Transcript
Speaker: Welcome to Critical Matters, a sound podcast covering a broad range of topics related to the practice of intensive care medicine.
Speaker: Sound provides comprehensive critical care programs to hospitals across the country.
Speaker: To learn more about our programs and career opportunities, visit www.soundphysicians.com.
Speaker: And now your host, Dr. Sergio Zanotti.
Speaker: Clinicians in the ICU often are required to discuss preference-sensitive care decisions with patient families and surrogates.
Speaker: How options are presented in these discussions and how issues are framed can introduce unrecognized bias and potentially harm patients.
Speaker: Today, we will discuss choice framing in ICU goals of care meetings.
Speaker: Our guest is Dr. Joanna Hart, a pulmonary critical care physician and assistant professor of medicine at the Perlman School of Medicine at the University of Pennsylvania in Philadelphia.
Speaker: She's a core faculty member of the Palliative and Advanced Illness Research Center and affiliated with the Center for Health Incentives and Behavioral Economics.
Speaker: Her research focuses on medical decisions under conditions of uncertainty, particularly preference-sensitive decisions in serious illness care.
Speaker: Joanna, welcome to Critical Matters.
Speaker: Thank you.
Speaker: It's a pleasure to be here.
Speaker: Well, I think that obviously, as you were talking before, this is an exciting topic for me.
Speaker: I think something very relevant to everybody's practice, even if they might not be thinking about it.
Speaker: But my first question is, why should intensivists care about this topic?
Speaker: Well...
Speaker: I think intensivists have a lot of hats that we wear and a lot of roles that we play in clinical care.
Speaker: And one very essential role in clinical care for an intensivist is to guide family members and patients through decisions that are very high stakes, that have a lot of implications for both the patient, obviously, but also the family.
Speaker: And it's really important that clinicians...
Speaker: sort of really assume that role, understanding how to communicate, why they're communicating, what the objectives are.
Speaker: And this topic, choice framing, is sort of essential to how we communicate, especially when we're guiding decision making.
Speaker: And so this is why it's sort of relevant for everyday intensivists.
Speaker: It's something that they're going to encounter and participate in on a regular basis as they help guide
Speaker: family members and patients through these decisions.
Speaker: Perfect.
Speaker: And there's a lot of evidence and a lot of science behind decision making.
Speaker: And before we talk about decision making in the ICU, I would love to get a little bit of more of a maybe a decision science 101 intro for our audience.
Speaker: So my first question is, what is decision architecture and how does it impact our decisions?
Speaker: Yeah, so I actually learned about decision science really, for the most part, from folks across the street at the Wharton School of Business here at the University of Pennsylvania.
Speaker: And I think in learning from those folks, I was really able to understand how consequential choice architecture and decision architecture and some of the science behind decision making that I didn't appreciate as a clinician and had not gotten done.
Speaker: to that point in my sort of clinical training and medical education.
Speaker: And so when we think about the science of decision making and particularly sort of choice architecture, decision architecture, this is really talking about the way that people are sort of moving through the decision making environment.
Speaker: So if you take, for example, when an architect is designing a building,
Speaker: they have to make certain decisions about where things go within that building.
Speaker: So you have to decide where the elevators are going to go, where the stairs are going to go, where the sort of open spaces for people to gather will be, whether or not you have sort of closed offices or sort of more of an open format office, where are you going to locate the bathrooms and the kitchen?
Speaker: And all of those decisions are going, or like choices, are going to guide how people move through that physical space.
Speaker: And so when you think about decision making, it's very similar.
Speaker: So the person setting up the decision for the decision maker is going to have to decide sort of where to put certain things as they're presenting that decision.
Speaker: And that is going to influence how people interact with those choices and ultimately make a decision.
Speaker: And so that's really what sort of choice architecture or decision architecture refers to, sort of the construct and that
Speaker: you know, sort of how that decision is arranged and presented to someone.
Speaker: And one really sort of critical aspect of decision architecture, choice architecture, is that just like in a building, there is no neutral architecture.
Speaker: I mean, architecture has to have some structure that's
Speaker: influence this function.
Speaker: And the same thing is true for choice architecture.
Speaker: So as much as we might say, oh, you know, and I don't actually think th is is true, but I hear this a lot from clinicians, that our job as clinicians is just to present everything in a very neutral fashion and not to influence people.
Speaker: The truth is you can't do that.
Speaker: Anything, any way that you're going to present a decision is going to necessarily influence how a decision maker interacts with the choices, the options, the information.
Speaker: And so it's really important to be thoughtful about how you're doing that, because no matter what you do, you're going to be influencing.
Speaker: And I think that's a super important point, because there's been a lot of also ethical concerns, right, with architecture choice.
Speaker: And when we say, oh, I present things in an unbiased way, we're kind of deluding ourselves.
Speaker: Exactly.
Speaker: You're exactly spot on.
Speaker: And I think instead, rather than trying to kind of aim for neutrality or, you know, aim for absence of influence, I think we would be better served thinking about how are we influencing
Speaker: and doing that in an ethical way.
Speaker: And I think there's, and we might get to this later, but there are sort of real roles for thoughtful choice architecture, for thoughtful supports for decision makers to decrease their cognitive load and help guide them through those decisions.
Speaker: But they have to be aligned with their values.
Speaker: You have to sort of match the influence with sort of what you're trying to accomplish and have
Speaker: sort of very clear sort of boundaries around that influence.
Speaker: And I think if we, you know, if we, like you said, delude ourselves into thinking that we're just trying to, you know, aim for that neutrality, what we've seen and what we'll talk about today is that I think we're actually influencing in ways we don't appreciate or recognize, and we're not being thoughtful about it.
Speaker: And I think when I talk about this with colleagues, one of the points I always make to them is that every day when they are using an EMR, there's choice architecture that's influencing their choices, right?
Speaker: Whether it's something that in an order set is pre-checked, it's much more likely to be utilized.
Speaker: Or if you have a list of options, whatever's on the top is more likely to be utilized.
Speaker: And I think that that seems to be something that we consume every day, yet we believe that when we offer people choices, we're doing it in a very, very balanced way.
Speaker: Absolutely.
Speaker: And these are functions that the health system has borrowed from other sort of more commercial entities because there's a long history of understanding that these functions
Speaker: ways of constructing choices are highly influential.
Speaker: And the health system and the electronic medical record can now harness that knowledge to influence clinicians' behavior towards things that are designed to either improve quality of care, improve safety,
Speaker: decrease costs, improve efficiency.
Speaker: And so they, the, you know, Penn has a whole nudge unit that sort of specifically focuses on how to sort of influence choice through some of these kind of nudges or these, you know, setting up choice architecture that sort of helps people move to making the quote unquote right decision.
Speaker: And much of that is clinician facing.
Speaker: Much of that is embedding some of these nudges and supports into the electronic medical record that change the way that clinicians interact with
Speaker: order sets, as you mentioned.
Speaker: One really early example from the Penn Nudge unit is something as simple as changing the default option to generic away from branded prescriptions.
Speaker: And within that, as soon as it was turned on, clinicians dramatically started selecting the generic option, whereas it wasn't true before.
Speaker: And it saves a tremendous amount of money
Speaker: in the health system to prescribe generics over brand new medications.
Speaker: And so thinking about that as, you know, something that we are, we are sort of the recipients of on a regular basis through the EHR and through sort of health system efforts is something that's just sort of like based into our clinical practice, whether or not we recognize it.
Speaker: Another concept that I think is very relevant to our discussion today, and I would like you to maybe give us a little bit of an overview, is the concept of loss aversion.
Speaker: Yeah, so there are a number of different sort of patterns that we use that are really, people will call them heuristics, they're cognitive shortcuts essentially, and they're part of us as human beings.
Speaker: And so there's sort of a, you know,
Speaker: an understanding that, you know, back in the day, economists and psychologists, you know, we would, we would really more economists would often think of humans as sort of like rational beings.
Speaker: And then, you know, in the sixties and seventies, there was sort of a recognition that
Speaker: humans behaved in very patterned ways when making decisions that suggested maybe they weren't as rational, meaning sometimes they didn't choose the thing that would sort of provide them the most benefit.
Speaker: But because the humans were behaving in these sort of patterned deviations from rationality, what
Speaker: these sort of were recognized as over time with psychologists and economists sort of creating this new behavioral economics sort of school of academic thinking and science is that human beings are sort of irrationally rational.
Speaker: So we use these cognitive shortcuts so that we can move through life more quickly and make decisions.
Speaker: So it's not that they, it's not that they're,
Speaker: it's unhelpful for us to have this sort of pattern deviations from rationality.
Speaker: It's because, and the example I often teach medical students about, so I walk to work and when you walk to work, it's really important to know like what the weather's going to be so that you can bring your umbrella or bring your raincoat out.
Speaker: But if we had to stop and every morning, you know, do a complex, rational decision analysis of whether or not I should take an umbrella that day, I wouldn't be able to move through my life.
Speaker: Like if we did that for every decision and instead I might rely on these cognitive shortcuts to,
Speaker: to assess whether or not, you know, if it rains the day before, I'm probably more likely to take my umbrella because I have recency bias.
Speaker: If I ruined a pair of shoes because I got caught in the rain one day, you know, that might influence my decision sort of next time.
Speaker: And so,
Speaker: Instead of sort of having a stop at the front door where I spend 30 minutes, you know, mapping out sort of a decision analysis to decide whether or not to take my umbrella, you know, I make an in the moment quick decision relying on these cognitive shortcuts.
Speaker: And, you know, when we're recording this, it's, you know, 915 in the morning for me.
Speaker: I've probably already made 100 decisions this morning and I couldn't do a sort of complete rational decision analysis on all of those.
Speaker: I wouldn't be able to get through life.
Speaker: And so instead we use these cognitive shortcuts or heuristics to really quickly move through life and be able to move forward and not spend all of our cognitive load on these decisions.
Speaker: What happens however, is then those, we still bring up with us those sort of cognitive shortcuts, those heuristics when we're making sort of more consequential decisions.
Speaker: And so when you think about something like loss aversion, it's really that losses loom larger for humans.
Speaker: than gains.
Speaker: So as a, and it's not necessarily true for everyone in every circumstance, but for the most part, we display these sort of patterns, deviation from rationality.
Speaker: That might be like, if you think about like a hundred dollars, so a hundred dollars has like an absolute value, but for humans, we will say like losing a hundred dollars, like losing a hundred dollar bill
Speaker: or having a hundred dollars stolen from us or losing a hundred dollars on bed, that is more painful as a pattern than if we were to gain that same hundred dollars.
Speaker: So it's the same absolute value, but gaining a hundred dollars is going to make us sort of less happy than losing a hundred dollars is going to make us sort of sad or unhappy.
Speaker: And so even though that hundred dollars has an absolute value, the loss is more painful than the gain is positive.
Speaker: And so for us, we tend to have something called loss aversion as sort of a human population where we want to avoid losses because we recognize those as painful for us.
Speaker: And so we're more likely to sort of avoid losses than we are to seek out gains, for example.
Speaker: And that's just one example of hundreds of cognitive shortcuts that can often even compete with each other.
Speaker: But it's these cognitive shortcuts that sort of help us move through life and make decisions quickly.
Speaker: But for critical care physicians in particular, it's really important to understand that those cognitive shortcuts and those heuristics can get in the way of decision making, particularly in sort of these heightened environments where people are sleep deprived, there's high emotional stakes.
Speaker: There's sort of life and death decisions.
Speaker: The environment can be traumatic, which sort of changes, all of this sort of changes our brain function and how we process and how we make decisions and actually mean that we may rely very heavily on cognitive shortcuts because our cognitive functions are overloaded.
Speaker: And this is the environment in which we as critical care physicians are helping people make decisions.
Speaker: And I think that when we add what we're going to talk about next is choice framing to those cognitive heuristics.
Speaker: It can create synergy, right?
Speaker: And all of a sudden, a certain presentation might take down a path very quickly for people making decisions that are biased and that might not always be aligned with what the patient would really want.
Speaker: And I think that's why this is so important.
Speaker: Could you tell us, Joanna, what is choice framing?
Speaker: Yeah, so choice framing is really part of that concept of like choice architecture, decision architecture, where the person presenting the decision or the entity presenting this decision or choice, as we talked about, has to make certain decisions about how it's presented.
Speaker: And so those choice frames can be classified.
Speaker: So there's lots of different ways that we can present choices or frame choices.
Speaker: And some of these sort of directly speak to the cognitive biases we know exist for people.
Speaker: These, you know, cognitive shortcuts are heuristics.
Speaker: And so the choice framing is something that we all do all the time when we're presenting choices for people.
Speaker: And we're probably not always aware of sort of the decisions that we as the choice architect...
Speaker: are choosing as we present sort of those options to a patient's family, say.
Speaker: And so how we present it is really the choice framing.
Speaker: And some of those choice framings can reinforce those cognitive biases that people are already bringing to the table when they're meeting with us to make decisions.
Speaker: And so we will go through some examples of those.
Speaker: And we talk about some examples in the paper.
Speaker: It's really imperative to understand that sort of when we are sort of bringing those choice frames to a decision maker, we are sort of again feeding into possible cognitive biases or heuristics that are sort of already present and we are sort of laying a foundation for them to use these when we present decisions in this way.
Speaker: Excellent.
Speaker: So let's move into talking about decision making in the ICU.
Speaker: And you already alluded to some of these, but I would like to reemphasize or if you could recap the challenges that patients, patient families and surrogates face when making important decisions in the ICU.
Speaker: Yeah, so it's a tough environment.
Speaker: And we know it's tough as clinicians, and it's even tougher when it's not
Speaker: you know, part of our sort of professional obligation, but instead something that is happening to a family member or someone that we love or to the patient themselves.
Speaker: And so
Speaker: The intensive care unit exists because there is a threat of death for the patient.
Speaker: And that is a highly emotionally charged situation.
Speaker: Obviously, I don't have to tell them to any intensivists.
Speaker: And we know that family members and patients or other surrogate decision makers are not getting sleep, are having a hard time usually doing just like, you know, daily functions of life and
Speaker: they, when I think about sort of stress and coping, we think about stress as something where a stressor exceeds our capacity to cope.
Speaker: And I think anyone of us who works in an ICU recognizes that the stressors in the ICU often exceed everyone's ability and capacity to cope.
Speaker: And so that changes the way that people are able to make decisions.
Speaker: And so the lack of sleep
Speaker: the emotional charge, the extreme stress that family members and patients are under.
Speaker: And the idea that the ICU is an environment that we know causes
Speaker: primary trauma.
Speaker: So family members and patients after the ICU experienced very high rates of things like complicated grief and post-traumatic stress disorder.
Speaker: And this is something that we as a sort of critical care community increasingly recognize over the past couple of decades and have not yet figured out how to solve it.
Speaker: And part of it's just inherent to the nature of our work and the nature of what the ICU is.
Speaker: But what that means is that us as clinicians also taking on this role of supporting decision-making have to recognize that we are asking people to make highly consequential decisions, highly complex decisions.
Speaker: in terrible decision-making conditions where people are not able to really carry that cognitive load.
Speaker: And it's not about intelligence and it's not about sort of even health literacy or literacy.
Speaker: And that's sort of not what I'm talking about here.
Speaker: I'm talking about just take anyone and put us in that environment.
Speaker: And that is a suboptimal condition for making high quality decisions.
Speaker: And that increases the burden on us as ICU communicators and sort of decision guides, you know, as intensivists to make sure that we protect as much as possible the integrity of that decision making process and figure out how to best support those decision makers in this highly challenging environment.
Speaker: And of course, there's also challenges on the side of the clinicians.
Speaker: Could you mention some of those?
Speaker: Yeah, I mean, we carry the same stress.
Speaker: So I have not yet met an intensivist who has figured out how to exactly cope with all of the stressors of being an intensive care doctor either.
Speaker: I mean, we have our own sort of emotional, I don't want to call it baggage, but we have our own emotional sort of inflections that we bring to the job and the task of sort of guiding decision-making.
Speaker: We as intensivists have to be amazing communicators all the time.
Speaker: We have to build trust with family members in a very rapid period.
Speaker: I was, when I was first going through training as a critical care fellow, I remember just being shocked by the amount of times sometimes where
Speaker: a patient would come in and, you know, sometimes within five, 10, 15 minutes, we had to make very consequential decisions.
Speaker: And in that period of time, we as a team had to establish trust with this person that had never met any of us before.
Speaker: And that is an amazing thing to have to do.
Speaker: It's amazing what to carry for us.
Speaker: And we have to figure out how to establish that trust, how to communicate well.
Speaker: We have to have the time and balance sort of
Speaker: the time spent communicating and making decisions against our other sort of clinical applications and procedures and sort of direct sort of clinical care that obviously decision making is part of.
Speaker: And we have to have the ability to elicit values
Speaker: And when I say values, I mean, some people might say like preferences.
Speaker: And so we have to figure out how to elicit values, which is in and of itself really challenging.
Speaker: And then some people, many people before ending up in sort of the ICU in a decision making role might never have thought through some of the nuances of what their values are with regards to sort of intensive medical care.
Speaker: And we have to understand and have a good command, obviously, of all of the available and relevant medical options, synthesize a whole lot of clinical data in order to sort of guide decision makers, patients, family members through sort of these options.
Speaker: And so it's a hard job for us to not just in sort of the medical capacity, but also as communicators trying to integrate all of these various competing demands.
Speaker: For sure.
Speaker: And you published recently in Critical Care Medicine a wonderful paper, Clinicians Use of Choice Framing in ICU Family Meetings.
Speaker: And I would like to go over that paper in a little bit more detail, but mostly as a roadmap to talk about this topic, because I think that obviously the results are very important, but also you introduce a lot of concepts that many intensivists might not be very familiar with.
Speaker: So my first question, Joanna, is what was the central question of your study?
Speaker: Yeah, so I mean, really the question of our study was how are clinicians presenting choices to family members of ICU patients?
Speaker: And the motivator behind this was really around this idea that integrating decision science and what we know about
Speaker: decision science and decision making and choice architecture, we know that these things are highly influential.
Speaker: We had done some work a few years ago that demonstrated that clinicians lack sort of competency in predicting the outcomes of these various ways of presenting decisions.
Speaker: And so we were interested to understand then
Speaker: which types of choice framing or choice architecture are clinicians actually using when they're talking to family members, making these decisions.
Speaker: And more than that, not only how they were using them, but also both the frequency of sort of how they use these choice frames, but also patterns across the different types of decisions that intensivists present to patients' families.
Speaker: Perfect.
Speaker: And what did you do in the study to actually answer the question?
Speaker: Yeah, so this was a very labor intensive project.
Speaker: So we were really fortunate to receive 101 transcripts of family meetings.
Speaker: So these were real transcripts.
Speaker: real family meetings that were happening from 2009 to 2012 that had been recorded and transcribed by some of my co-authors had led that work.
Speaker: And we took advantage of the ability to use that data to look at this question in particular.
Speaker: And so we had de-identified transcripts of these family meetings that, you know, covered a whole range of topics.
Speaker: And the first thing that we did is we said, well, we need to figure out
Speaker: which decisions are preference sensitive or value sensitive.
Speaker: And preference sensitive decisions are the ones that you need to know something about the patient and the family in order to understand the sort of quote unquote right choice.
Speaker: So for example, whether or not to put someone on a ventilator
Speaker: has a couple of sort of things involved in that.
Speaker: First, to know if that's the right decision, it has to clinically make sense.
Speaker: And that's sort of on us to identify whether or not it clinically makes sense to involve sort of a mechanical ventilator to support that person.
Speaker: But you also have to understand whether or not that person is aligned with that person's values to introduce this form of life support.
Speaker: Whether or not to treat a pneumonia in someone is a sort of value sensitive or preference sensitive decision.
Speaker: It depends on sort of what else is happening with that person, what their sort of values are for medical care and their sort of quote unquote goals of medical care might be.
Speaker: And that part is sort of preference sensitive, but sort of which antibiotic to use is no longer, that's not a preference sensitive decision in sort of the vast majority of the time that's based on sort of, you know, clinical information.
Speaker: And so we looked to the literature to identify sort of what decisions in an ICU are sort of standard preference sensitive or value sensitive decisions.
Speaker: And so we looked to some work that had been done by Allison Turnbull and colleagues that had predefined this based on sort of previous work that they had done.
Speaker: And so we took their list.
Speaker: And we added sort of if there was some other topic that was not pre-specified on this list of preference sensitive ICU decisions, but was sort of clearly presented as a preference sensitive decision.
Speaker: And so the types of things that would wind up on that sort of preference sensitive critical care decision topics are things like mechanical ventilation,
Speaker: resuscitation, meaning CPR, chest compressions, code status, what happens to the patient after the ICU, so discharge to a location other than sort of going home, comfort-oriented care and hospice, sort of feeding tubes,
Speaker: dialysis and other forms of renal replacement, urinary catheters, venous catheters, pulmonary artery catheters, and then any sort of, you know, cervical intervention or procedure that you typically have to, you know, have some kind of informed consent for.
Speaker: And so we went through these 101 transcripts and identified where all of these preference-sensitive decisions showed up.
Speaker: And so that was sort of our first, is to sort of narrow...
Speaker: the amount of data we were dealing with down just to the preference sensitive decision so that we could take a deeper dive into those decisions.
Speaker: And so then for the second phase, now that we had sort of these preference sensitive decisions, we then went through and we had, you know, four coders of four research staff on the team who went through and we identified where these sort of a priori choice frames that we had identified from the decision science literature,
Speaker: where we were able to identify those happening.
Speaker: And so we went through and combed through and read very carefully all of those preference sensitive decisions and identified where these sort of pre-specified choice frames were appearing.
Speaker: And we did this both from a quantitative standpoint and then also, you know, obviously this was qualitative analysis, content analysis of these, um, you know, transcriptions.
Speaker: but we were quantifying it, sort of putting it into, sort of grouping them into buckets based on, you know, what type of choice frame was used to discuss these various sort of decision topics.
Speaker: And then once we had sort of, you know, bucketed all of the data that we had, we then looked at sort of patterns across the different decision types and across the different types of choice frames to understand sort of where the
Speaker: where clinicians were using particular choice frames more.
Speaker: Um, and we'll talk, I think a little bit more about that.
Speaker: Could you give us a little bit of an overview of some of the choice frame and their definitions?
Speaker: Yeah, absolutely.
Speaker: So, um, one of the ones that might be familiar to a lot of clinicians, for example, is, um, a default option.
Speaker: So a default option means that just as it might sound,
Speaker: Whatever option is presented will be selected or continued if you don't actively choose another option.
Speaker: So if I say X is going to happen, unless you say, no, we're not going to do X, we're going to do Y, we will continue on with X. Another sort of related choice frame is a polar interrogative, which is just really a fancy way of saying a yes or no question.
Speaker: And so this is if I was going to say to you,
Speaker: I'm going to give you X as an option and you can say yes or no to that.
Speaker: And so do you want us to continue with dialysis, your options to say yes or your options to say no?
Speaker: And we'll talk maybe a little bit more about why that particular choice frame is very similar to a default option.
Speaker: Another is to have two options.
Speaker: So you present X and you present Y. You very explicitly allow someone to make a decision between X and Y.
Speaker: Multiple options is very similar.
Speaker: You just add additional options at that point.
Speaker: It is worth noting that multiple options can be overwhelming to people.
Speaker: So maybe a choice set of
Speaker: Three is something that people can wrap their heads around.
Speaker: But once you start getting out to choice sets or options that include, you know, four, five, six, seven, eight options, that then also becomes overwhelming and makes it harder for people to make decisions.
Speaker: So there is a little bit of a sweet spot in some of these choice frames in terms of how they're going to influence decision making.
Speaker: Another choice frame we identified was open-ended, probably familiar to most people, you know, open-ended questions where someone is presenting questions about a specific choice and requires sort of an elaborated answer, can't answer it with a yes or no.
Speaker: And then we get into some that are a little bit more nuanced.
Speaker: So a choice frame that involves commission is a presentation of option that favors action rather than inaction.
Speaker: So this would be a bias towards sort of doing something as opposed to not doing something, which you can frame the same option either way, either as commission or as its alternative, which is omission, which is presentation favoring inaction rather than action.
Speaker: So it might still be a choice between action and inaction, but depending on how you frame it, you might be framing it as sort of commission or framing it as omission.
Speaker: Endowment is also one that I first learned about from, you know, decision science.
Speaker: And this is a presentation where you're going to frame an option for someone or an intervention
Speaker: as sort of being used currently or like quote unquote owned by a patient.
Speaker: And so the choice would be to like take away that from the patient.
Speaker: And the classic sort of decision science example, not in medicine is like, let's say there's a mug.
Speaker: So I know endowment can be a little hard to understand, but and how it might influence people.
Speaker: So let's say there's a mug, a coffee mug, and I might be willing to pay $5 for this mug and then the mug's mine.
Speaker: But now if you're going to say, oh, I'd actually really like that mug.
Speaker: Now that I own this mug, it's now worth more than $5 to me.
Speaker: The mug has an absolute value maybe of $5, but now that I own it, it's now taken on more value and I'm not going to give it up unless you give me 10.
Speaker: It's not about making a profit for me.
Speaker: It's really about sort of, I now value this more because I have it.
Speaker: And so a choice frame that includes sort of this endowment effect would be one where
Speaker: The clinicians framing it as something that the patient has that we're not going to take away, which means that it's harder to do that for a decision maker to give that up if it's quote unquote owned by the patient now.
Speaker: And then gain, loss and mixed frames are really about whether or not you're emphasizing the positive outcomes from doing a particular option.
Speaker: or the negative outcomes from sort of avoiding that option.
Speaker: And then mixed frames are when you use both gain frames and loss frames.
Speaker: So you're emphasizing both the sort of positive outcomes and the negative outcomes.
Speaker: And then a clinician, yeah, go ahead.
Speaker: Go ahead, sorry.
Speaker: And then the last two that we looked for were clinician recommendations and social norming.
Speaker: And those are probably pretty familiar to most listeners.
Speaker: Clinician recommendation, I think we sort of understand what that is.
Speaker: And then social norming is really a presentation that emphasizes how sort of like peer comparisons or how other people would behave or choose.
Speaker: And social norming is something we, in our previous work, we know clinicians sort of recognize the influence and effect of that.
Speaker: Absolutely.
Speaker: And I think it's fascinating, Joanna, how all these definitions, right, are when you hear them are very precise.
Speaker: We recognize them, yet in the midst of a conversation, they're not on our radar, right?
Speaker: I mean, we're not thinking I'm going to use a default framing right now or, oh, I just used a positive, a gain framing, right?
Speaker: And yet this has been well described.
Speaker: And I think that's one of the beautiful things of your studies that you were able to really
Speaker: match these to what people do in real life and see what happens more frequently in ICUs or at least in the ICU that you studied.
Speaker: So could you tell us what you found, the results?
Speaker: Yeah, absolutely.
Speaker: So we found that we had
Speaker: Across these 101 transcripts, we had 202 decisions that we were able to classify and look at.
Speaker: And it was the most common topics were the ones that are probably familiar to most intensivists.
Speaker: So mechanical ventilation, overall goals of care, dialysis, sort of post-ICU discharge plans were sort of the ones that came up most frequently, but we saw all types of decisions.
Speaker: And when we looked at classifications,
Speaker: What sorts of choice frames were used most commonly?
Speaker: The default and polar interrogative, which remember is the yes, no question.
Speaker: Those are the two choice frames that were used sort of overwhelmingly across multiple decision types.
Speaker: We also found that, and I'll talk a little bit more about that in just a moment.
Speaker: We also found that endowment was used quite frequently.
Speaker: And then most clinicians across the decision episodes use sort of mixed loss and gain framing.
Speaker: And so when we go back and think about sort of default framing in particular,
Speaker: This is sort of the main takeaway that I had from the study is how often default options were used when talking about these preference sensitive decisions and how although we think of default as a nudge, in this case, I'm not convinced that our use of defaults here, and I say our meaning, you know, obviously I wasn't one of the clinicians in the study, but I think, you know, we as a community probably follow similar patterns.
Speaker: is that that use of default is when a clinician presents that specific option to a family member or a surrogate decision maker.
Speaker: And I'm not sure we meet the burden of ethical use of default framing because part of what should happen when you present a default is that the person understands that there are other options.
Speaker: Well, first that there's a decision being made at all
Speaker: The second, that there are other options.
Speaker: And third, how those options can be selected, meaning how can you deviate from the default and pick something else.
Speaker: And in most of the decision episodes that we looked at, clinicians were not always explicit that there was a decision being made at all.
Speaker: So sometimes we were.
Speaker: So there was one example that we include in the paper where the clinician says,
Speaker: sometimes we take the tube out and here he's talking about endotracheal tube.
Speaker: Sometimes we take the tube out and people don't do well, we put it right back in, you know, that's our default.
Speaker: And so here he's saying, you know, this is a default, meaning there could be other options that like we're not talking about here, but at least it signals for the family member that there's, there's a decision being made that there are other options, but this is sort of what we would typically do.
Speaker: But most of the time,
Speaker: clinicians did not explicitly indicate that this was a default option.
Speaker: So it was something more like the clinician saying, if we're able to get the patient off the ventilator, then either his kidneys will continue to improve.
Speaker: So he says the kidney issue will either come to a head by then and he'll either be better or he'll be on dialysis.
Speaker: And so...
Speaker: There's no, like dialysis there is the default option.
Speaker: The family member is not presented with the idea that there could be other viable options that that family member might select that would potentially be more value aligned with the patient's goals.
Speaker: And that was how overwhelmingly clinicians used default framing.
Speaker: And the polar interrogative is interesting because it's very similar to the default, meaning there is a single option that the family member can either accept or reject by saying yes or no.
Speaker: And it helps...
Speaker: a little bit because it shows that there is a decision being made.
Speaker: So it signals to the family member, hey, like we're making a decision here about, you know, whether or not to do dialysis.
Speaker: But if you only present dialysis and then the family member can accept it or reject it, it doesn't present a sort of full choice set or at least any other choice set and how the family member might be able to opt for something else or what should go into that decision.
Speaker: And so some of the
Speaker: some of it was the family members actually pushed back and said, well, what's the alternative?
Speaker: If I say no to this, then what happens?
Speaker: And so that, I think, signals that family members are really trying to understand what is the choice set when they're given a single option to accept or reject.
Speaker: And I think for default and polar interrogatives, one thing that's really important to recognize, too, is that
Speaker: Clinicians wield a lot of power in these circumstances.
Speaker: And there are a lot of, there's a lot of sort of social and conversational norms that are happening here too.
Speaker: And so if you're presenting a default or a single option to accept or reject to a family member, it can be hard for them to
Speaker: sort of disagree because being sort of non-compliant, and by that I mean sort of that social norming, I don't necessarily mean like medical adherence, but to be sort of non-compliant or disagreeable would be to sort of try to deviate from the default or say no.
Speaker: And we also know from the decision science literature that most people receiving a default will assume that the
Speaker: that the choice architect, that the person presenting that default to them is recommending that option.
Speaker: And that might be true in many circumstances for intensivists, but I'm not sure it's always true.
Speaker: I'm not sure that intensivists always mean that that default and have thought clearly through, this is my recommendation.
Speaker: There may be lots of other things happening for an intensivist that lead them to present that default, even if it's not sort of an explicit clinical recommendation.
Speaker: But it's important for us as communicators to understand that family members are likely reading into that, that
Speaker: whatever option you've presented as the default or the, you know, polar interrogative, the yes, no option is what you recommend.
Speaker: And I think also that a lot of the interventions that we, we, we, um, we provide at the bedside, uh, our default set, right?
Speaker: So if somebody is crashing and you don't know, you go ahead and intubate.
Speaker: If somebody is crashing and somebody codes, you do CPR.
Speaker: right and I think it just kind of pushes us in that direction and like you said I mean we might not be aware but it definitely has a tremendous impact on how people perceive and I guess to some extent it's just kind of like the law of inertia and physics right if something is there already it's a lot harder to change that that course or decision or patients families might not understand that that's even an option.
Speaker: Yeah, exactly.
Speaker: And there's some really cool work being done on that idea of clinical inertia and how do you stop the train once it's started and recognizing that
Speaker: you know, every day is a new opportunity to make a new decision, but I think it often doesn't feel that way.
Speaker: And there's a lot of, one sort of cognitive bias we don't talk about in this paper is like some cost bias, this idea, well, we've done so much already, we might as well just keep going.
Speaker: And I think when we looked at these decisions, we found that
Speaker: you know, clinicians were doing a lot of things well.
Speaker: So when they specifically were trying to elicit or clarify patients' values, they often use open-ended questions, which is, you know, what we're, I think, trained to do and what feels right to do is to ask these open-ended questions where we're eliciting values, trying to understand more about the patient to help match ICU care with these sort of more specific decisions.
Speaker: But we don't use them.
Speaker: Clinicians do not really use that much, these open-ended questions, except in these kind of more broad topics like overall goals of care, end-of-life care, you know, prolonged dependency on medical support.
Speaker: And I think we could do a better job of integrating those open-ended questions and values elicitation into these kind of more specific preference-sensitive topics.
Speaker: life support, major medical intervention decisions where, you know, we're not really doing that.
Speaker: And I think that even though there's no evidence to support it, just from my personal experience, in any difficult conversation, the more curious you truly are to learn and offering open-ended questions always helps.
Speaker: helps understanding, helps rapport, helps alignment.
Speaker: And I do believe that you're right.
Speaker: I mean, we should listen more and perhaps talk a little bit less in these end of life or these goals of care meetings so that we can really appreciate what are the real preferences that that patient probably has.
Speaker: I think you're absolutely right.
Speaker: And I think, you know,
Speaker: when you talk about sort of that curiosity and, you know, really trying to understand the perspective of the people affected by sort of the medical crisis, I think part of that is, you know, really establishing trust and having sort of authentic empathy.
Speaker: And I think
Speaker: I agree with you, I think that goes a long way, not only to make that family and that patient both feel supported and actually be supported, but also allows us to gain sort of insight into how to match the critical care that we're providing with like what makes sense for that person.
Speaker: And that person and that family are the ones who like really know what makes sense for them.
Speaker: And we're trying to puzzle piece it together as their clinician.
Speaker: Yes.
Speaker: I wanted to ask you, Joanne, about choice sets with two or more options and gain and loss framing.
Speaker: I know that those were far less frequent than the default or the polar interrogative, but they were also present, right?
Speaker: Yes, they were.
Speaker: So clinicians use choice sets with two or more options.
Speaker: Again, kind of talking about overall goals of care, resuscitation options and, you know, post discharge plans.
Speaker: So, you know, SNF versus LTAC versus home, things like that.
Speaker: And so, well, they didn't use, like the use of these explicit choice sets were fairly rare, but when they used them, clinicians really sort of highlighted the options available to patients given sort of whatever clinical situation it was.
Speaker: And so it allowed patients
Speaker: Just the presentation of a choice set seemed to lead directly into, you know, like how to choose between those things in a way that default options and those polar interrogative sort of necessarily did not.
Speaker: And so clinicians, I think even just presenting a choice set is saying here are two options.
Speaker: I think it prompted clinicians to sort of have a more detailed discussion about how someone functionally makes a decision between those two options and led to more of that values elicitation.
Speaker: And with the gain and loss framing, I think it's also important, right?
Speaker: Because we frame things more in a loss perspective with loss aversion that might push people in that direction.
Speaker: And you can be saying the same thing from a gains perspective.
Speaker: It might not have the same impact.
Speaker: Can you talk a little bit about how they use gain and loss?
Speaker: Yeah, absolutely.
Speaker: So clinicians were actually quite balanced in how they presented gain and loss framing.
Speaker: So most of the time they did not present only loss framing or only gain framing.
Speaker: They presented them sort of mixed.
Speaker: So both of them together.
Speaker: And I suspect one of the reasons that clinicians did that, because they did it across all types of decisions, is that this...
Speaker: I think felt very similar to like an informed consent process where you talk about like risks and benefits of any particular intervention.
Speaker: And so I think probably because of that sort of more traditional training around how we present informed consent, clinicians were actually very balanced in their use of gain and loss framing and typically used both in a single decision episode.
Speaker: Well, I believe that ultimately the goal of research is action, not only knowledge.
Speaker: So if you had to put things all together from what you've learned with this paper and all the other research that you've done in terms of applying it at the bedside, what are some common pitfalls that you recommend we avoid and goals of care conversations and meetings in the ICU?
Speaker: Yeah, I think for me, the biggest takeaway here is trying to avoid
Speaker: defaults and yes, no options for our ICU decision makers, meaning patients and family members making these preference sensitive decisions.
Speaker: And I think it's very in some, I think most of the time it's actually important for clinicians to make recommendations if that's helpful to the family.
Speaker: And I think in my experience, a lot of the times it is helpful to make recommendations
Speaker: But I think those recommendations need to be based on first understanding sort of the values of the patient.
Speaker: And I think one of the reasons why defaults in polar interrogatives or those yes, no questions can be harmful is that they don't give us that pause point to...
Speaker: say to ourselves, okay, I am making a recommendation here.
Speaker: And I think it's a little bit of a easy way out to give a default option.
Speaker: It's like, well, I gave them the choice, but I'm not sure that that's actually clear that the patient and the family member are receiving a choice or feeling as though they're participating in a preference or value sensitive decision.
Speaker: So I would say in general, in clinical practice, I would try to avoid using the defaults unless it's very explicit and you're at your
Speaker: presenting sort of what the non-default would be.
Speaker: And instead of presenting it as a default or a yes or no question, instead present it as a recommendation.
Speaker: Like if that's your recommendation, present it as a recommendation to explain why and what the alternative is and have sort of a discussion, a collaborative conversation with the family and avoid sort of that single option for these, you know, known to be preference or value sensitive decisions and interventions.
Speaker: I also think it comes back to sort of what we
Speaker: have already talked about around ask the open-ended questions, get to know, you know, what's important to them, what makes life worth living, what are sort of the essential features of life for that person.
Speaker: And, you
Speaker: I think step back for a moment and recognize how challenging that decision making context is for patients and families, which is why sometimes a recommendation can be really important coming from a clinician who has the ability to kind of sometimes see a bigger picture, understand the patterns of critical illness.
Speaker: and be the one who helps link the patient's values and what makes life worth living and what the essential features of life are with what we think is likely to clinically transpire given a particular option.
Speaker: And so I think it's totally ethically justified to make recommendations and support decision-making, but it has to be thoughtful and it can't be just sort of a knee-jerk way of communicating.
Speaker: And I think that the importance there also, Joanna, seems to be that we should be making recommendations that align with the preferences that we understand the patient has based on what the family has shared or the patient has shared with us and explaining why those recommendations align with those goals.
Speaker: I think that is often where I think clinicians make a mistake is they think of what they think should be done as opposed to what are the options that align with the goals of our patient.
Speaker: Absolutely.
Speaker: I couldn't agree more.
Speaker: So in addition to using, and the other point I wanted to make before we leave the pitfalls totally is that the default framing and the yes or no questions was 80% of the time, right?
Speaker: So this is, I mean, that's the pitfall.
Speaker: And what this is telling us is that we are,
Speaker: falling in pitfalls the vast majority of the time.
Speaker: So I think something for our listeners to pay attention to because the vast majority of the time we're utilizing choice frames that probably potentially could cause harm or not the best way of presenting information for these complex decisions.
Speaker: So I think that's an important point.
Speaker: You did talk about open-ended questions.
Speaker: Are there any additional pearls that you would give for conducting better family meetings using decision science tools?
Speaker: I think the other is to recognize that a lot of times when, and this might be a little bit outside of this particular paper, but I think as clinicians, we should also recognize that cognitive shortcuts, heuristics, cognitive biases, whatever term you want to use, when we see those displayed by family members, I think it's important to understand that that is a coping mechanism for the
Speaker: overwhelming challenge of making decisions in the context that they are, you know, forced to make decisions.
Speaker: And that the emotional strain and challenge with brain processing and sort of just cognitive functioning because of the ICU environment, the threat to mortality and so forth that we already talked about, that the demonstration of those cognitive biases is not a
Speaker: failure of the family member to be reasonable or to get it or all the kind of phrases that we use to describe sometimes when we see those cognitive biases come out and recognize that that is like an evolutionary feature of us.
Speaker: That that is a way that we as human beings cope with decision making stress is to reduce our decision making load and reduce our cognitive burden by using these cognitive shortcuts
Speaker: And we as clinicians have to help them overcome them if we think they're harmful, if we think that they are counterproductive in this circumstance.
Speaker: But if we don't recognize those cognitive shortcuts or those heuristics as sort of a feature of human beings, then we rely on, oh, this family's just unreasonable.
Speaker: And I think that's the wrong way to approach it as clinicians, guiding people through these really complex decisions in really poor circumstances.
Speaker: I agree.
Speaker: And I think it summarizes something that I often say and think that in the ICU, it's not usually that we have difficult families.
Speaker: We have families in difficult positions and we should be a little bit more empathetic with that.
Speaker: Yeah.
Speaker: So it's been a fascinating conversation, definitely learned a lot.
Speaker: I find it very interesting that it was a pair of psychologists who started the whole behavioral economics movement, and now we're talking about applying it to critical illness.
Speaker: So I think it just shows you how interconnected disciplines are and how
Speaker: we should always be expanding our horizons and looking for innovation or better ways of doing things outside of our little niche in the ICU.
Speaker: So to close the podcast, Joanna, we usually like to ask our guests a couple of questions unrelated to the clinical topic.
Speaker: Would that be okay?
Speaker: Absolutely.
Speaker: So the first question relates to books.
Speaker: Is there any book or books that have influenced you significantly or a book that you have gifted often to other people?
Speaker: So I'm a big believer in reading.
Speaker: And I read nonfiction, but I actually read mostly fiction.
Speaker: And I believe a lot in the value of reading fiction.
Speaker: I think it helps us, and research actually does show this too, that it helps us become more empathetic.
Speaker: And one book that I would recommend and that I've given to trainees and friends and colleagues is Demon Copperhead by Barbara Kingsolver.
Speaker: It's a long book, but it's a very fast read and it incorporates a lot of aspects of
Speaker: including opioid use disorder and caregiving and health equity.
Speaker: And so I think it's a really fascinating read and a very captivating read and a great work of fiction that I would recommend.
Speaker: Perfect.
Speaker: We'll definitely link it in the show notes.
Speaker: And I have not read it, so I definitely will pick it up.
Speaker: The second question is, could you share something you changed your mind about over the last couple of years?
Speaker: Yeah, so I...
Speaker: You know, I wondered whether or not I should give, you know, an academic answer to this.
Speaker: But to be honest, the thing that popped into my head was, you know, I grew up
Speaker: with a brother who played baseball.
Speaker: And I thought baseball was one of the more boring games that existed.
Speaker: And now I have an 11-year-old son.
Speaker: And in the last few years, I've been convinced.
Speaker: And I think all of us out there who are parents with kids transitioning into their tween and teen years know that you've got to follow their interests.
Speaker: And so the last few years, I've gone all in on Phillies baseball.
Speaker: And last night, I even played in the...
Speaker: that every season the last practice is a parent versus kid baseball game.
Speaker: And I got on baseball times I batted.
Speaker: So I'm very proud of myself.
Speaker: And that's the thing I probably most convincingly changed my mind about in the last few years.
Speaker: Well, that is awesome.
Speaker: And I'm just laughing because I think we share that.
Speaker: I grew up in South America.
Speaker: My family's from Italy.
Speaker: And I still have a hard time explaining my dad why I enjoy baseball so much.
Speaker: Like I'll comment, I mean, I went to a game or talk about the World Series and he does not get it.
Speaker: He just does not get it.
Speaker: Perfect.
Speaker: So the last question is, what would you want every intensivist listening to us to know could be a quote or a fact?
Speaker: Yeah, so I think one thing, one fact that I find really, really compelling is around family caregivers.
Speaker: So family caregivers are the single largest workforce in health care.
Speaker: And they provide 600 billion with a B dollars of uncompensated healthcare labor in the U S every year.
Speaker: And I think recognizing the importance of family members in
Speaker: Not only caring for their loved ones, but also just maintaining our health system is something that I think our health system and intensivists should increasingly recognize and support their roles in our patients' lives and in our lives as clinicians.
Speaker: I think that's a perfect place to stop.
Speaker: Joanna, thank you so much for sharing your expertise and being so generous with your time.
Speaker: I look forward to having you back on the podcast to talk about this and other critical care related topics.
Speaker: Thank you.
Speaker: This was wonderful.
Speaker: Thank you for listening to Critical Matters, a sound podcast.
Speaker: Make sure to subscribe to Critical Matters on Apple or Google Podcasts and share with your network.
Speaker: Sound's transforming the way critical care is provided in hospitals across the country.
Speaker: To learn more, visit www.soundphysicians.com.
