It's my absolute pleasure to welcome everyone back from the holidays and invite today our speaker, Dr. Kruser. So Dr. Kruser is an assistant professor of medicine at the University of Wisconsin, as you probably have heard us talking about. And I got interested in her work and started reading some of the things she was publishing about the role of language in an ICU system. And I just thought this was really interesting, some things I wasn't necessarily thinking about. And so I'm really happy to have her here today to share some of her work and her thoughts on this. And so without further ado, Dr. Kruser, I will hand it over to you. Thank you so much for being here with us on this Friday afternoon.
Great. Thank you so much, so much for the invitation and the opportunity to talk with you today. And hopefully we will be able to sort of engage in some discussion and I'll hopefully save a lot of time at the end. And I hope to be somewhat provocative here as we're going through the talk today. So I have no financial conflicts of interest to disclose. I would like to acknowledge that some of the work I'll be presenting today has been funded and resourced by
the National Institutes of Health and also the University of Wisconsin and Northwestern. I'd like to make a couple of self-disclosures as well to start so you understand my perspective and where I'm coming from. So the work I'll be sharing today is strongly influenced by my experiences and perhaps sort of more often than I'd like to admit some of my linguistic missteps as I practice. So I'm a critical care physician and I practice primarily in a medical ICU and my clinical practice and the work I'll be sharing is for the most part from the academic medical center context. So I'd like you to keep that in mind as sort of my lens and my lens into the reflexivity of the work that we'll be presenting.
And I'd also like to disclose, and perhaps this will make a little bit more sense to you as I go on, that I used to use the term need and the phrase goals of care all the time in my clinical practice and in my role as a health services researcher. And I study the interface in the ICU between life extension and medical technology and end of life. And so these were common parts of my language. And I hope to at least give you the rationale and perhaps convince you by the end of today about some of the reasons why I now try to avoid this language most of the time. And so I'm going to really focus on these two specific parts that I call parts of our default language in the ICU. First, need. And I'd like to reconsider with you the language of need and how it's used in critical care. And then I'm going to pivot and discuss with you attention that our work has begun to expose within the phrase goals of care. And again, specifically how that plays out in the intensive care unit. And sort of within these two sections of the talk today, I'm going to be making a sort of bigger argument.
and perhaps try to convince you that I think that this focus of language and the ability to examine it carefully is not just about banning words or sort of policing and patrolling the language that we use and sort of narrowly defining things and being precise in how we communicate with each other and with patients and families, but really that this careful examination of our default language is a unique way to study and to try to uncover some of the latent and really embedded features of our ICU care processes that are so... common for us and so part of our day-to-day work that are often hard to recognize but largely influence the way in which care tends to unfold for our patients. And I'm going to suggest to you with some of our data that these embedded ICU care processes that we can start to examine through our default language often bypass moments that are truly well suited for deliberation and for thoughtful choice with our patients and their families.
So let's start with need. So to need is to lack something essential. And so for patients with acute serious illness, almost by definition, our patients in the ICU lack something essential to sustain their life. This is often how we justify the rationale to admit a patient to the ICU. They need something, some form of medical technology that only we can provide in our highly specialized contexts in a really unique environment within the hospital. So it's really unsurprising to me, at least, that the word need rolls off our tongues every day as clinicians in the ICU. If her breathing gets any worse, she will need to be intubated.
He needs a central line, a special IV catheter in his neck, so we can give him blood pressure medicines. If she doesn't make any urine soon, she'll need dialysis. And if she can't be excavated soon, she will need a trach. But the way we use this term can be misleading, especially when we talk with critically ill patients and their families. And I'd like to illustrate this point with a story of a patient that we learned about during one of our research studies. We were interviewing surrogate decision makers of patients who had previously been in the intensive care unit and had received mechanical ventilation. And we were trying to gather their perspectives and learn about their experiences in those decision making moments, or at least the processes of care that unfolded that resulted in the patient.
receiving mechanical ventilation. And so the daughter of a patient who was being cared for on a cancer care floor told us that early in the morning, the physician called and said, either your mom needs to be on a ventilator or she's going to die today. So this patient's daughter during the interview went on to tell us about her mother. And she said, my mom kept fighting. She was never going to be put on a ventilator. That is the worst thing that could possibly happen to her. But the daughter went on then to tell us that in fact, she, in conjunction with the patient's ICU clinicians, ultimately decided to intubate and start mechanical ventilation for her mother. And she said, I trusted the doctors much more than my mom.
And so let's step back a little bit and dial in what are we talking about here when we use the word need and why are we focused on this particular phrase? So need isn't simply used in our language. in contemporary language to state a fact about what is or isn't lacking. So let's take in a very simple example from a non-medical context. So here's a fact. This figure does not have any water, so he doesn't have water. And we can make this statement, and outside of any particular context, the statement doesn't make much sense, or at least doesn't imply a particular course of action ought to be done. But if we say this person needs water, a very similar statement, but a different verb, takes on a whole new meaning. It's a call to act, and not only is it a call to action, but the use of this term need and the way it's used in a sentence.
conveys the specific action to be taken. Get this patient some water or get this person some water, this stick figure. And so really the crux of our argument and our concern about the use of the word need is that it creates an inextricable link between the problem at hand and our available medical solution. So as clinicians, we think about this type of link first. In fact, this is one of the ways we're educated through our training. Does this patient have the type of respiratory failure that is severe enough to need mechanical ventilation and intubation? Then we take a step back in the clinical context to consider whether we should intubate.
And we may be assuming that patients and families can follow our logic, but after hearing that a patient needs to be intubated, they might be assuming that it's the right thing to do. After all, who wouldn't want to give their loved one something that is needed? So when we set up and start a conversation with a family or start an ICU stay with a statement that a patient needs a particular life-sustaining treatment, and then we try to walk back and deliberate with a patient or family about whether it should be done, it isn't surprising, at least to me, that we often end up in a space of conflict, and I worry that this is perhaps perpetuating some issues of mistrust. And so we proposed an alternative framing for this exact same clinical scenario. So instead of saying the patient's breathing is getting worse, she needs to be intubated, we've suggested two alternative phrases that make space and break up this inextricable link. So the problem with her breathing is getting worse. Can we talk about what this means?
So what this means is a signal to patients and their families and truthfully is a reminder to me as a clinician that this is a big change in health, that this is going to have a major impact on the patient now and in the future, even if it's undoubtedly the best step to pursue life support for this patient. It's a moment that we can explicitly summarize where the patient is, the illness and its trajectory. And I think that this understanding and this moment for sort of brief summary is really crucial for all critically ill patients. And then I think in addition, this type of framing creates space. to attend to patients' and families' emotions in distress before immediately jumping to solutions and actions. So here in the case, the example of the patient with cancer that I told you about where her daughter received a phone call that said your mother needs to be intubated or she's going to die, you might say, the infection in your lungs, what this means is the infection in your mother's lungs is very severe. Her cancer treatment is making it hard for her immune system to control it. We don't know yet if antibiotics will help, and I'm concerned that she could die. I'm worried.
What to do next frames separately then an opportunity for deliberation. It's open-ended instead of a predetermined need, and it creates space to discuss what's acceptable to and prioritized by the patient. And I know that there's often time pressure in the ICU setting, and so what is said here is always tailored to the specific patient situation, and it's sometimes very brief. So let's talk about what to do next. We can intubate her and provide mechanical ventilation, a form of life support, sometimes called a breathing machine. This will give us time to see if the antibiotics are helping. For some patients, though, being on life support like a breathing machine is not okay.
Another option we have is to provide care that focuses exclusively on her comfort, knowing she may die from this infection. And so I want to take a step back now and sort of allude to that argument that I started the talk saying is that focusing on language like need is not just about sort of striking and redefining and reformulating the words that we use, but it's actually a lens into some of the broader system level embedded processes that are so common and so everyday for us that sometimes it's hard to recognize when there are influences at play. And so this is a quote from a medical anthropologist, Sharon Kaufman, who had a very influential career at UCSF studying end of life care in the American health care system.
And what she said is that patients get put on a train of interventions that is very difficult to stop. Diagnostic tests confirm the need for interventions, and then procedures become appropriate by default in this organizational scheme. And others have sort of described this particular phenomenon in care. In this case, the metaphor is a train. Some people call it a conveyor belt or even an airport walkway with a moving walkway with sides so high that you cannot get off until the end. But all of these different images are designed to convey this notion that once you get started on a clinical pathway,
the momentum is such that it is very difficult to stop the organizational scheme that by default produces action and procedures. And so we've worked in this space for a little while, and we developed a model, a conceptual model that we call clinical momentum that draws from work by Professor Kaufman, as well as other medical social scientists and those in psychology and decision science and behavioral economics. And so to illustrate what clinical momentum is and to perhaps sort of grow from this quote from Professor Kaufman in our space, I think it's useful to consider a common patient exemplar from the ICU. So a patient with respiratory failure, like one we've already been talking about, that's intubated, mechanically ventilated. And as ICU day 14 approaches, there becomes a recognition that there's time to discuss tracheostomy. But I think we all know that this reductionist diagram really doesn't capture what we're doing every day on rounds for the patient and their clinical course. And instead, it looks much more like.
this, that after they get intubated, their blood pressure is low, they get started on vasopressors to fix that problem. They have anemia, so they need a blood transfusion and the family gives informed consent to proceed. Their urine, they stop making urine and be diagnosed with kidney failure and they need now hemodialysis. And sort of over time, this latent system level influence of the speeding train and the clinical momentum is building. And this influence is permissive of the use of interventions. And again, sort of describes a variety of analogies, the speeding train and the conveyor belt and so on.
And this clinical momentum, what we worry is that it influences patients, families, and clinicians to accept and tolerate additional ongoing interventions without pausing to deliberate about how the interventions and what it means to need a particular intervention. And it makes it much more difficult to consider alternate options. And we wonder if this helps explain why when we use our best communication skills, and even with in the moment, it feels like we're working upstream sort of against an influence of the momentum that's already taken hold. And so by the time we all sit down for the family meeting and the patient is dependent on mechanical ventilation and dialysis, the daughter has done what was needed in each circumstance for her mother. And perhaps it's difficult to see the difference between a tracheostomy, which may be needed to help prevent upper airway complications for this patient and all of the decisions that she made along the way.
So another way that this clinical momentum idea has been described or modeled is like a snowball sort of racing down. Perhaps they're trying to visualize the side of a mountain here. And as individual abnormalities and clinical problems come up, influences like cascade effects and the way we make decisions, which is often recognition-primed decision-making, which is sort of an automated way of introducing the appropriate action to a particular problem at hand, the momentum has grown over time and is quite influential. And I think it's really important, I've sort of said this once already, but I think it's really important to underscore how it is possible that this momentum is having divergent influences on ourselves as clinicians and the patient's families and sort of decision makers. So this is a model that was from the PDAC.
ICU literature, but I think probably represents really quite well what is happening for our patients and families in the adult world as well. So you can see how they have sort of, we start at two ends of the spectrum, where at the beginning of an ICU stay, which is on the left-hand side of the screen, here we are as the clinicians and the ICU team, and we're really focused on individual disarticulated problems that the patient needs to be intubated, they need a central line. And we need to monitor particular labs. And when the family comes in and when the patient comes in, they are worried about the loved one that they've known perhaps for many years who has an entire life sort of outside of this hospital context. But as time goes on in the ICU, as clinicians, because of our experience, we recognize that for each additional needed form of life support that this patient needs, we become increasingly worried that the patient may not recover. And so we have behind the scenes what this means to us and what we're sort of slowly coming to terms with is perhaps that the patient might be nearing the end of life.
But what we worry about is that the families may be having a different experience. They become acculturated to the idea of how we talk on rounds. They learn what it means to monitor the daily fluctuations of creatinine and PEEP and FiO2. And they go through sort of a very quick mini medical school and slowly learn that that's what we're focusing on. And they become focused on it over time in the ICU. And there's this important inflection point, which I think perhaps can give us an interesting way to think about conflict in the ICU, where...
where at some point we've changed and all of a sudden we're ready to really start focusing on the patient as a whole and what our range of expected outcomes might be. And the family instead has taken our place in the problem focused arena. And so to come back to this issue about need, I think this is one lens into this larger system level idea that even our language reflects the fact that our ICU system is specifically and expertly designed to deliver highly specialized and lifesaving technologies to patients with individual problems. And so now I want to pivot a bit and think about a different phrase that's common for us in the intensive care unit and examine the tension that's exposed by this phrase.
And so I suspect this isn't a point I need to belabor to this audience, as most of you here today either know or routinely use goals of care in your everyday professional work. But I think a quick Google search does help us illustrate just the notable reach of this phrase, from patient-facing logos asking patients if they've had a goals of care conversation with their health care team, to statewide organizations that are designed to promote right types of care for patients, to the branding of large health system initiatives, and even in this freely available or publicly available slide set from a particular legal and risk management clinical in-service on goals of care. And so all of this is just to say that this phrase is very common. We conducted a systematic review in 2019 looking at the specific phrase goals of care in the healthcare literature, and we found an exponential rise in the use of this particular term starting in the early 2000s. And one of our purposes was to actually to sort of discern whether there was a shared understanding of what goals of care means. And in fact, we found a really consistent definition of how scholars talk about goals of care.
So the overarching and how they define goals of care is this, the overarching aims of medical care for a patient that are informed by the patient's underlying values and priorities established within the existing and a specific clinical context, as in facing a particular illness or in addressing a problem as a hospitalized, as a patient in the hospital, and then ultimately used to guide decisions about the use of or limitation on specific medical interventions. And so we put together this visual diagram to illustrate how goals of care Care is described as sort of sitting at the center and being a really operative notion as a way to combine something underlying and that could be outside of a clinical context, namely a patient's values and their individual priorities. And when you put it into a particular specific clinical decision, you can formulate goals of care that ultimately are used to guide decisions and courses of action. But we wanted to understand a little bit more carefully how goals of care is used.
in the clinical context. And to do that, we conducted two different studies. One was looking at electronic health record notes from about 50 patients in ICU. So actually, we ended up looking through thousands of pages of notes, and we undertook a qualitative discourse analysis. So trying to understand how goals of care was being used in the natural environment in the health record. But recognizing that that's an incomplete view on the clinical context, we also conducted a variety of focus groups with 59 ICU clinicians who represented nine different professional roles in the ICU.
And what we found in both studies was actually fairly similar. And first, that clinicians use goals of care to signal end of life. And so this is just a representative quote from an ICU nurse that joined one of our focus groups. So for us, the phrase goals of care is synonymous with an end of life, end of the road discussion. We've tried everything. Now let's have a goals of care discussion with the family to explain where we're at.
And similarly from the electronic health record, a consultant wrote about an ICU patient. At this point, we've established that all hemostatic interventions have been futile, agree with the primary team's efforts to address realistic goals of care in the context of this end-stage multi-system organ failure scenario. And clinicians also tend to use goals of care to signal conflict. So an ICU physician again told us, I've seen goals of care used in situations where the care team feels that prolonged ICU care will be non-beneficial, but the family is not in agreement. Oh, this family needs goals of care. It's implying that the family is not on board what we think is right or appropriate.
And similarly from the health record, you can see this notion of conflict embedded in the way the language is used. We'll continue to press for resolution of goals of care. And here we're looking for a family meeting with ethics to try and work towards a more mutual goals of care decision. And we also found, and clinicians told us this in the focus groups, that they use goals of care as an implicit signal. And in fact, they generally felt like it was best to avoid this particular phrase with patients and families. So one ICU physician said it's a phrase that's often used amongst providers. We all sort of have this secret and implicit understanding within ourselves as to what we mean when we say that.
And so the tension I hope to be highlighting for you so far is that on one hand, we have a straightforward and hard to argue with scholarly view on goals of care. And perhaps sort of a more nuanced argument on that side is that promotion of goals of care through campaigns, for example, like this one that asks whether patients have had a goals of care conversation with their health care team is an avenue to achieve better individualized care for patients. And then on the other hand, we have a complex arena of the ICU where the same phrase has taken on. quite a different meaning, and the meaning in some cases is latent and even described as secret among clinicians.
And now I think one perhaps really tempting explanation for this tension or tempting approach at reconciling the tension would be to argue that we need sort of more education for our ICU clinicians. And perhaps ICU clinicians don't understand the meaning of goals of care if we're using it as a euphemism for end of life. But we did when we conducted the focus groups with ICU clinicians, what we heard from them consistently in all the focus groups is that, in fact, they knew this ideal scholarly view of goals of care. And yet they still tended to use goals of care in the clinical context in this other way. So it didn't seem to us and it didn't seem to the participants in our focus groups as if this was a problem of misunderstanding. And in fact, it's perhaps a problem of how the ICU system and the embedded processes of care tend to evolve. And then our language ends up reflecting these defaults. And so I want to take this moment again to step back and again, reconsider sort of how this tension might both reflect and
show us an example of how we might consider differently how to approach the underlying ICU processes of care. Now, process maps, most of you probably have seen a form of a process flowchart, or sometimes this is called a process map, in a variety of different contexts. In healthcare, they're sometimes used in quality improvement projects and in process-level workflows and helping people move through different processes of care. This particular tool originally came from a field known as systems engineering. And I'm showing you an image here from NASA. So actually, NASA and the space industry was one of the spaces in which systems engineering was first took hold. And so NASA was a real early adopter of the use of process engineering and operations engineering tools in order to guide their complex space missions.
And so process maps are a visual diagram of a system. They demonstrate the temporal link between steps and they show how individual steps unfold in sequence. So this is perhaps sometimes contrasted to a checklist or a series of steps because these maps and these models show relationships between steps. So at its core, a process model is used to show how any individual system produces outcomes. It's also a very powerful tool that in many industries is used to capture complexity in systems and the variety of different layers of control and process failures and process fail safes that are used to ensure consistent outcomes. So you may be looking at this and say, like, sure, NASA has quite a few steps here. But what I'd like to point out is that this is actually just a small upper corner of the process map from NASA. And this is the whole thing. So this is sort of how they've layered in the complexity of how they design the process flow for their missions, their space missions.
And so in systems engineering, one primary purpose of a process map is to identify dysfunctional steps and to find points at which a current process fails. So these are also described as system level vulnerabilities that are targets for improving the system to produce better outcomes. And so why am I telling you all this about NASA and space travel? So we decided to use this tool to try to understand the processes of care in the ICU. And I think it is.
Not something novel to say that, you know, it's been recognized now for decades that failures at the health system level are more likely to be caused by system level, sorry, failures and outcomes or undesired outcomes in health are more likely to be caused by system level failures than by individual level human error. And yet our tools to do research on the system and to really identify ways in which we can apply system level change have really lagged behind that recognition about. the importance of health system design. And so we decided to sort of take this modeling approach from engineering to reconsider whether we could have a different view of our health system.
So we sought to generate a process map of the embedded care process for patients with acute respiratory failure who are admitted to a medical ICU. And so we conducted this study at two different academic medical centers, and we included members of the interprofessional ICU team, the palliative care team, as well as surrogates of patients who died and patient survivors. We derived an original map, a preliminary map from our analysis of the health record and direct ICU observations by health systems engineer. And then we revised and refined and improved and validated and tested that map through. multiple series of focus groups and semi-structured interviews. In total, we had 70 participants who were involved in producing and validating the map, and they represented 17 distinct stakeholder roles within the process of care delivery in the ICU.
And this is a summary map of what we found of the ICU care delivery process. And I call this a summary map because we have very detailed process maps that represent each of these three different phases of ICU care. But here today, I'm just going to focus based on our conversation today on sort of this overall process flow. And so there's a variety of features of the map, but I'm going to focus your attention on just a couple of them today. So I think what we heard from the people in our study is that the power of this map is that it really represents care for any patient who's admitted to the ICU with respiratory failure. So you could take any patient from step one, which is the yellow oval, so someone presenting with severe acute respiratory failure, and all the way to their ICU endpoint, and the map will show you how they got there, whether the endpoint is end-of-life care, recovery and extubation, or the receipt of prolonged mechanical ventilation.
And as I alluded to, the process that we generated in the map that we used to summarize it has three phases. So initiating clinical care, the early moments where you're developing an initial plan of care, and then phase three is really a cyclic process that recurs daily in the ICU as the patient's physiology evolves. And one of the features of the map that I would like to highlight for you today are these four distinct deliberation periods. And so using standard and traditional process mapping icons, the diamonds typically are used to represent a decision point. So a place in the process in which the flow of a particular process could go one way or the other. And what we learned from...
the patient examples we observed and all of the stakeholders that we talked to is that these are naturally occurring moments that deliberation occurs for patients in ICU and deliberation about whether or not to initiate, continue, or add life-sustaining therapies for an individual patient. The reason why they are represented here with dashed lines is because what we also heard is that even though these are naturally occurring and perhaps well-positioned moments for deliberation, they're highly inconsistent at the individual patient level. In other words, they're often bypassed these moments for deliberation. And that is what is represented on this map as the blue main line coming through. So if deliberation is bypassed, we initiate mechanical ventilation, we continue mechanical ventilation, and then we are in this cyclical cycle.
here, where depending on the patient's physiology, we reevaluate and continue or add and escalate life-sustaining treatments until one of these deliberation periods is triggered or recognized. And in the qualitative analysis of the data that we collected, we learned a lot about why these deliberation periods were so inconsistent. And so I'm just going to share with you several of the main themes that we learned from our research. First, from a surrogate decision maker who at one of these moments in the early deliberation period really felt like continuing mechanical ventilation was not okay and was not consistent with what her cousin would have said. What she told us is, I didn't want to sound like I was giving up on him or that I didn't trust their ability to cure him. Me bringing it up was not the right thing. So at that moment, the deliberation was bypassed and his care continued until, in fact, several weeks later when a different deliberation period was brought up by the ICU team.
An ICU nurse told us that if the attending that's on or the fellow that's on, they don't really do a good job connecting or empathizing with patients and their families, I might not address it with them. So this nurse was describing scenarios in which they felt that the deliberation period was appropriate and perhaps they had evidence to suggest it should take place, but they made a decision based on their perceptions of clinician level variability not to pursue deliberation in that moment. And I think what resonates with me quite strongly as an ICU clinician, and I recognize this in my own practice, is that there is a huge interruption at the moment of staffing rotation at the physician level. And this was an example from a nurse. We'll drop the ball in a way and say, well, I'm only on for the weekend, so I don't want to get involved. I'm only here for one shift.
And so what happens when these deliberation periods are bypassed is that the strong blue through line becomes our default embedded process of care. And what happens once we're in this cycle of continuing to add or escalate treatment because of worsening physiology, there ultimately becomes a point, a hard stop in the process when the physiology progresses to a point in which the deliberation period really cannot be ignored or bypassed any longer. And that's the moment where we saw a lot of these conversations about. addressing realistic goals of care. And here you can see how that's described. And even the rationale here, it's emphasized that this is an end-stage multi-system organ failure scenario. And so to summarize so far what we've been talking about today and what I've been suggesting to you is that our language, and as one example, the language of need in the ICU creates this...
inextricable link between identifying physiologic problems and our increasingly impressive ability to use technology and action to address those problems, to solve those problems. And when we put a variety of problems together for our patients in the ICU who are with us over the course of days and weeks, that these individually linked problems, the need to be intubated and the need for hemodialysis produces because of... the way we make decisions and our experiences and impressions of ICU care over time produces a clinical momentum. And that momentum certainly can be disrupted and we do it all the time, but perhaps this is a moment and a reason to think, is there a different way or is there a different perhaps approach where it doesn't feel as if when we use our best and highly skilled communication practices that we're trying to undo an understanding that has already taken hold.
And second, that the phrase goals of care, while it is uncomplicated and easy to agree with in the healthcare literature as a way to make sure that our care aligns with what's most important to the patients in any given clinical scenario, what we find in the ICU because of how our systems and our staffing rotation and our complex teams and the unfamiliar environment and the problems with agency, these moments for deliberation where perhaps we could have thought about goals of care earlier on are bypassed. At the moment of physiologic failure is often when we invoke this notion of talking about goals of care, and perhaps there's a moment to consider whether that could have been done earlier or even whether goals of care is the right phrase to use in that context.
And so I want to sort of leave you and I'll just have a few more slides here and hopefully we have some time for discussion with another potential solution and one that for me, at least over the course of considering this language and this work that we've been thinking about about the ICU system is perhaps ideally suited to be a new approach to care and a new way to consider how do we address these system level embedded processes. And so what I've been describing to you so far is a variety of implicit latent processes that are really common in our intensive care units and where these. Moments for deliberation are characterized by dashed lines and moments that are really easy to bypass because of the place and the context in which we work. And we've been really interested in the concept now of time-limited trials as an opportunity to formulate a more explicit plan where instead of being vulnerable to the bypass of deliberation moments, trying to bake in and to establish key moments for deliberation that don't get skipped. And so we undertook...
With ATS, we put together a workshop committee and over the last couple of years, this hasn't been published yet, but we've been developing a consensus definition for what we mean by a time-limited trial for patients with critical illness. And this should, hopefully our document will be published for all of you to read sometime next month. But this is what we've decided on our consensus definition of time-limited trials. So this approach is a collaborative plan among clinicians and a patient or their surrogate decision makers to use life-sustaining therapy for a defined duration. So this idea that we make a plan together with our patients, that it's explicit, and we define the duration of how long we're going to be using life-sustaining therapy. And after this plan, the patient's response to therapy informs the decision to either continue care directed towards recovery, transition to care focused exclusively on comfort, or perhaps in some cases, extend the trial's duration.
But one of the issues that I hope all of you would recognize, despite the fact that we have a lot of expert opinion on time-limited trials and there's growing enthusiasm for this approach in our literature in critical care, that there's very little evidence about how to use time-limited trials and in whom they should be offered. And really sort of fundamentally to ask the question and push back a little bit on whether they are different than how we're currently providing care. And if they are different, do they have a positive impact or are there any unintended harms or consequences from using poorly conducted time-limited trials? And so this is just an example about why we need more data and more research about time limit trials before we sort of unleash this strategy and recommend it broadly. So I was part of a research team led by Liz Viglianti, who surveyed a bunch of ICU clinicians.
and provided different hypothetical vignettes about cases. And then she asked the critical care clinicians, would you offer a time-limited trial in this particular patient's case? And if so, how long would it last? And so these are just two of the vignettes, two examples from the research study that she conducted. But one was a 79-year-old with IPF who had hypoxemic respiratory failure, and they were considering invasive mechanical ventilation. So almost 75% of the respondents felt that it was appropriate. But I think it's important here to recognize that a quarter of the respondents did not. And among those who felt like a time-limited trial was appropriate.
there was a really wide variation in how long that hypothetical trial might last. Someone said three days. Some others said seven days with an average around five. And I think it's a really key question that if we don't know how long the time of the trial ought to last, I think you do bring in the question of whether you could be causing some unintended harm by either trials that are too short to understand whether the patients had meaningful response to our therapy or too long and perhaps would be ineffective in addressing some of the issues we've been talking about today. And so I'm just going to leave you with my next steps, which is really to start to use some of these tools, both from systems engineering and process mapping and our experience conducting qualitative research, to ask the question about what is the best way and the optimal way to conduct a time-limited trial. And we're really excited to be getting started on a five-year project supported by the NIH this year that will be at five different ICUs across the country examining the use of time-limited trials and their impact on patients and on families and on outcomes and moral distress in our ICU team members.
And so I just want to acknowledge Justin Clapp and Bob Arnold and Margaret Shorzy, who have really been key contributors to the intellectual framework that I've been discussing today and a large number of collaborators that have contributed to the papers that I've referenced. And I want to sort of call out in particular two of my mentees, Michaela Reif, who is currently a pulmonary critical care fellow who did medical school at the University of Maryland. And she's really an amazing research, burgeoning researcher. And so maybe you'll see her name again more someday. And then Dr. Secunda, who is a professor, assistant professor at the University of Pennsylvania, who also helped lead some of the work.
while we were both together at Northwestern. So I am going to leave you again with a really nice picture of our lakes here in Madison and look forward to some questions.