go. University of Michigan has >50,000 anesthetics in their database. Search for S Kheterpal in pubmed.

coprolalia · Mar 27, 2009

proman said:
I've had patients say the same thing. What do you think you did differently (no Benadryl??)? I've mostly blown off statements like that.

It's actually a fairly rare occurrence, in my estimation, where you have the opportunity to do almost the exact same case twice in a relatively short timeframe on the same patient (excluding patients who go back-and-forth for various washouts, dressing changes, etc.) where a full-blown anesthetic is given for more than a couple of hours.

I was actually concerned that the guy might have had some kind of reaction or had some aberrant GABA receptor (or something) based on how he and his wife had expressly stated how long it'd had taken him to "wake up" from the prior anesthetic. They live about an hour away by car, and the wife told me that he was goofy the whole way home the previous time... and well into that evening. She said he was saying crazy things, and wasn't making any sense even after discharge during the car ride (made me think later... did he have an anoxic hit?).

That's why I looked, in detail, at the previous anesthesia record. No Benadryl was given by the CRNA. In fact, I did almost the exact same anesthetic (midazolam in holding, LMA, desflurane, anti-emetics, no ketamine, etc.), except that I gave about twice as much fentanyl as she did.

So, I'm not sure what the difference was, at least as could be discerned on the previous record.

That's why I think it's hard to detect differences quantitatively. For example, I know this particular CRNA, though, and she's a bit of a "cowboy". I can't be sure that everything that was done was recorded (i.e., I walked into her room one time to give her a break, to which she refused... the patient was under a full TIVA GA, whether she realized it or not, without an airway satting at about 88% with a full jaw thrust... "do you need some help?" I asked to which I got, "nah, I've got it under control"... I just shook my head and left the room... hell, I'm just a resident... I don't know anything, right?... ask Noyac).

Without an electronic anesthesia record that can't be altered and a good, full pre-operative health history that codifies all co-morbidities, I think it's really hard to quantify the differences.

-copro

Noyac · Mar 27, 2009

coprolalia said:
I did an outpatient case today. The patient was a follow-up case, and had had the same procedure done this past November. He and his wife commented to me pre-op that it "took a long time to wake-up" after the past procedure. The procedure today, the exact same repeat one on the exact same body part for almost the exact same length of time, I performed the anesthetic for, whereas one of our more seasoned CRNAs had done in November. Today, less than an hour after the procedure was over, the patient was sitting up in the PACU, smiling, drinking a soda, eating crackers, and telling me how much better he felt compared to last time. That's a God's honest true story.

Now, I just don't know how you meaningfully quantify that.

-copro

How about I help you quantify that, You are amazing. Seriously, you must be the best.

Noyac · Mar 27, 2009

coprolalia said:
That's why I think it's hard to detect differences quantitatively. For example, I know this particular CRNA, though, and she's a bit of a "cowboy". I can't be sure that everything that was done was recorded (i.e., I walked into her room one time to give her a break, to which she refused... the patient was under a full TIVA GA, whether she realized it or not, without an airway satting at about 88% with a full jaw thrust... "do you need some help?" I asked to which I got, "nah, I've got it under control"... I just shook my head and left the room... hell, I'm just a resident... I don't know anything, right?... ask Noyac).

-copro

FULL TIVA

Is there a 1/2 TIVA?

How about FULL Inhalational vs 1/2 inhalational?

Yes, you are a resident. And you are showing everyone that you are a resident. And there is nothing wrong with being a resident except when you are the resident who thinks they know everything. RIGHT!

BTW resident, can you explain "some aberrant GABA receptor (or something)"? :uhno:

bigdan · Mar 27, 2009

More as a question for discussion rather than a suggestion of what might happen:

- Would data be skewed by others (circulators, surgeons, CRNAs themselves) calling for Anesthesiologist backup for the "Independent CRNA" the moment things look like they'll go South? I imagine most hospitals would not deep-six all the Anesthesiologists...so I envision you won't get tons of increased mortality, and perhaps not much more morbidity, because folks would still enact some sort of safety net. Or not?

dc

coprolalia · Mar 28, 2009

Noyac said:
FULL TIVA

Yes, as in under a full total intravenous anesthesia general anesthetic, not "sedation" (as is wrongfully stated by a lot of people who think you can push propofol to the point that the patient loses their airway reflexes, and it is still considered a "MAC").

Or, are you now going to tell me that you haven't heard people ask you to "just do a MAC sedation with a little propofol" before?

-copro

coprolalia · Mar 28, 2009

Noyac said:
Yes, you are a resident. And you are showing everyone that you are a resident. And there is nothing wrong with being a resident except when you are the resident who thinks they know everything. RIGHT!

I don't know everything. I do, at this point in my training, know how to practice anesthesia independently.

-copro

coprolalia · Mar 28, 2009

Noyac said:
BTW resident, can you explain "some aberrant GABA receptor (or something)"?

You've never wondered why, when you give 2mg of midazolam in the holding area, some people will sit there and continue to chat away and make you wonder if the supplier made a manufacturing error in that vial and it only had saline in it... yet that next patient who is a 250lb weight lifter and gets the same dose is completely snowed?

Have you ever actually seen "paradoxical agitation" after a dose of midazolam? Do you know what causes it?

Are you aware that there is allelic heterogeneity in the GABA-A subunit that cuases people to react differently to benzodiazepines, barbiturates, ethanol, and other GABA-A specific effectors?

So much for your "clinical and scientific curiosity", Noy.

But, no, to answer your question, I can't explain it. I searched the previous anesthetic record to try to discern a difference, both before and after the fact, in the anesthetic technique as to why the patient had a significantly better outcome this time compared to the previous time when I had, in fact, only given more fentanyl according to the record. The rest of the anesthetic was the same.

This leads me to the following potential conclusions, in this case, that:
(1) This CRNA didn't record everything that occurred in the case record during the previous anesthetic (which, staying on topic and not simply posting in attempt to assassinate my character) would prove it difficult to collate this data and do a comparitive study).
(2) The intervening time the patient had a clinically significant, but unquantifiable, change in his baseline health that affected his reaction to the anesthetic.
(3) A significantly higher degree of pain in the previous procedure, that manifested itself as delirium and/or was "over treated" in the PACU, caused his reaction on the way home.
(4) Some other cause.

Bottom line is that I gave this CRNA the benefit of the doubt in this case, and I was careful to observe how the patient reacted to the anesthetic this time based upon his and his wife's explicit statement in the holding area. Long story short, he reacted differently to what, literally on paper, was relatively the same anesthetic he got four months ago. These are the things that would make it difficult to do a comparitive study of such outcomes.

-copro

Noyac · Mar 28, 2009

coprolalia said:
Yes, as in under a full total intravenous anesthesia general anesthetic, not "sedation" (as is wrongfully stated by a lot of people who think you can push propofol to the point that the patient loses their airway reflexes, and it is still considered a "MAC").

Or, are you now going to tell me that you haven't heard people ask you to "just do a MAC sedation with a little propofol" before?

-copro

Sorry, but there is no such thing as a "full TIVA" except maybe in your practice.

TIVA=general anesthesia
Have any of you ever heard of a "Full General"? :laugh:

Noyac · Mar 28, 2009

coprolalia said:
You've never wondered why, when you give 2mg of midazolam in the holding area, some people will sit there and continue to chat away and make you wonder if the supplier made a manufacturing error in that vial and it only had saline in it... yet that next patient who is a 250lb weight lifter and gets the same dose is completely snowed?

Have you ever actually seen "paradoxical agitation" after a dose of midazolam? Do you know what causes it?

Are you aware that there is allelic heterogeneity in the GABA-A subunit that cuases people to react differently to benzodiazepines, barbiturates, ethanol, and other GABA-A specific effectors?

So much for your "clinical and scientific curiosity", Noy.

But, no, to answer your question, I can't explain it. I searched the previous anesthetic record to try to discern a difference, both before and after the fact, in the anesthetic technique as to why the patient had a significantly better outcome this time compared to the previous time when I had, in fact, only given more fentanyl according to the record. The rest of the anesthetic was the same.

This leads me to the following potential conclusions, in this case, that:
(1) This CRNA didn't record everything that occurred in the case record during the previous anesthetic (which, staying on topic and not simply posting in attempt to assassinate my character) would prove it difficult to collate this data and do a comparitive study).
(2) The intervening time the patient had a clinically significant, but unquantifiable, change in his baseline health that affected his reaction to the anesthetic.
(3) A significantly higher degree of pain in the previous procedure, that manifested itself as delirium and/or was "over treated" in the PACU, caused his reaction on the way home.
(4) Some other cause.

Bottom line is that I gave this CRNA the benefit of the doubt in this case, and I was careful to observe how the patient reacted to the anesthetic this time based upon his and his wife's explicit statement in the holding area. Long story short, he reacted differently to what, literally on paper, was relatively the same anesthetic he got four months ago. These are the things that would make it difficult to do a comparitive study of such outcomes.

-copro

So now you are trying to explain your BS. Yes I know why some people react differently to midazolam that others. And for your information it is very predictable. ETOH, Benzos for sleep or chronic pain, antidepressant, etc etc.

ANd I don't need to "attempt to assassinate your character", you are doing it all by your lonesome.

proman · Mar 28, 2009

How come we always end up in personal vendettas? Seriously, take it to PMs, please.

Consigliere · Mar 28, 2009

And people wonder why this speciality is failing...

dfk · Mar 28, 2009

coprolalia said:
Without an electronic anesthesia record that can't be altered and a good, full pre-operative health history that codifies all co-morbidities, I think it's really hard to quantify the differences.

-copro

i'm assuming you've used EARs, but even with these, you still can't guarantee complete accuracy. yea, vital signs are continuously recorded, but with the EARs i've used, you need to click on boxes to signify what you've done/used. very easy to not document something done (or even document something not done).

EARs are not infallible.

Noyac · Mar 28, 2009

Consigliere said:
And people wonder why this speciality is failing...

This specialty is failing? I wasn't aware of this. I seem to be doing pretty well. How about you Jet? And the rest of you in PP, are you doing OK?

I think it is healthy to challenge each other.

Licoricestick · Mar 28, 2009

Licoricestick · Mar 28, 2009

Consigliere said:
And people wonder why this speciality is failing...

I'm looking at a nationwide shortage of about 2200 anaesthetists come 2028....things look pretty good to me right now.

coprolalia · Mar 29, 2009

dfk said:
i'm assuming you've used EARs, but even with these, you still can't guarantee complete accuracy. yea, vital signs are continuously recorded, but with the EARs i've used, you need to click on boxes to signify what you've done/used. very easy to not document something done (or even document something not done).

EARs are not infallible.

Agreed. And, some systems are certainly better than others.

I think this is only part of the solution, though, and the numbers needed to detect any differences would still be huge.

-copro

Consigliere · Mar 29, 2009

Licoricestick said:
I'm looking at a nationwide shortage of about 2200 anaesthetists come 2028....things look pretty good to me right now.