FEMALE SPEAKER: Welcome to Mayo Clinic COVID-19 Expert Insights and Strategies. The following activity is supported in part by an independent medical education grant from Pfizer Inc and is in accordance with ACCME guidelines.

OGNEN GAJIC: OK. Good morning, everyone. Welcome to METRIC 2020 Ritual Critical Care Conference. Today's topic is a tough one of Shared Decision-making in the Context of Critical Illness. And I really have a privilege to talk to a real panel of thinkers in this field.

I would just like to have anyone-- everyone introduce themselves. So I'm Oggy Gajic, intensivist from Mayo Clinic and have a privilege to moderate this session. Victor.

VICTOR MONTORI: Hello. Victor Montori. I'm a consultant at Mayo Clinic as well and researcher with the Knowledge and Evaluation Research Unit on Patient-centered Care.

OGNEN GAJIC: Alison.

ALISON TURNBULL: Hi. My name is Alison Turnbull. I'm an epidemiologist in the Pulmonary and Critical Care Department at Johns Hopkins University.

OGNEN GAJIC: Scott.

SCOTT ABEREGG: My name's Scott Aberegg. I do pulmonary critical care at the University of Utah.

OGNEN GAJIC: Thank you. So to put this in a framework, can someone tell me a little bit about decision or a definition of shared decision-making, so what it actually means? Victor.

VICTOR MONTORI: Yeah. I can take that. Thank you, Oggy. So shared decision-making is a conversation between patients and clinicians with the intent of figuring out what to do. That's the simplest definition.

It's a conversation. It's shared because there is at least two parties, could be more. It could be more people on the clinician side, more people on the patient's side. And the objective of that conversation is to figure out what to do.

There are four types of shared decision-making. The first one is the consideration of a matching process between one's preferences and the features of the available options. So options are presented with their pros and cons with the intent of figuring out which of the available options matches the preferences of the patient.

A second one, which has to do with resolving conflict, where a patient may have internal conflict, or there might be a conflict between the patient and the clinician. And the main goal there is to reconcile and arrive at a consensual decision in which it becomes clear, what is it that the patient wants? So the first one is preferences. The second one is wants and desires.

The third type, which becomes a little closer to what we're discussing, I think, today, is a problem-solving approach. Here it's not an issue of matching preferences to options. It's an issue of figuring out, what is the aspect of the patient's situation that demands action, and what are the-- what is the action that the situation demands? And it's an iterative problem-solving conversation in which patients and clinicians embark to-- until the right way forward emerges.

And finally the fourth approach, which is also pertinent to our discussion today, is the kinds of discussion in which what we're trying to do through dialogue is to figure out what optimizes our-- the humanity of the patient. What allows the patient to proceed in a way that makes the patient whole? And these are common discussions at end of life. These are common discussions in transitions, for instance of transgender patients that are going through transition surgery.

These are the kinds of discussions that affirm who the person is. And how is it that they belong to the larger picture of their family, their community, and the world? So there are four kinds of shared decision-making.

This will be the last thing I'll say. The main goal of shared decision-making is to arrive at a plan of care that makes intellectual sense. So it is consistent with the best available research evidence, and it matches the situation of the patient quite well.

It makes practical sense. So it can be implemented. A solution can be implemented either in the health system or in the life of the patient, which is the most consequential in my field, which is chronic disease. So it's practical sense, it's feasible.

And the third thing, it has to make emotional sense. It has to feel like the right thing to do right now in this situation.

OGNEN GAJIC: Thank you, Victor. So obviously, humane, patient-centered care is by-- patient-centered critical care is a holy grail for everyone if we get sick ourselves, for our families, and for our patients. So we're all here for it, and it's tough. So Allison, you've been doing a lot of research on shared decision-making in the context of critical care, not in the context of current pandemic-- we'll get to that-- but in the context of critical care in general. So what are the specifics, and challenges, and thoughts on shared decision-making in the context of critical illness?

ALISON TURNBULL: I think one of the first challenges that often gets overlooked is recognizing when a situation actually calls for shared decision-making and what does not. At least here in the US, I think there's a tendency to be a little too broad in what we think requires shared decision-making with a family, particularly when a patient lacks capacity and is critically ill. So the first step is recognizing what is a preference-sensitive decision where we need to take into account the values and goals of a patient or their family before deciding on the plan of what to do, as Victor identified.

In the ICU, I think the next issue that we often run into is figuring out who should actually be involved. So if a patient lacks capacity, then you run into the kind of morass of who is legally somebody's proxy, health care proxy. And the rules for identifying that person varies enormously state from state, state by state. And the person who is the legal health care proxy is not necessarily the person who actually wants to be involved in this process or is best situated to be.

I think another thing to think about is that we talk about shared decision-making as a sort of holy grail, as the aim, the goal. But the ATS and SCCM guidelines on shared decision-making make it quite clear that a number of different ways of making decisions with patients or their families are appropriate. And that can be more physician-directed, and it can also be more family-directed.

So while research has shown that the vast majority of families fall somewhere in the middle and want a shared decision-making process, as particular decisions become more value-laden or more difficult, some families will want more control. And other families actually want physicians to take more of a directive approach and to make clear recommendations or to even take responsibility for really difficult decisions. So I see the second challenge in the ICU as understanding what role a family member actually wants to play in a specific preference-sensitive decision.

And then this is a long process. This is an incredibly emotionally fraught process often. There needs to be trust, and that can be hard to build.

Unlike in a chronic care situation where you have a long term relationship, longitudinal relationship as a disease develops, people are often meeting an ICU physician for the very first time. And given the way schedules work and the number of other people involved, they may see a different doctor as a shift changes or a specialist comes in. And it can be very hard to develop a trusting relationship for a high-stakes decision. So I see those as being the fundamental challenges to shared decision-making in the ICU setting.

OGNEN GAJIC: Thank you, Alison. So I think we've got to this, so first there is a time pressure. Second thing, the patient may not be able to do it themselves. So they may not be competent or able to participate himself. And the third issue is a trust that is very difficult to get on a fly with this.

So Scott studies decision science in general as well and does this for life in his clinical practice. So given these challenges, my sense is that this often goes wrong. And the decisions made are not really appropriate, coherent in a sense that Alison mentioned, that they should be logically coherent and appropriate within the framework of being really patient-centered and to the all ethical principles of that. So Scott, what are your-- what is your experience or thoughts on the shared decision-making in the ICU in general?

SCOTT ABEREGG: Well, my expertise if I ever had one is more along the lines of the formal study of decision-making, which-- because earlier in my career, I did some survey-based research. So those principles, I think, are useful to help guide families through decisions to a limited degree. For example, one model of decision-making promoted by Jon Barron, who is a psychologist at University of Pennsylvania, is that thinking and deciding, making decisions is about searching.

And there are three things that you search for in his search inference framework as he calls it. You are searching for goals, and then you have to evaluate those goals and consider the trade-offs that must be made with all of the different goals, which is I think germane to what we're talking about here. After the search for goals is a search for possibilities or alternative ways of satisfying those goals.

And then finally is a search for evidence, to look for factual information that can be brought to bear upon the possibilities to evaluate which one of the possibilities is most likely to achieve the goals and in a holistic, practical way that balances competing goals. That's something, I think, is useful for me anyway to fall back on when I'm thinking about any kind of a decision.

OGNEN GAJIC: Thank you, Scott. So one thing that comes up often and-- is having an advance directive as something that would help these emergent decisions in the future. Personally, I've found that not particularly useful. And I in particular did not find useful the DNR and DNI discussions.

The only thing I've asked people who are in emergencies, god forbid if you can't speak for yourself, who can speak for you? Because decisions are so contextual. And while I'm building trust, that was something.

But obviously, there are different opinions, and you guys have done a lot of research on this. So what is the role of advanced directive, if any, and do not resuscitate and do not intubate discussions when they come, like before you are becoming critically ill? So not once you are critically ill.

I'm talking, let's say, we want to prepare everyone. Now everyone is-- all our physicians wants to put advance directive, being scared of what they can-- what they may go through if they get sick with the COVID, the idea. So who is going to take it?

VICTOR MONTORI: Yeah. I can take a generic thing. And then we can go to critical care with my colleagues since I don't do critical care. Just like with shared decision-making the process, that most value in any of this advance directive or other things is in the-- in listening carefully to the reasons people have for what they're saying. It's a way of discovering what seems to be motivating our approaches. I think Scott mentioned the issue of goals.

And it's sometimes-- I remember filling out advance directives with my father before he died. And he was pretty clear when he was wanting certain things. The interesting thing was not that he wanted this or the other-- is the reasons he would give for wanting those things.

And in listening to those reasons, one can uncover not only what matters to people, what is at stake for them. But also, you can pick up, for instance, misunderstandings that might actually be affecting how they're making those calls. And you can-- you get to potentially inform them better. So that's one generic point to make.

The second point to make is a little more contentious, and it's the issue of being able to imagine alternative futures and how it is that you can imagine how you will fare in those alternative futures and how you feel in those alternative futures. Some people have called this affective forecasting. And the ability to be able to put yourself in that situation and make calls for when you will be in the situation, let alone a proxy, is actually very limited and fraught with problems as well.

So one concern I have about the focus on figuring out what your goals are is that it forces you to think about the future, where we might not be able to do that very well cognitively. And that's why I think, just like Scott falls back on goals and trading them off, I fall back on problems. What is a threat? What is going on right now? What is it affecting? How is this affecting who you are and who you want to be?

And then, how might we then identify ways forward that will advance those things, solving the problems that you're facing now or addressing the threats to your situation, in ways that-- and this is where I think we all agree-- in ways that not only are consistent with your values and preferences, which again is another challenging concept, but rather ways in which you will find that it makes intellectual, emotional, practical sense to you?

OGNEN GAJIC: So thank you, Victor. I've heard here two things. So one is you want to understand the goals more like motives.

You cannot say a specific thing because it's more important to know the reasons for wanting or not wanting specific things that you may not even understand what it is. So understanding the real kind of goals of what is important for them. So diagnosing a person and the personality and what is really important for them, so that's the challenge that is needed.

And then you said being able to imagine something that has not happened yet is a real challenge. And to that extent, I really-- my personal experience and my reading gets to that it's very difficult and limited to have. There is some role, but just a pure advance directive in any way-- yes, go ahead.

VICTOR MONTORI: So just to add a little more precision, one of the-- so what-- that's from the patient side. But the clinician who is interacting with the patient in that process has to have a curiosity, has to want to know this person, wants to discover who this person is. And I think Alison mentioned that, for all sorts of reasons-- maybe it's the clinician's first time with the patient, maybe there are time pressures, maybe there is efficiencies that are built into the system-- we undermine the opportunity for curiosity, to discover who this person is or, if we're talking to a family member, who this person was when they were able to express themselves and manifest themselves in more humane ways. And so curiosity becomes an important thing.

And then the other corollary of that, it's not about the menu. It's not about, what is it that we could do, and you check the box, and you tell me what you want. It is about the process of figuring out what's at stake and making the person whole.

OGNEN GAJIC: And it will come to that when we talk about trust. And obviously, I-- it cannot be said better than said [INAUDIBLE]. The [INAUDIBLE] came for the patient that you actually care. And if you care, then you care about the person as well. And you would wonder what that person is, too. So you can try to help them as a healer.

So Alison, these three things, so in a context of doing it before, being able to forecast, being able to diagnose-- to not diagnose-- to be curious and understand what is-- who is the person in front of you with their goals, so you can match the recommendations to their goals, and doing it upfront or maybe just designing someone who can speak for you when it happens.

ALISON TURNBULL: Yeah. So advance directives are tough. I generally agree with you that they-- we have pretty good research that shows having an advance directive, even one that says do not do x generally is pretty ineffective at stopping initial resuscitative attempts, is not associated with decreased likelihood of being admitted to an ICU. But what they do tend to do is, or what they've been shown in research to do, is often decrease the time that you're there.

So it's very hard. That piece of paper is unlikely to stop the regular processes. And I have to say, frankly, we ignore people's advance directives on a frustratingly high level and frequency here.

What protects you or what can change that course of events, that normal course of events, is a family member who is very clear about your wishes and is very outspoken. And you get that family member by having conversations. And those conversations can often be started by saying, I'm going to complete an advance directive. As a family, let's talk about what matters to me and who's going to speak for me, because a family member who shows up in-- during an admission or in an acute situation and says, this is what needs to happen, or this shouldn't happen, especially if they've been named as the health care agent, can be very effective.

But if that family member is saying something discordant from an advance directive, we almost always go with the family member. It's pretty rare otherwise. So are they useful? I think they're a useful tool for starting a conversation and for thinking yourself about the future.

I realize affective forecasting is an issue, but simply saying somebody may feel differently in the future should not negate their ability to make a decision now. So just because you might feel differently in the future doesn't mean you should not be permitted to state preferences or make decisions now.

We can never know how any of us will feel in the future. That's just-- there is never going to be certainty about these decisions. And if you require 100% certainty before taking action, you just won't. So they're a useful tool, but we have good research showing that the extent to which they actually end up influencing care down the line varies enormously by institution. And that makes them not a perfect tool by any means.

The other thing I think that they are good for is I think they are helpful to family members, particularly if you are stating that at some point you want to stop treatment or stop the use of life support, because asking-- even if it's in your advance directive, we always ask family members to consent to that anyways. Whether or not we should, we do. And that is a decision that family member will have to live with for the rest of their life. And they will have to have a story that they tell themselves about how and why they did this.

And having a piece of paper that they can point to that says I was carrying out my loved one's wishes can do an enormous amount for their mental health in the long run. So I do see it as something that is worth doing for your family members, if nothing else because they're the ones in an acute ICU type situation who have to take on the psychological burden of making extremely hard decisions.

OGNEN GAJIC: Thank you, Alison. Anything, Scott, to add to this piece, like doing it before it happens, like advance directive? Or any other tips or trick, did you find them useful in any sense, or anything else to add?

SCOTT ABEREGG: Well, following on the heels of what Victor said, the affective forecasting is beset by several known biases. One that I always think about is called projection bias. I'm thinking of a paper by Ubel and Lowenstein. It's about, oh, going on 20 years old but demonstrates people's inability to actually-- to accurately be able to predict future affective states. And when your situation changes, and when you're closer to certain outcomes, they increase in salience. And so we have to be aware of that when we're talking about advance directives.

The other problem, obvious problem, with advanced directives is the inability to present enough information about the myriad different situations that people might find themselves in so that they could give goals that might actually be applicable to a real situation that they find themselves in rather than vague boilerplate. And then all-- the only other comment that I have about it, too, is when we're talking about patient preferences and shared decision-making, I'm reminded of a paper that was in Medical Decision Making in 2005 called Cure Me Even If It Kills Me. It's a fantastic paper that presented patients with a suspicion of cancer and then gave them options, for example, watchful waiting versus surgery to remove it, with associated estimates of what their outcomes are going to be in terms of mortality and whatnot.

And that paper found that patients were willing to accept significant risks of death just to get the tumor out of them. It was the psychological burden of knowing there might be a tumor in you that was so salient to you that you would rather statistically worsen your outcomes by pursuing surgical removal rather than, for example, radiotherapy or watchful waiting or something like that.

And so when you learn about studies such as this, you begin to recognize how nuanced and complicated these decisions really are and how that the models that I used for studying physician decision-making-- a normative model, for example, based on utility theory or consequentialism or whatever-- how that it might be difficult if you tried to apply those models to patient decision-making, because you might actually be surprised at some of the what you might say are irrational choices based on the numbers that patients make. And those arise because of values that people are assigning to certain outcomes that you might-- might be different from yours or hard for you to predict or even understand once they're stated.

VICTOR MONTORI: Can I make a comment on Alison's [INAUDIBLE]?

OGNEN GAJIC: Yep. Go ahead.

VICTOR MONTORI: So I actually agree that the fundamental role of the completing the document, again, is in the conversations that completing the document triggers, sometimes even for the first time, and the consequences of having that in place have for the psychological well-being of the family members after the fact. So I'm on top-- I think that's-- but one of the things that people have worried out loud, which I think Allison has negated effectively by saying that most of the advance directives are ignored, is that people do not want to be held accountable to what they thought before they were sick when they're sick.

In other words, it's not necessarily true that they're-- because of all the reasons that Scott just mentioned in terms of affective forecasting-- that what they thought was going to be terrible now becomes acceptable because they're in a different situation. So there's been a shift in the way the situation is appreciated. And those options now become-- the values associated with their initial expression of goals may have also shifted along with those.

So it's actually almost a good thing that we don't necessarily hold people to their expressed goals or their advance directives when they're sick, because it's likely that things have shifted. And again, we go back to what we learned in the conversations that were produced while making those to look in there for the things that might actually be still pertinent now that the situation has shifted.

OGNEN GAJIC: Thank you. So I think as we move to-- I'm going to keep some of these things from advance directives now to the real time decision-making when the patient is in a context of critical illness. And from what I heard with advance directives, there is a limited role.

However, the conversation triggered, and maybe as Victor said for the first time when you do that between you, your family, you as a patient, your family, your trusted doctor, is something that is useful to start. And other than that, the other-- the paper itself and the form has limited value that the families can relate to it, can remember when they've done it together.

And so I've done this with my mom. At the time, we did it together. And it was more important. The paper was totally useless, but having her and I having a conversation about it was-- got to this really important stuff.

It's not the actual procedures but the reasons why she would or would not know it. And similar to Scott said, she underwent completely unnecessary surgery for a cancer. And she said, I don't want to do-- I don't want a word cancer in me, period. So exactly to the point of Scott. And I didn't-- I went along with it. That's just fine.

But on the other hand, for some of the-- when she came to the end of life, she for a year didn't go to the hospital. She was cared by me here in her bedroom and caregivers, which was I-- and which I'm very proud of. And I found it very gratifying. But these things that are challenges of affective forecasting, wanting to and being able to understand that the person, the patient is a person, building trust, and then communicating it in the emergencies, those are all the challenges that were brought even for advance directives.

So maybe we-- I start this time with Scott. So now this is-- forget about, you may or may not have advance directive. Maybe it's somewhere hidden, and you can't even find it.

And the patient is there. And hopefully the family member is there, or the patient can tell you, still mumble that who can speak for you. And you said, you're safe here. You're trying to build a trust. Who can speak for you?

So how do you actually approach these challenges of, A, affective forecasting, trying to understand who this person is so that you wouldn't really violate their reasons for they would want or do not want some kind of a projective future, understanding that this is difficult to imagine? How you communicate that, please give us your tricks and tips. And maybe then we'll have the researchers and deep thinkers, Alison and Victor, come in to rescue with regards to frameworks, how to deal with it.

So what do you do, let's say, again, not in a context of this particular moment and the COVID pandemics where we may have even not enough resources to treat people, but in general, in critical illness? OK, you have a patient and how you approach the aspect of needing to have them enrolled. It's a patient-centered-- cannot say, OK. I'm going to do this and forget about you. So what do you do? Scott?

SCOTT ABEREGG: Just to be clear that I don't go on the wrong-- down the wrong path, can you restate the question just once more?

OGNEN GAJIC: We talked about advanced directives. We talked about trying-- so we-- and what is the shared decision-making. So being able to have the basically patient understand the potential options, have a logical options, and have a decision that is informed so that the patient is-- so you can talk about it. We could-- be it is informed consent, is this what?

But there is an absolute need in a patient-centered care to understand the person's goals or reasons for certain goals in a context of critical illness, being it what are the potential consequences if I do one versus the other option and then how you personally deal with it. So the patient comes. They have critical illness, and you are to give them consent, assent, any kind of approach that you are trying to achieve.

So what is your-- just say what you-- it doesn't mean that you're doing it good. No one is doing it good. That's why we have this shared decision-making.

What do you do? And then by having you saying, just imagine the patient, what you do every day. Just tell us that with us, Victor and Alison, to see, to come to a rescue.

SCOTT ABEREGG: Well, one of the things that I found most helpful is first getting older, but in terms of gaining experience and understanding what-- how to focus the conversation on what patients really want to know. And I think it's only with experience that you figure out what they really want to know. And they don't oftentimes want to know about the next couple of hours, days, or even weeks. They want to know where this journey is going to take them.

And one of the difficulties about that discussion are the inherent uncertainties of prognosis. But giving some kind of guide or estimate, I think, is sufficient because they recognize that you're not omniscient and that you can't predict the outcomes very, very far into the future. But for many patients, learning that a tracheostomy is going to mean a transfer to a long term acute care hospital and that the-- and what the prognosis is of survival and return of independent function at the end of some time period, six months or a year-- I think guides like that make the choice more, as I said previously, salient or meaningful what the questions you're asking them are actually getting at.

It's not whether or not you want a tracheostomy. A lot of patients will tell me, no, I would never want a tracheostomy, because they have an affective response to that. But they don't know what that means and what the alternatives are.

So I think focusing the conversation on where you are likely headed, and whether or not that place is going to be acceptable to you, and what you're going to have to go through, experience in order to have a chance at getting at that end result, whatever it may be. That's my two cents on the question. And I'm not sure I answered what you were asking.

OGNEN GAJIC: No, it's not the answer. You're telling it what it is. So I think what comes through that's obviously, as you mentioned before, the words matter. And we'll get to the communication.

So the same as the word "cancer" matters in decisions to what you're going to do about cancer, the same certain words matter, like tracheostomy. And people may not really know what it is, or long term, or a nursing home for that matter. So the words would be he or she would have never want to be in a nursing home and things like that. And not necessarily these things are very specific.

But projecting, rather than a specific procedure, like tracheostomy, projecting to some understanding of the, as Victor would say, before. Why? Not the it but the-- not the what but the why the person would or would not accept certain procedures. So what are the projected goals that they want?

They maybe don't want to be a burden to their children. And knowing that they don't want to be a burden to their children, or their spouse, or someone like that is very important to know. That's something to respect.

It's not the actual procedure, but it is what, how much? So trying to find a real-- which I think that's what has come up to it. But the anything you know [INAUDIBLE].

So now we are in emergency room, we're in the ICU. So we have a critically ill patient. And Alison, what do we do to get to these? So we don't have effective way to project ourselves in the future.

We may not have a trust, but let's say we have a fantastic physician who actually cares about the patient. So there is actually-- that's not an issue. Let's say that we are like that. So we have that, and they trust us. Let's say we will see what we can do to improve just how to approach this conversation so that we achieve or move closer to matching the available options to patient person and goals.

ALISON TURNBULL: Yeah. So I want to second a lot of what Scott said about families, or patients especially, not being focused on the immediate but wanting to have an idea of what the long term trajectory looks like.

So there's a famous study by Terri Fried in the New England Journal of Medicine in 2002 interviewing elderly people who already had COPD, or heart failure, or cancer, that asked what longterm outcomes they'd find acceptable. And things change dramatically if you said "survive with all sorts of new disabilities" versus "survive and get back to your baseline." People made completely different decisions. And I think that's a lot of what people are trying to understand, is not, what's happening right now, but what does this mean for the long term?

I'd also-- I completely agree that not focusing on the procedure is really important. And I know that as a physician, the immediate things that matter to a physician are, what am I going to do, and when can I discharge this patient? So what's the thing I need to do? And then are they going to survive to discharge?

And unfortunately, that often frames these conversations. I think you're going to make it. What do you think about this procedure? And that's totally anathema to the way the patient is thinking. I think it makes a lot more sense to step back and talk about outcomes, which long term outcomes are important, are acceptable. And then based on knowing that, make recommendations about procedures that will help them achieve that.

So framing the procedure as something that will help you get home, or something that will help you be alive for that wedding or whatever it is, or something that will help you feel less uncomfortable and framing everything around that long term outcome that is important to the patient. That's a framework that families can understand and patients can understand whereas going into the details of what's involved in a procedure is going down a medical rabbit hole that tends to not help.

So as hard as it is to get out of the physician mindset of, "What am I going to do, and am I going to be able to discharge this patient?" and into the patient mindset of, "What do I want? Where do I want to be, and what will help me get there? Will this help me get there?" is ideal, I would say.

VICTOR MONTORI: I have a concern about-- so this is very helpful to me as a non-intensivist to learn from you what you-- how you see it. And we've done some work in both ICU and emergency settings where this issue of trajectory, also in end of life with patient who's chronic insufficiency, the issue of trajectory. Also with elderly people coming in for exacerbations of chronic conditions to the emergency department and the notion of being admitted, readmitted, or sent back, or not even transferred to the hospital become issues of choosing to go through certain paths with the idea that the paths have a chance of leading to some outcomes, like going home or not going home, or dying peacefully or dying here, or et cetera.

But it's the nature of the-- what people experience at the end of the day is not that end outcome, which they will-- hopefully, they will arrive in some fashion. But what they experience the most is the road, the way there. And there is some balance that has to be kept, which is tough, between, how is it that the day to day will look like or the hour to hour look like and whether it helps or not get you to where you want to go.

So I like the notion of framing everything that we do in terms of what we're trying to accomplish. I think that is very helpful. And but also recognize that for the patient, it is an experience of-- or and the family, it's an experience of day to day, that it's-- this is-- that reflects what path they chose.

The second thing that I don't think maybe this is the right time, Oggy, to discuss or not, but is the notion that I think Alison brought up which I think is essential, which is the mood-- this might be too light of a word-- but the mood of the intensivist. Is the patient stuck in the trajectory? Is the patient declining? Is the patient getting better?

That sense of where things are at determines sometimes the kinds of conversations we want to have and the kinds of-- and how we frame that conversation. And it may be worth discussing, what is it your sense of trajectory, with particularly the notion of the patient's stuck, the patient that is not going anywhere and the fact that we have to define it. How does that shape the kinds of discussions that we have with patients about what to do?

OGNEN GAJIC: [INAUDIBLE]

ALISON TURNBULL: Can I add one--

OGNEN GAJIC: Yeah, go ahead.

ALISON TURNBULL: Can I add something along that idea of mood? I find in my research, where I'm studying what physicians actually say in the ICU, that the mood that is most troubling or that tends to have the worst impact on conversations are the people who are really on-- the physicians who are really uncomfortable with uncertainty, who feel that until I am certain, I must say nothing or be vague enough that we continue.

And Scott brought up the importance of being transparent about uncertainty. And I couldn't agree more. Saying I'm not sure is really powerful and important, but also not letting that uncertainty prevent us from talking about best guesses around prognosis or making recommendations. Lots of people simply will not go there until they have an extremely high level of certainty.

OGNEN GAJIC: Thank you. So one of the things that has been recommended by most of the guidelines and by experts in the field is so-called "time-limited trials" that you deal with uncertainty. So given that uncertainty, the time-limited trial of some burden, some intervention is something that you can always offer. And that's been recommended, and it makes sense.

I don't know if anyone has some-- the thoughts of that not being useful. If you all agree that time-limit, that concept of time-limited trial is useful one, we can move forward to another topic. Is there anyone who has a problem with what it means, time-limited travel in a-- as an offer, as one of the options with regards to uncertainty?

VICTOR MONTORI: I can try this from a non-intensivist perspective and see what happens with my colleagues and how they respond. So in chronic disease, that's all we do. It's we experiment. And even if we as clinicians don't sanction the experiment, patients are experimenting all the time with trying different things, different times of the day, different skipping different things and so forth.

And I think the reason we tend to do this, the reason we tinker is to learn, as are most experiments, in which we become clearer about what the impact of this change in direction might have on people. And that provides information, and sometimes it reduces the uncertainty as we are trying to sort out what direction to go. So for that, if it's the purpose is to clarify so that patients can and patient clinicians can make better decisions, and they're more information-related, not to patients like this but to this patient, I would think that's a legitimate tool.

OGNEN GAJIC: OK. So we have 13 minutes. And we have some more stuff to cover. So let's say that this is-- we established trust. We do care. We tried to glean at the person behind of what would be potential goals for it.

We have ways of transparently communicating uncertainty and then even ways to deal with it if they're uncertain, that we cannot-- we can try. And then in two days or in a week, we'll be more certain about the chances that this burden will be lower than the potential benefit of what's your goal.

So I want to get to something that I called informed consent. And I think it's important because the burden of what we do in the ICU and-- is not to be neglected. And that's, I think, what is important to communicate to the patient as well. So we have a final goal, but there is also the burdens that you need to go towards the final goal where the goal is.

And then the concept of informed consent, like we do for surgery or chemotherapy, where you understand that there is not only a benefit. So it's not that critical care always saves you. It can actually shorten your life.

Our interventions can make it not just more miserable but also shorter. And being more conservative can allow you to get to that goal. So it's not just matching it to the goals. It's the problem that the way it is, critical care versus no critical care, it's not 100% you have a chance. And you don't do it, you don't have a chance.

Even in respiratory failure and shock and intubation, not everyone who doesn't get intubate dies, not at all. Depending on a condition, it could be up to 80%, 40%, 20% of people who actually go home after expressing the wish not to be intubated.

So how do you present the burdens or the combination of risk and benefit in a sense of informed consent and, I dare to say it, given that the patient is sometimes not fully able to participate, in a clear mind in these situations, even if they are awake and competent, informed assent? Like having an appendectomy on 12 year-old and explaining what it means, you cannot shove it and take them to surgery. You have to get some assent from them. So thinking of [INAUDIBLE] and making informs of risks and burdens, given the uncertainty and the fact that these are not necessarily always beneficial interventions and then informed assent. Scott.

SCOTT ABEREGG: Yeah. Well, informed consent might not-- in many cases might not be aptly named. Whether or not the patient actually does have the information and the understanding to integrate that information into their value systems, that's a huge problem.

And I think the way this-- hopefully this is not anathema, but what I have learned to do over the years is to not be agnostic about what the patient's preferences will be. I begin with the starting point of what my best guess-- I think I'm getting an echo, right? Sorry. I don't know why that started.

OGNEN GAJIC: I think your-- maybe do you have headphones?

SCOTT ABEREGG: I don't.

OGNEN GAJIC: Now we can hear you. OK, go ahead just a bit.

SCOTT ABEREGG: I start with my best guess that a reasonable person under those circumstances would want based on experience, my own, my own values, try to assess how close the patient's values might be to my own or to the average person. And I think doing that is, I think, a good starting point for at least being able to recognize, among all the myriad pieces of data and all of this ocean of data, and facts, and everything that you would have to actually convey to a person to reveal their preferences. I think that starting by just asking yourself the simple question, what do I think that this person or a similar person would need to know in order to approach this decision?

Because there is the issue of information overload, which I have been very guilty of in the past, of overexplaining things or going into too much detail, because there is a limited capacity of all people to be able to accept and integrate a wealth of or a deluge of new information. And you can be sometimes very shocked to have what you think is a very detailed, and cogent, and understandable explanation. And then when you leave the room, you find that all that was remembered was the stuff that was the extraneous details in retrospect.

And so as I said, my approach is to not approach-- I go with an open mind. I do not assume that their preferences are going to be similar to the average person or to my own, but I do begin with a starting point and don't approach it agnostically. And I think that helps me. It's very difficult to describe it.

Intuitively, I think insightful people-- anyway, I hope that I'm insightful-- can during these conversations get clues that maybe this person isn't going to have the average set of preferences. They seem to have hangups about this or that. And so that-- by responding to those intuitive clues, you know to explore the situation further. That sounds all very vague and nebulous. And I apologize, but--