System Wide Goals of Care Implementation: A Podcast with Ira Byock, Chris Dale, and Matt Gonzales

Eric 00:12

Welcome to the GeriPal Podcast. This is Eric Widera.

Alex 00:12

This is Alex Smith.

Eric 00:13

And Alex, who do we have on with us today?

Alex 00:16

We are honored and delighted to welcome back Ira Byock, who is a palliative care doctor and author and founder of the Institute for Human Caring, and he is joining us from Missoula, Montana. Ira, welcome back to the GeriPal Podcast.

Ira 00:30

Truly a pleasure to be back with.

Alex 00:32

And we’re delighted to welcome Matt Gonzales, who is a palliative care doc and friend and chief medical officer of the Institute for Human Caring and Providence, and he is joining us from Southern California. Matt, welcome.

Matt 00:45

Thanks. Nice to be with you, friends.

Alex 00:47

And we’re delighted to welcome Chris Dale, who is an intensivist ICU doc and is the medical director of Clinical Innovation and Virtual Care and Digital Health at Providence. Chris, welcome to welcome to the GeriPal Podcast.

Chris 01:01

Hey, thanks for having me. Awesome to be here.

Eric 01:03

So we’re going to be talking about a New England Journal Catalyst article that we will have on our show. Notes this podcast on basically creating a goals of care initiative in a large healthcare system and what the results of those initiatives are. But before we jump into this topic at hand, Ira, I think you have a song request for Alex.

Ira 01:25

Yes, indeed. Alex, I’d like you to play, “Just dropped in to see what condition my condition is in.”

Eric 01:33

Who sings that again?

Ira 01:35

Kenny Rogers.

Eric 01:36

Kenny Rogers. Ira, why did you pick this song?

Ira 01:40

Well, you know, we’re talking about goal alignment as key to people’s quality of care, quality of life with illness. And I just thought that this kind of evoked a sense of really looking at one’s condition and what the. What life holds.

Alex 01:58

And this is a fun song. This is not the Kenny Rogers many of you may know. This is before that. Kenny Rogers. Here’s a little bit.

Alex 02:07

(singing)

Eric 03:03

I never would have guessed that was a Kenny Rogers song. 100 million years.

Ira 03:08

It’s like a guaranteed earworm, too. [laughter]

Eric 03:11

Yeah, it is.

Alex 03:12

Yeah, yeah, yeah. Oh, yeah.

Eric 03:16

Psychedelic for Kenny Rogers, I feel.

Alex 03:18

Yeah. I think our first podcast with Ira was on psychedelics.

Matt 03:23

That’s totally shocking. [laughter]

Ira 03:27

Yeah, yeah.

Eric 03:28

But this is not a psychedelic podcast. We’re Going to be talking about New England journal Catalyst again, article on goals of care conversations, a system wide innovation. Really great study. And we will put it again in the show notes, encourage all of our readers to talk about it. But I’m going to take a big step back because I heard something that I saw in the article, something called the Institute of Human Caring. Ira, I’m going to go to you first. What is the Institute of Human Caring?

Ira 03:58

So it’s the Institution for Human Caring.

Eric 04:01

For human caring.

Ira 04:02

And it is an effort within Providence to align what we do with what people want and to improve whole person caring. We saw an opportunity back in 2014, frankly, to get in front of the leadership of this large western health system and use the knowledge, attitudes and skills of palliative care to embed within the health care system writ large to really turn focus from solving medical problems only to caring well for people as whole persons. Now, just to take you Back in 2014, we all thought that Obamacare or the ACA was about to be the new model of health care across the country.

And what I had the opportunity to do in speaking to the leadership, executive leadership and board of Directors of Providence was to make a business case for making sure that what we’re doing for our patients is what they desire and that they’re making informed decisions. The idea being that there’s a business case to avoid highly burdensome and ineffective treatment and also treatments that inform patients simply do not want.

Eric 05:23

So was the Institute for Human Care, was that created in 2014? Yeah.

Ira 05:28

Yes, it was.

Eric 05:29

All right. And around that time, Matt, if I’m reading your article correctly, congratulations on the first author. This is great article.

Matt 05:38

Thanks.

Eric 05:39

This project started around 2014. 2015, as far as.

Matt 05:44

Yeah, I think we really ramped up around 2015.

Eric 05:47

Yeah. And I’m just saying, what were your hopes for this? Like, what were you trying to hear a little bit from Ira about you?

Matt 05:54

Well, you know, I think look, to many of us in this field, AIRA has been a long term inspiration. And so frankly, just like buying into his vision and trying to see the world envision a world that we all want to live in, to be able to get healthcare in a place that actually listens to you, that actually sits down and has conversations. I think one of the cool things about the institute has been we’re an interdisciplinary team of a different type.

And so we brought folks who had a background in writing in newspapers or we had folks that were good at community engagement and really trying to think about this problem from all different ways to really change fundamentally our health care system and incline it for the better towards having these conversations. So we started out small, I mean, a lot of places, right? We started out thinking in 2015, how can we do this in just a few hospitals? As a chance, as a lab of innovation, to be able to see what it could be like to be able to focus particularly on having these goals of care conversations more often at higher qualities. Particularly focused on the ICUs, because that was, that was an easy conversation for us to start with.

Eric 07:08

And Chris, how long have you been in the ICU part of the Providence system?

Chris 07:11

Oh man, way too long, like 15 years. But I mean, I think actually to Matt and Ira’s credit is like the Institute for Human Caring. It’s not like a, it’s not a think tank, right? Like it’s not a bunch of people and it is based out of Southern California. So I imagine there is some surfing and some hang out on the beach and some bonfires that go into it. But it’s not this thing that people do and talk about it. They actually tried to change how health care is delivered across a very large, very, very diverse.

I mean, we have like hospitals from Kodiak island in the middle of nowhere in Alaska, or at the edge of the middle of nowhere in Alaska, all the way down to, you know, the LA Basin or whatever L A is. I’m not even sure I’m from Seattle, so I guess. But like, to do that across our system is really, really challenging. And so I think, I don’t know, you guys can talk about it, but you guys create like you, you kind of like, you know, ear wormed in to a structural element of how do you use a common EMR to actually affect change across this like very, very heterogeneous system. I thought it was cool.

Matt 08:07

Well, I agree.

Ira 08:08

So it’s actually more grandiose than the article even would appear to be. In a sense, this was a top down project. Not that we’re at the top of Providence, but we went to the top of Providence and we basically got their buy in so that we could mount a lot of different initiatives and quality improvement tactics, all of which reinforced a new model of quality. So we actually got them to affirm that quality at Providence Health requires a good faith effort to do what people actually want to do. That there’s listening, that there’s documentation of an actual conversation.

Eric 08:52

A goal is a care conversation.

Ira 08:54

Goals of care conversation. There are other initiatives to make it easier to have and document a conversation about who you would want to Speak for you, you know, kind of what we call advanced care planning, those sorts of things. Aligned with this goals of care conversation, we made it clear that we were not going to skimp on criteria for quality. We used, you know, other specialty services definition of quality from the icu, from College of Physicians, from College of Cardiology. And then we made it easy for clinicians to do the right thing. But we heard lots, we’ll talk about the barriers.

We heard lots of complaints about not having the time double, not wanting to document all of that stuff. But we were well prepared with buy in from the top and a willingness to continue to learn and iterate and get it make the right way easier that we were able to make huge progress.

Alex 09:56

And I love how in your article you emphasize, and it’s the first point I think that you emphasize is this is mission aligned. This is aligned with the mission not only of like the whole institute and the whole healthcare system, but you as a clinical provider. This is something we think you care about. That’s my interpretation of what that first bullet point was. And it’s interesting, you know, Eric, we just happen to be recording with Victor Montori tomorrow, whose website is like Caring Revolution and there’s this sort of like Institute for Human Caring.

There’s this movement that this is, these institutes are trying to tap into and some of it’s pushback against, you know, like documentation for billing purposes, leading to burnout and stress and whatever. And this is like a counter movement that is like, let’s get back to what brought us into this profession in the first place. Caring for other people. And how can we align what we do in a very practical sense with that mission?

Matt 10:54

100%. I mean, I feel like that’s why, I mean not only why, but it’s a huge part of why having these conversations has been so powerful. When we go into places when we used to travel more often and do large scale launches, because we would do that, go in, spend a day with hospital leadership, have them come to a communication training with us and not only come in and stay for a few minutes to introduce us and leave, but actually participate in the communication training with us.

I think what you said, Alex, is so smart. It’s that clinicians want to have these conversations. And so when we recognize that and recognize that the bond between a patient and a clinician actually is, is true anti burnout is actually so powerful in terms of being able to move this stuff forward.

Eric 11:47

So what did you actually do? Because I Remember so around 2015, your article does a lovely Job of breaking it down. You started your pilot project, you start talking to doctors and nurses about yes, like you said, Matt, having these conversations are important, but there’s all these barriers. You got time. We’ve never been trained on how to do kind of goes into EHR netherland of where these, where is this documented? So what did you do with that information?

Matt 12:19

I mean, so I think leadership buy in was, was critical. The first step, the second step for me was EHR optimization. And that’s not just because I was a software engineer in a past life. And to me, EPIC is like something I adore and not a four letter word like it is to most people because I recognize that’s true. Like it’s not, it’s not set up well out of the box necessarily to function the way that we need it to function. But we made it a lot easier to be able to like record these conversations up front. You know, before it was like the down button was everybody’s favorite button before because you’re scrolling through all the old notes looking for what mattered to the person and that takes so much time.

So we really like fundamentally change the way that we organize those. So for instance, now there’s a single source of truth that existed that we created in 2015 that organizes all those conversations into one place so that it decreases the time for clinicians being able to use that. We spent a lot of time doing the EHR stuff, but then also teaching people those pilot sites, going through and using the serious illness conversation guide framework that we adapted to become our advanced communication training. Truth be told, for your audience, it’s good communication training. Part of our branding strategy was to call it advanced and not basic because no one signs up for basic communication training. Right. And so a lot of this was getting engagement and teaching people how to feel confident in having these conversations and then knowing where to put them and where to find them down the road.

Ira 13:55

Well, there were other elements here. We also early on invested. The institute grew largely in data and analytics personnel. We had a whole team for data and analytics and we built dashboards that were refreshed very frequently and could show at any level of the organization, from the floor or unit of a hospital to an individual clinician, all the way up to a region to the enterprise.

Show how you were doing in terms of getting these conversations documented in the medical record and how you were doing last quarter or last month and how they were doing in another part of the health system during the same period of time and what the deltas were and if you were unhappy or if your CMO or CNO was unhappy, we could help you improve that in any number of ways in putting a program manager at your team’s elbow so that they could use the EHR tools to whatever we had recently tweaked and improved to education and just in time, education about how better to start this conversation to any number of things, how to bill for it, all of which we’re trying to make the right way easy, but without skimping at the goal of getting these in the chart. And we continue to iterate and listen and make it easier.

And then, you know, I want to get you in here, Chris, because we really outlasted a lot of the resistance, but also listened and drew in leaders from across this seven state health system who really did believe passionately that good quality care required, you know, listening and having a conversation and documenting what the patient had decided.

Chris 15:52

Yeah, I mean, just listening to you, I’m struck by kind of a thing that you said and the thing that Matt said too. Like I’m a, you know, an ex chief quality officer next cmo and I like, you know, like a lot of us, I think a lot about healthcare delivery and how to change and improve healthcare delivery. And I think that there are two things that you guys did that are like very catalytic and very, very cool. Like there’s a quote from Charlie Munger. It’s something like, take a simple idea and take it very seriously. And I think you guys did that. You live in this space of, hey, we have an idea that people can have better conversations and we can use technology to reinforce them and that can net produce better care and a better caregiving environment.

But I think it’s so cool. One thing I’ve been struck with Ira that you said is, and I think it’s a page for maybe the Press Ganey business playbook where you demonstrate the problem with data and then you sell the solution. Like, both those things kind of go together. I thought you guys did a very good job out of the gate saying, like, hey, there are some problems, like there’s heterogeneity across this, you know, 51 hospital system in terms of how frequently people in ICUs are having goals of care discussion, like having the data and the metrics change the conversation. Because then then CMOs or, you know, even chief executives of hospitals or regions were coming saying, hey, you know, dear ICU team, like, why do you guys suck at delivering goals of care? Having goals of care conversations like your, your partners in California, like, they’re awesome.

Like, why are you bad up in Seattle? You guys got to improve. Like, and then, Ira, like you’re saying there about like, oh, well, then we could meet them where they are in terms of what they needed. But that’s hand in hand with Matt, something you said at the beginning, which I’d be very interested to know if this is like, you guys had actual conversations about the catalytic nature of what you did in epic. Because I agree that like, having like the simple things, this is not like AI taking over the world or like, you know, like modern technology. It’s like a very simple idea. Like, there should be a place you can look to see code statuses over time. Like, wouldn’t it be cool if you could know the journey of code statuses over time? Wouldn’t it be cool if you could see the results of all these previous Goals of Care discussions? And like, that’s the, you know, the meat and potatoes of the thing you guys put in epic. I just curious, like, is that something you guys kind of had conversations about from the get go is like, oh, we need a place. It needs to have all these things.

Matt 18:03

Yeah. That felt like the top level program problem for clinicians. Right. Is like, back in 2015, when we started this, there wasn’t a place for that. And I truly believe we were one of the first in the country to be able to organize all that information in a thoughtful way in colorful tiles to make it aesthetically pleasing to look at and not wonky and spread over the chart.

Chris 18:26

There’s only so far you can go with epic. Honestly.

Eric 18:32

Like, if I went on to epic, what is what. Where is the Goals of Care conversation? Like, if I want to quickly.

Ira 18:38

There’s a tile right on the opening page.

Matt 18:41

Yeah. One click. You can find it from in our EPIC builds great.

Eric 18:44

And it pulls up the last Goals of Care conversations, or all of them.

Matt 18:47

Or organized in sort of reverse chronological order so the top one is the most recent one. And then you can work your all the way down. And I think that was really helpful. But just say that, Ira, something you said is that getting this right has required a lot of listening and iterating over time. If you’re listening to this and thinking, oh, how would I do this? I can guarantee you that you’re going to make missteps along the way, but using the skills that we have, particularly in talking to the palliative care audience here, of listening and understanding the underlying things of what resistance people have and then working to break those down.

So I’ll share two lessons that We’ve had one is early on we thought physicians and you know, Ira and I are both physicians, so this made sense. We thought that they were going to be the ones to be able to make the most amount of change early on. And so we invested time and money in training clinicians, training physicians. I have a vivid memory of hosting a communication class between 6 and 9pm in our offices because it was the only time they would show up and we bought them fancy dinners to come to the training session. And I have a very vivid memory.

Eric 20:08

Of I, a pharma company, cleaning up.

Matt 20:11

Boxes of food at like 9:30 at night in our offices. And I gotta tell you, that didn’t move the dial. I mean it kind of did, but it really barely did. We were recording a few dozen maybe conversations a month at that point in time. And what really changed the game was training the nurses, training all of the acute care nurses in the hospital. We reached 91% of all the nurses in the hospital, putting them through this two and a half hour communication training.

And that really changed the culture because for instance, on ICU rounds when they were doing their walk rounding and their interdisciplinary rounding at the, the hospital that I was at, they started asking like, oh, have the goals of care conversation been done? And then later, not was it done, but what are their goals of care? And that being kind of a standard part of the integration of the workflow. And I think for us, recognizing through the data and listening to folks that physicians weren’t going to change this, but nurses could really partner with us and help them make that change was one of our biggest learnings.

Eric 21:14

Can the nurses, are they part of the, like if they write a note about goals of care, does that show.

Ira 21:19

Up in the nurses, social workers, even chaplains?

Eric 21:23

Eric, all flows in there, right?

Ira 21:25

And I thought there would be huge resistance to that as a, as an old time doctor, I thought that’s my job. I thought that’s a doctor’s job. And I really expected that we would get pushback. Remarkably, we didn’t. I think the doctors all agreed, at least what they said out loud throughout the system, that this was important, that the having goals of care conversations before you intervene in some, in somebody’s health or treatments particularly invasive is absolutely part of quality. We didn’t get any pushback to that. But then they found any number of reasons not to do it themselves or whatever. So we said fine. You know, we’re, you know, medicine these days is practiced in a team. You’re the chair.

You know, as a doctor, you’re the head of the team. But one way or the other, the, the system is making a commitment directly to its patients that we’re going to have goals of care conversations. And, you know, and whoever has that, we’re going to train up all the people who are encouraged to have those conversations, but we’re going to have those conversations. And, and some places it was just the nurses, some places the hospital medicine docs got, have gotten involved in a big way and it’s, it’s theirs. I should also say that in this initiative, we required a goals of care conversation during the hospitalization to meet the performance criteria for ICU stays. But throughout the health system, goals of care conversations that were entered in the office by a primary care physician counts to other metrics, other quality metrics, because it’s just been done. It’s person centered. It doesn’t have to be done by the oncologist. It can be done, done by the primary care doc or the hospital medicine doc if the patient’s admitted.

Eric 23:13

Chris, did it feel like usual, this interprofessional goals of care conversations or did this feel different kind of boots on the ground?

Chris 23:22

Well, I think the main thing about it is the widespread acknowledgement in the culture of the importance of having goals of care conversations and also recording them too. I think, I don’t know, maybe Matt, you can talk about the difference in quality of goals of care conversations that you guys have looked at over time. But definitely from out of the gate when this, like Ira is talking about this, if you stay in the ICU for five nights, you need to have a goals of care conversation is the metric.

And so it’s the usual wailing, screaming, gnashing of teeth about metrics, like, oh, the data are wrong, you haven’t defined ICU correctly. And I’m just the ICU consultant. It’s a strange surgeon should be having the conversation, like, you know, whatever, blah, blah, blah, like a bunch of excuses. But this is like part of the healing process, right? It’s like part of the, like the change process. You gotta, you gotta, you gotta go through it. Yeah. So eventually, yeah, they kind of take change. And then the docs in our institutions up here in Seattle area, maybe a little less so the nurses, but really very, very aware of the importance of it.

Eric 24:22

I can see it going two ways though. One is people say, oh, this is a great mission. Like, this is important to document. We’re gonna dive straight into it. The alternative is like most hospitals do med rec notes, stuff like that, which is abysmal. I’m gonna say it’s like, it’s this idea that we’re gonna do these high quality medication reconciliation. Ultimately, it’s just like putting in a quick note about medication reconciliation that is absolutely useless. Where does this kind of. Was that a concern? Is this just documentation for the. The sheer act of documentation, like a bed rec note, or is this like. No, this is. This is high quality goals of care documentation.

Ira 25:08

So it’s both, in all honesty. I mean, there is some perfunctoriness to it and some gamesmanship. And we knew it going into it back in 2014 and 2015. We knew we were going to get gamed. I mean, this is not the first time I’ve tried to do this stuff and none of us, you know, so. And we saw initially when. When there was a specific goals of care note type that you had to open to get it counted, we saw some percentage of notes that said, talk to Joe about goals of care, wants to remain full code. Yeah, okay.

Eric 25:41

I think in the paper it was patient sleeping more than.

Ira 25:46

More than one or two. Right, exactly. You know, so we tightened that up and in some hospitals, I mean it varied. The proportion of those varied dramatically between hospitals. Some hospitals, that wasn’t trivial, you know, and they’re clearly that the hospital medicine teams have learned that whatever the hell they write is on the goals of care template, you got credit for it and they left you alone. So we then went back to up through governance. And I gotta just say, in a 51 hospital, seven state health system, you know, there are so many layers of governance that you just. We just presented to everything that didn’t move any.

Any agenda that would have us. We presented to. And you knew you were done when the popcorn stopped popping. You knew you were done when there was nothing, nobody else to present to antecedently, proactively. Nobody within Providence could tell you what the pathway to getting this into a formal policy or formal. Formal, you know, approved as a formal performance measure. There was no roadmap. You just presented and got approvals and went to the next council and next council and regionally and finally you outlasted them. And damn, like, it’s a performance measure.

Matt 27:01

Can you all tell we have a little bit of PTSD from it?

Alex 27:06

This is hard work and important work.

Ira 27:09

So persistence, persistence, persistence.

Eric 27:12

So real quick, so the results are. Matt, there are a lot more goals of care conversations. Can you give me the top level results of your. Your initiative?

Matt 27:21

Yeah. So last year in 2024, we hit just a little over 84% of our patients with ICU, length of stays, five days or more having these quality goals of care conversation. And just to expand even further, that’s actually 27% of all patients admitted to our hospitals last year had a goals of care conversation, which is like remarkable to me. We’re talking about last year alone close to 100,000 of these conversations that met our sort of like baseline quality criteria for that.

Eric 27:54

What was the baseline quality criteria? What counts as a goals of care conversation?

Matt 27:58

Yeah, so kind of as we were talking about, right, we went from a binary system as is there something in the chart, yes or no? And then refining it. Chris was a huge part of the committee that helped us think through these guidelines, right. Of like, do we know who you talk to? So can’t just be. There’s got to be a name or a relative, somebody, do we know, A brief description of what was talked about and then one of four levels of medical care. And those four levels of medical care in the paper. But simply, right, it’s like full treatment on one end, comfort care on the other. And then in the middle you’ve got a trial of full treatment on selective care, something.

Eric 28:34

And is that a drop down or do people type that in?

Matt 28:37

Yeah, those are a drop down. And part of this is that we really, we really wanted to sort of like set the litmus test of like, would a cross covering intensivist or cross covering hospitalists know what to do if someone decompensated?

Eric 28:49

So almost like a pulse note in a way?

Matt 28:51

Yeah, a little bit, yeah. Middle part of section B of like the California Pulse. But, but some more robustness also in just like who was there and some descriptions of things and, and you know, they vary. Sometimes there are just a few words or details that are added in and sometimes they’re like really robust, beautiful, you know.

Ira 29:08

And we’re teaching the Serious Illness Conversation Guide, which has all these whatever it is, nine categories, but we’re accepting as min specs, you know, who, who did you have the conversation with a little bit about what was discussed and then choose one of these four. That’s pretty, that’s pretty flexible. But we thought that at least was meaningful enough that we could count the darn thing.

Matt 29:31

That’s not to say that more isn’t important, right? Like, of course it values exploration is really important. Talking about hopes or worries, trade offs strengths, sources of strength, like all of that is encouraged. But it felt like it was kind of like setting the bar a little too high in terms of just trying to figure out like where we actually have room to. To grow long term.

Alex 29:50

Yeah. And I guess I’d channel Louise Aronson here. I’m not. I shouldn’t say that. She’s going to get mad at me. Wait, he said, what about. I remember we had a podcast with her after she wrote this New England Journal piece about how she felt like her, I think it was her mother was treated without, like, attention to nuance in the sort of care that she would want to receive because they interpreted sort of a reductionist order to imply all sorts of things about her goals of care. So I guess the critique would be that if at a minimum you have these like four drop down levels of care ranging from do whatever you can to keep me alive as long as possible, to focus on comfort, allow natural death, that you are binning people into bins that just are a mismatch for so many people out there who have, you know, nuance and subtlety to their wishes, values and goals. Any thoughts about that? You probably agree, you gotta do something. Yeah, go ahead.

Chris 30:54

How did the bins evolve over time? Because I actually thought the bins were clinically quite helpful because a lot of people do fall into one of those things, which is like, there are people on one end of the spectrum, which is just comfort and allow natural death, and on the other end is kind of do everything until everything can be. Until humankind has exhausted everything that can humanly be done. And then that middle ground, the reasonable trial drop down, it seems very applicable to a lot of workflows and, and particularly as an ICU doc, like where a lot of patients and families are in their journey. I don’t know. And it seemed like. Did those change over time too? Did they morph? I forget.

Matt 31:26

I don’t think they morphed. I mean, I think in our discussions over time with the time limited committee that you were on, like, those conversations evolved over the exact wording of them. And so maybe we got a little more clarity around them. Alex, What I would say to that is that it’s a common question that we get in these communications training classes is like, all right, four things is not really enough to encapsulate the human experience. And that’s really why we encourage people to add more robustness, particularly when we’re in those middle two categories, that it wasn’t just a drop down by itself and we’re supposed to act off of that. But there’s a description that goes along with that that hopefully adds robustness. If this, then that or these are the things that they’re worried about or make sure to call the daughter if we get in a crisis situation because there’s trade offs that we need to weigh. Certainly I think there’s far more that we can do in that over time.

Eric 32:22

And it certainly doesn’t seem like it was just focused on code status, which feels like a lot of places there is a lot of focus on code status, which was like one probably the least important of all of these decisions you have to make.

Ira 32:35

But we also iterated around code status, by the way, because we realized in this 51 hospital health system when the institute began and we began to, you know, grow quality colleagues like Chris and so many others around the health system that we had a problem with code status, that in some of the hospitals there were still partial code orders that were. That were being written and accepted. And so we took on a politically dicey initiative to refine code status policies and procedures consistent with American Heart association and College of Cardiology. And that wasn’t easy in getting rid of the partial code.

We also instituted when a do not resuscitate or do not attempt resuscitation order is written, that there is a menu that pops up. Correct me if I say this wrong, Matt, that has a desired level of medical treatment list so that you’re talking about what the patient would accept short of code status. And we gave them credit in a goals of care note for filling that darn thing out because it had to write that legitimately. You had to have had a conversation with the patient.

Alex 33:48

Before we move on, I just want to also just go back briefly and talk about what the rate was in 2016. I think our listeners will probably be impressed that the rate was over 80% of people who had ICU stays in more than five days having goals of care conversation. But I think it was less than 10%. Right. In 2016.

Matt 34:09

Yeah, for sure.

Alex 34:10

So huge increase. Huge increase, yeah.

Ira 34:13

Now, there may have been conversations that you couldn’t find, although I’m not sure that there were that many, but. But we were blind to it. We couldn’t analyze it. We couldn’t count them.

Eric 34:23

Well, Chris and Boots on the ground perspective, like, were these conversations happening and they just weren’t being documented in an easy find note, or do you think this has actually improved documentation and potentially quality?

Chris 34:37

I think it’s improved. Yeah. I think it’s improved the quality.

Eric 34:40

I mean, so this has like face validity, these numbers?

Chris 34:43

Yeah, yeah. But in an important way too, because, yeah, everyone knows that you can kind of game the system by dropping its. You know, dot GOC into your note and then like, you know, like what you guys are saying, like patient was sleeping or like important thing, I’ll talk about it tomorrow. Or you could even be like, I had a nice bagel at the cafeteria today. The metric doesn’t count what you put in there. So you can put whatever you want. But if you do that, like I think they’re like second and third order effects. Like the, the first is, I mean just like the organization, the institution is saying that this is something that is important.

Eric 35:16

Yeah.

Chris 35:17

If I’m going to write kind of a BS goals of care note, like I know in my heart it’s not good enough. Like I’ve done like I’ve done this right. Is like, like okay, things are going on. There’s a lot of stuff going on. It’s the last admin I want to get out of here. I’m just going to drop this goals of care thing. But then I think to myself, oh, when I come back tomorrow, like I got to have a bigger discussion here because this is an important thing. I don’t know if that seed would have been planted as well if this weren’t something we measured and thought.

Eric 35:40

Do you also think there’s some behavioral nudging where other people are going to see that you had a bagel yesterday as part of your goals of care conversation with patient and maybe just having it so easily accessible will like push people to do the right thing?

Alex 35:55

Well, more so I would say that like the public reporting by is it by clinician and by health, hospital or level?

Eric 36:01

Right, right.

Ira 36:02

All the way up, Scale it up, scale it down and scale it across the other parts of the health system.

Eric 36:10

The, the, the social accountability of it is a nudge that forces people to do things from just social accountability. Like am I doing a good enough note? Because other people are going to see it to these dashboards and metrics. What do you think, Chris?

Chris 36:25

I think it’s super interesting because I never really thought about that before. The way it is displayed. Matt, I think alluded to this but it’s in reverse chronological order. So these goals and care notes in the section that they’re in an epic you can read just backwards and they’re just a hyperlink where you click on it and then it shows you the text but right next to it it shows you who the author’s name is. Like there’s a colleague of mine, Leah Smith in our group who’s a fantastic doc, trained at ucsf, very, very person centered and I know that If I see Leah’s name next to a goals of care conversation note, it’s going to be a good note. Right?

Eric 36:55

Yeah.

Chris 36:56

We also, on the other end of the spectrum might have colleagues amongst us who may be a little less like that. Right. And so you know, like who the good notes are. So I wonder if there is some subtle psychological thing that I want to be like Leah and like leave my friends in the future or even me if I’m coming back like a year later to have another goals of care thing.

Alex 37:11

Well, speaking, can we stick with the nudges for a moment? Because I believe that you did incent, you didn’t incentivize the individual providers to write these notes. Other than this is an important metric. They’re seeing what other people do it, it’s aligned with their mission. But at the top level, you were talking about the top buy in at the top. Weren’t the, at the hospital’s executive C suite level, weren’t they part of their financial compensation was impacted by their hospital’s rate of documentation? Am seeing nods. Yes.

Ira 37:42

Well that’s why I say that, you know, we went all the way up these endless layers of governance to get it approved, but once approved, man, it kind of was a top down project and we were just helping everybody at all levels excel. We want you to excel. You know, we were going to hospitals and the nursing and medical teams within the hospital tell us what we can do to help you excel. We’ve got all kinds of tools and education and you know, tweaks to EPIC and all that to make this easy for you to excel.

Matt 38:14

I’ll just like kind of wedge in that. It wasn’t, we did have some of like the C suite folks had their compensation at risk, but it wasn’t about the outcome of this conversation. Right. So it wasn’t about trying to make people more DMR or to limit more care. It was, did the conversation happen and was it a quality conversation? And if you hit your thresholds, then yeah, those folks were eligible for more bonus. What I would say is that was super incentivizing to folks. I still get calls from chief medical officers within our health system that are like, why are we not performing on this? I’m a little bit worried about our numbers. How can a new we partner together to get them back up to where they should be.

Alex 38:52

It’s really interesting to think about the compensation at risk at that chief medical officer level, which is different from putting the individual ICU doc, for example, or nurse working in the critical care Unit compensation at risk. It both appeals to me. It also makes me wonder how transportable this is to other settings because, you know, for listeners out there who may be like, you know, practicing palliative care in some hospital and wondering, well, how am I going to change the compensation package of my chief medical officer? Like what? I don’t have that power, right? Like, how do you, what do you say to them?

Ira 39:33

You know, all I can say is this was really methodical and, and the opportunity to develop the institute and to get buy in across the organization in a way that was perceived as integral to the mission of the organization at all levels was, was one that was too good not to, not to take part in a energetic way. And we did. We were, we were fortunate, but we very strategically work to make this.

So I wanted to say before we get too far further in the conversation, that there is a table within the paper that shows a leading indicator of change, which is at what percent does code status change for hospitalized patients with a serious chronic illness, one or more serious chronic illnesses? How frequently does code status change during a hospitalized. So an acutely ill patient? What you’ll see in there is that there is a positive association between goals of care conversations during that hospitalization and the change in code status. We were agnostic to how the code status changed, but only that the change in code status was reasonable to, to assume was moving towards something that was more aligned with a patient’s own values, preferences and priorities.

Eric 41:01

So the general theory with palliative care is we’re not going to tell people what they want, but if we actually have these discussions, more people are going to choose potentially not the default in medicine, which is we’re going to do everything, including cpr. If you were dying.

Ira 41:17

Yes. And what we saw was a positive association and even in patients who never saw palliative care during that index hospitalization.

Eric 41:26

And more people got palliative care consults. Right. And the goals of care conversations, is that right? Or there’s another table looking at palliative care consultations completed, not completed. I think they’re. Maybe it was the. I do know that those who had a gold care conversation documented more people had a palliative care conversation, 59% versus 41%.

Ira 41:46

You know, another part of this large grandiose initiative or of the institute was to improve the access to specialty palliative care and the quality of specialty palliative care. And that’s a whole other set of tactics, but they’re aligned deeply, they’re distinct.

Eric 42:07

So let me ask you this from a sustainability standpoint, Dashboards, they’re up and running. You probably don’t have to do a lot of work on that. Goals of care notes. Like, it’s in the chart. Like, you just get providers to do. They’re supposed to be documenting any. It’s there. The one thing that seems very hard to sustain is these goals of care conversations. Like the training. Like, are you still doing the trainings? How is it built in? Like, how many people actually did it? Like, where are you with things?

Alex 42:35

Still meeting with people at 6 to 9 at night.

Eric 42:37

Yeah. Bringing good food.

Matt 42:40

Thankfully not. Yeah. The training’s an interesting story, for sure. We’re continuing to do them, but we’ve transitioned during the pandemic, we transitioned a bunch of them online to asynchronous learning, and now we do virtual role play as well. So that’s helped. Right, in terms of, like, we’re not flying all over these seven states trying to host these and do these conversations or do the trainings. I think that’s.

Eric 43:03

That’s been virtual role play.

Matt 43:05

Yeah. Like, the three of us would be in a room together. I’m an observer. Ira plays the patient, which she’s had to do in the past early days, and Chris is the physician. And we work together all under the watchful eye of someone from our team. So we’re looking at ways to scale that, but we’re not seeing static either in terms of even the ways that we’re thinking about these things in the ehr. So, you know, Deb Unger, who’s our director of medical informatics at the institute, just recently put out last year a poster, and we’re in process of publishing this.

But, like, for instance, you know, dot Goals of care, as Chris mentioned, is great, but it has limitations. Like, you have to remember to do doc Goals of care every time you have one of these conversations or embedded in your note. And now we’re moving from that to having embedded GPT4 into Epic. So that just reads these conversations. It reads the documentation and identifies.

Eric 44:01

So, like, an AI agent comes in and scours through EPIC and summarizes goals of care.

Matt 44:07

It doesn’t summarize, but it tags the conversation. It tags the notes that contain a goals of care conversation. So we can stop, for instance, training new people when they show up. They don’t need to write type goals of care soon because AI will just find it. And it gets to your question, Alex, of like, are these clinically meaningful? The AI can screen out. Is this just a code status conversation or not? In fact, her tool is amazing. It’s got a 95% specificity and an F1 score, if you know, AI of 0.76, which is great. We have it running in four of our hospitals right now. We’re going to go up to all 52 in September because.

Eric 44:47

Wait, so you’re using AI to see where are these goals of care conversations and to tag those conversations. And it’s really good at it. It sounds.

Matt 44:56

It’s really good at it. And I think that.

Eric 44:59

Is that going to be another New England journal, Catalyst paper right there.

Matt 45:02

But we’ll see. That’s definitely, definitely coming. I would just encourage folks just to say that, like, we wanted to put this out there because it’s so critically important that we get this right. And if folks are thinking, oh, I couldn’t do this, you can. And you can learn from the effort that we’ve done. You don’t have to start from where we started 10 years ago. You don’t have to reinvent the wheel. A lot of what we did is in the playbook that we published. And we’re always happy to help and talk and think through things because it’s important that we get this right for all of us.

Eric 45:35

Chris, I’m going to go in boots on. Do you still work in the icu?

Chris 45:40

Yeah.

Eric 45:41

When you’re on the weekend, things are crazy. Do you find these things helpful? Oh, yeah. This is where we are.

Chris 45:49

Oh, this is impactful a million times, yes. Like last night I do. Telecritical care is one of my things. And so last night I was working from the Seattle area, covering prav Anchorage, and there was a guy who had unfortunately, a bleed in his brain and his sodium was rising. And so there was a discussion amongst the nursing staff this morning about what we’re going to do. And like, I opened his chart, I clicked on the goals of care tab, and I saw a note from. He actually had, like, there were actually two really good notes in there. One from the ICU flavor physician and then just like above that. So, like, more recently, the palliative care team had dropped a note and it very clearly said, like, they’re waiting for his son to come in from rural Alaska. We’re not escalating care. And I was like, okay, so I’m not going to write for D5W. We’re just going to watch the sodium, like, every single day.

Eric 46:34

Go ahead, Ira.

Ira 46:36

So I would just say that the sustainability model was designed from the get go as to lift from current state to future state where this was a heavy lift that required lots of different education and buy in and perform incentives and all kinds of dashboard, blah, blah, blah. But we’re now at the future state where this is kind of the new normal and within Providence while you still have to educate new employees and you have to refresh things and build skills and iterate quality, right now it’s kind of like, well this is just the way we do things here and it is sort of self perpetuating. Well, we really have gotten over a big hump.

Eric 47:18

Last real quick question, Matt, you got 10 seconds to answer. What is the next future state in 10 seconds? Yeah, maybe I’ll give you 30 seconds.

Matt 47:28

The next future state is moving out of the ICU for us. We’re partnering with all of the other service lines, hospital medicine, oncology. It’s so clear that when we have these conversations earlier and more often, it’s really impactful and helpful to patients and families. So that’s what we’re doing.

Eric 47:47

That’s awesome. Well, with that maybe we’ll get a little bit more Kenny Rogers to tell me what my condition is.

Alex 47:53

(singing)

Eric 48:50

Ira, Matt, Chris, thanks for joining us on this GeriPal Podcast.

Chris 48:53

Thanks gang.

Ira 48:54

What fun.

Matt 48:54

Thanks you guys.

Ira 48:55

Thanks for having us.

Eric 48:56

And thank you to all of our listeners for your continued support.

***** Claim your CME credit for this episode! *****

Claim your CME credit for EP366 “System Wide Goals of Care Implementation”
https://ww2.highmarksce.com/ucsf/index.cfm?do=ip.claimCreditApp&eventID=16311

Note:
If you have not already registered for the annual CME subscription (cost is $100 for a year’s worth of CME podcasts), you can register here https://cme-reg.configio.com/pd/3315?code=6PhHcL752r

For more info on the CME credit, go to https://geripal.org/cme/

Disclosures:
Moderators Drs. Widera and Smith have no relationships to disclose.  Guests Ira Byock, Chris Dale, and Matt Gonzales have no relationships to disclose.

Accreditation
In support of improving patient care, UCSF Office of CME is jointly accredited by the Accreditation Council for Continuing Medical Education (ACCME), the Accreditation Council for Pharmacy Education (ACPE), and the American Nurses Credentialing Center (ANCC), to provide continuing education for the healthcare team.

Designation
University of California, San Francisco, designates this enduring material for a maximum of 0.75 AMA PRA Category 1 credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

MOC
Successful completion of this CME activity, which includes participation in the evaluation component, enables the participant to earn up to 0.75 MOC points per podcast in the American Board of Internal Medicine’s (ABIM) Maintenance of Certification (MOC) program. It is the CME activity provider’s responsibility to submit participant completion information to ACCME for the purpose of granting ABIM MOC credit.

ABIM MOC credit will be offered to subscribers in November, 2025.  Subscribers will claim MOC credit by completing an evaluation with self-reflection questions. For any MOC questions, please email moc@ucsf.edu.