No Easy Answers in Bioethics Podcast

The Patient Preference Predictor: Tomlinson and Stahl - Episode 1

September 7, 2017 Devan Stahl photoTom Tomlinson

Center Director Dr. Tom Tomlinson and Assistant Professor Dr. Devan Stahl discuss the Patient Preference Predictor (PPP), a tool they are developing that could potentially supplement advance directives or surrogate decision makers for patients, or that could aid decision making for the "unbefriended" patient who has no family or friend to make their wishes known. Could the PPP be even better than a patient's family member at predicting what a patient may want in a given scenario?

This episode was produced and edited by Liz McDaniel in the Center for Ethics. Music: "While We Walk (2004)" by Antony Raijekov via Free Music Archive, licensed under a Attribution-NonCommercial-ShareAlike License.

Episode Transcript

Liz McDaniel: Hello and welcome to No Easy Answers in Bioethics, the podcast from the Center for Ethics and Humanities in the Life Sciences at the Michigan State University College of Human Medicine. Today we're joined by Center Director Dr. Tom Tomlinson and Center Assistant Professor Dr. Devan Stahl. Dr. Tomlinson and Dr. Stahl discuss the patient preference predictor, their work in progress that relates to making decisions regarding patient care and that has potential to be a tool that could supplement advanced directives and surrogate decision makers. Let's have a listen to learn more.

Tom Tomlinson: Well hello my name's Tom Tomlinson, I'm the director of the Center for Ethics and Humanities in the Life Sciences at Michigan State University. I have been working in bioethics for thirty- five years, it seems longer sometimes.

Devan Stahl: And my name is Devan Stahl. I'm an assistant professor of Clinical Ethics here in the Center for Ethics and Humanities in the Life Sciences and I have been here for two years, so not quite as long as Tom.

TT: [laughs] I hope you will be here as long as I am.

DS: Yes, I hope so too. [laughs]

TT: So I'm going to start us off. We want to talk today about patient preference predictors or P.P.P. No snickering please. So what our patient preference predictor is, so what’s the idea behind them. So the idea is that, is to wonder whether or not it's possible to use a nationally representative sample, probably a very large sample of people, to see if there are associations that you can find between their preferences regarding a variety of life sustaining treatments at the end of life, or if they are demented, say, and various demographic and other characteristics that they might have. So that, um, if such if a patient were to be in a hospital, say, and you didn't know anything about them, you'd be able to use the information you did have about them—their age, their background, their race, their educational level, their income, you know a whole variety of factors that might be relevant from the statistical point of view, so that you could predict, with some degree of confidence, what they would want with regard to being put on a ventilator, say, or with regard to what they're going to be resuscitated if their heart stops. So that's the basic idea behind a patient preference predictor, is the potential usefulness of such a tool, you know, in the hospital or other health care settings.

DS: I'll just add in there that, and I think we’ll come to this when we talk about some of the debates about the P.P.P., but, um, so the sociodemographic factors that people would typically think of in a survey will be applicable. We might also ask about their current attitudes and values. So, you know, what kinds of things do they value and how can we sort of quantify that, or qualify that, for future use of the P.P.P. We might also even ask about their relative personal experience with medicine. So if a person has ever been medically hospitalized or if they have experiences caring for somebody who's been incapacitated by their illness that might affect how they, sort of, imagine their future being incapacitated and needing medical care.

TT: Yeah, or if they have a chronic illness, or something like that, which might intuitively at least might be associated or have some influence over their attitudes toward various life sustaining treatments.

DS: Right, and then we'd also ask things, you know, like what do you value most and we'd have to, we’re figuring out ways to ask that. But if someone is particularly religious and that might influence how they respond to these or, you know, that family’s the most important thing for them, or maybe it's not …so those might be also things that we can ask about to help get more refined about the kind of person we're dealing with and how we might generalize information about particular groups of folks.

TT: So, um, you know just from a methodological point of view, there are huge complexities and difficulties in the way of actually accomplishing this. But I think for today we're going to, sort of, set those ones aside and assume that it is possible to construct something like a patient preference predictor with some degree or another of reliability.

DS: Alright, so Tom, I bet people are thinking, well why do we need something like this because we already have these things called advanced directives. We have living wills in which you are able to say exactly what kinds of treatments you might want in the future if you were to become incapacitated and those are documents that our doctor should be following, and we can even assign certain people, so surrogate decision makers, to be our proxies during that time so people that are close to us that we think could make good decisions for us, you know, that's a sort of really easy way to get across what we might want in the future so why would we need something like a patient preference predictor, if we have these really solid documents saying exactly what an individual would want.

TT: Are you asking me or you going to tell me? [laughs]

DS: [laughs] Well I might tell you, too, I just wanted to phrase it as a hypothetical there, but I think we can think of a lot of good reasons why this might be helpful beyond advance directives. So first to just say, most people don't have advance directives. Only twenty to thirty percent of the population has even filled out a living will and that's after decades of really promoting the living will.

TT: So the federal Patient Self-Determination Act was passed something around early 1990s, maybe even 1990, which requires hospitals and health care institutions to ask about peoples advanced directives. So there's been an effort ongoing to encourage the creation of advanced directives for at least that long.

DS: And, you know, numbers have gone up with, when there's been a big push by certain groups to kind of try to get more people to fill out these documents and sometimes it works, but in general it just seems like most people don't have physical documents stating what it is that they would want in the future, so this is going to work for, you know, most people because most people don't have advance directives so we need some other kinds of evidence about what they might want. So we also know that even when patients do have advance directives, sometimes they're vague, so it's hard to predict...

TT: Sometimes?

DS: Well, lots of times, okay. [laughs] Lots of times they're vague and Tom and I know this from working as clinical ethics consultants in our hospitals, that it's very hard to predict exactly when you're going to need these advanced directives. So what are the conditions you're going to be in when you lose capacity? That's hard to predict and it's also hard to really wrap your mind around the kinds of life sustaining treatments we're talking about. So most people still don't really understand what it means to be intubated, what it means to be on a ventilator, what a feeding tube does. These things are, they're complex and they're hard to understand. So when we get these documents they don't necessarily apply to the specific situation in which the patient’s in, or it might be unclear exactly what they meant in those directives, so they're vague.

TT: Like so for example, in a POLST form—a physician order for life sustaining treatment—one of the options people are given is to say whether or not they want to be resuscitated, if they have a cardiac arrest. And so someone might check “no I don't,” but that form doesn’t tell you sort of what scenario that they were playing in their head when they made that decision, and so you don't know whether or not what they intended to communicate when they checked that box in fact applies to the situation they're in now. Maybe they were imagining that they were, you know, at the end of life with some terminal illness and that's when they didn't want to have resuscitation, but that means this is in fact not the situation they're in currently, but we have no way of knowing what they meant and the advance directives can't tell us.

DS: Right advance directives typically don't give rationales. Right? You don't—that would be, you know, sort of a long form. Usually it's just a check box and so we have no idea why they checked that particular box and what circumstance they were imagining. Most people imagine the end of life, but in most states advance directives apply whenever the patient loses capacity, not just at the end of life, making it a little bit more complicated when they apply and when they don't. And so we also know that in reality physicians don't pay a lot of attention to advance directives. Even though we push them and we say that they're so great, physicians tell us that they don't seem to influence their recommendations very often maybe because they're vague...

TT: You know, but also because I think, you know, physicians’ tendency is to is to continue doing whatever they think might be useful and when they get to the situation where continued treatment is not going to be useful or the next line of treatments are not going to be useful, that's when they begin asking about the advance directives, because they are hoping it is going to support a decision to stop.

DS: Right, so we also know that most families don't talk about the kinds of decisions that are really relevant to making these end of life decisions or decisions about life sustaining treatment if the patient's incapacitated. It’s just kind of a cultural thing—people don't like talking about these things and even though we push for families to have these conversations, maybe over Thanksgiving, it's just not a very pleasant conversation and so a lot of families just really have no idea what their loved one would want.

TT: And just as an addendum to that, very few people talk with their physician about their advance directive, either.

DS: Right. So we know that—we also know that families are not great predictors of what patients say they would want. So in these studies where, you know, we ask a particular person what they think they'd want and then we ask the family what they think they would, that patient would say, families aren't great at predicting that. They're maybe only slightly better than chance.

TT: Yeah, but better than the physicians.

DS: Better than physicians, but not, you know, nowhere near one hundred percent. Maybe more, just a little over fifty percent, and that's even when they have had a conversation about what the patient might want at the end of life. So families aren’t really great at predicting the sort of exact preferences of that patient for particular treatments.

TT: But then on the other hand when you asked patients why they want families to make decisions for them one of the reasons they give is because they believe their family knows what they would want.

DS: Right, so they don't talk about it and families aren’t great, but we still have this presumption that our family members would know us best, would know how we would choose, and that might be a faulty assumption. We also know that it, um, making these kinds of decisions on behalf of somebody else really stresses people out. So when it's up to you, and often physicians put surrogates in this decision, this kind of decision-making. You know, what do you think that your father would want in the situation and if you don't know, or even if you do know, it can be a very stressful situation--it feels like you're making life and death choices for somebody you really love and that's hard on people.

TT: And of course this is also one of the rationales given for people completing advance directives, is that it lifts a burden off family members who have to make these really hard choices and if they somehow know what mom would want, it would be much easier on them and so mom is motivated then for her family’s sake, for completing the advance directive, hoping that will somehow make it easier.

DS: And surely for some families, that does, but that's just not universally true. You know, even if you think you know what Mom would have wanted, to make a decision that leads to the end of her life is still a stressful hard decision to make.

TT: Feels like you're killing Mom...

DS: It can feel like that, yeah. So there is some evidence—so the P.P.P. is in its infant position right now. We haven't done a lot of, there hasn’t been a lot of survey data on it, but there is some evidence to suggest that it actually might be better at predicting what a patient would actually want than family members themselves. So we have kind of this nascent data showing us that it might be helpful in situations, which it's not super clear to families or surrogates what the patient would want.

TT: So let's imagine that's true. Let’s imagine that we've got a P.P.P. instrument that at least across some range of interventions and clinical circumstances can predict what people say would have said they would or would have wanted. How could, how can we use that? I mean, should we use it? I mean how should we use it? So, and just to start off, one thing, one situation that occurs to me that you and I have seen quite a bit is that, you know, an elderly patient in the nursing home probably pretty severely demented who has no family, no friends, and family have long since died or drifted away, so there's nobody who knows her. She doesn't have a living will or durable power of attorney. She'll probably end up with a court appointed guardian who knows nothing about her. And so, if you had enough information about her, from her medical record and other things like that, with a P.P.P. you might be able to say, with some certainty, whether or not she'd want to be put on a ventilator. And it could be very useful information in some of the situations we've been involved in.

DS: That's right. We call these patients “unbefriended patients,” which is actually a terrible term—we need to come up with a new one. But they're “unbefriended” meaning they just don't have anyone to speak for what kind of person they were and what they think they would want. So in those scenarios it's very difficult often to get any physicians to agree to stop life sustaining treatments because they're afraid that they just don't know. There's really no evidence of what they would want and those are sometimes the toughest cases because, like you said, the court appointed guardian will have never met this patient so they don't know what they would want either. And then we have a—can often get involved in the courts and it can be complicated.

TT: Right, so I mean, the court appointed guardian may be pretty risk averse about making these life-or-death decisions and so the default option is just full speed ahead. Out of P.P.P. the evidence that a P.P.P. could provide, might actually tilt the balance back in a direction that’s more likely to be in accordance with what that person would want.

DS: So that might be the sort of most obvious use of a P.P.P. We also certainly have experienced families who are asked to make these decisions for a patient who doesn't have an advance directive who really, like we said before, have never had this kind of conversation and really just say “I have no idea what my family member would want,” and so this could provide them with some extra kind of data, some empirical evidence showing what the patient might have wanted and it can give them some more assurance, if they have no idea.

TT: And you can also, I think, that information might also serve as a way to draw them draw their attention out of their grief, which may in fact incline them to continue to do everything. You give them some other focus for their decision making, which might help them make a decision that's probably going to be more in line with what the patient would want.

DS: Right, these are stressful times on families and often times being asked to make a really important decision is difficult in the midst of that grief. We also know that lots of families are very conflicted, you know, maybe they've had the conversation, maybe they haven't, but when it really comes down to it, what that person would want, and the particular circumstance is difficult, and so you might sort of waiver back and forth—“well, you know, my loved one said this but they also said this and at the end of the day, I don't know what that means about the specific treatment that we're talking about, whether they want it or not,” and so the P.P.P. could help just to give some more evidence to that family as well. So there also might be families though that do think they know what the patient would want and yet they're, they remain conflicted only because it's a tough decision. And so having that kind of extra evidence would help them, as well, even if they aren't as conflicted as the previous family—one more piece of evidence to say “yes I'm doing the right thing,” and that can be a potential comfort to families.

TT: So how about another situation. I'm imagining a situation, which we also sometimes encounter, of a patient who has an advance directive that is pretty specific about not wanting something, whether it might be a ventilator or whatever, and they have a durable power of attorney and the durable power of attorney, who may be a family member, often is, says “oh don't pay attention to that.” You know, “she didn't really mean that” or “you know, we've since had some other conversations and so I'm pretty sure that’s not what she would want, if we could ask her.” Can I use the P.P.P. there to determine what the likelihood is that the durable power over attorney is really representing what the patient would want? If I can produce evidence that confirms that the patient would want exactly what she said she would want.

DS: This is where I think it gets a lot more controversial. So initially, I think the P.P.P. was conceived as a supplement to shared decision-making between the surrogate and the clinicians, which is ideally what we want—is everybody involved helping to make the right decision for the patient. To override what a surrogate, or even a DPOA would say, gets you into much more treacherous territory, there. Because, you know, sometimes it is the case that, as we said, advance directives seem to say something, or we might interpret they're saying something, we might interpret them to be saying something, but then a family member or that durable power of attorney can say “wait a minute I was there when she filled out this advanced directive, we had a conversation about it, and actually how you’re interpreting it is wrong.”

TT: That’d be the exception rather than the rule though, right?

DS: [laughs] It would be. Yeah, sometimes people who are assigned as a surrogate or a DPOA flagrantly disregard them, in favor of their own preference, which is not what they're supposed to do. Really, the job of the surrogate decision maker is to make decisions that the patient, him or herself, would have made—not to make decisions for them, but to make decisions they would have made. And so when that's their obligation and they're making decisions the patient wouldn't have made, or we have evidence to think that they're making decisions contrary to what the patient would have wanted, that's when it's, you know, you really have to get with that surrogate and remind them of their responsibility and potentially even remove them as a surrogate, if they're not making good decisions for the patient that the patient would have made.

TT: When I said I mean, I want to back up a little, because I want to make a distinction between using the P.P.P. to override the surrogate’s decision, or using the P.P.P. to advise the surrogate’s decision or to maybe even sort of confront. This may be too strong of a word maybe, but so not override because to P.P.P. is not going to have any legal authority, certainly. Advanced directives do have at least some legal authority, so we're not going to do that, but we might be able to say “whoa wait a minute, you know, we have reasons to think that given the kind of person your mother is, what we know of her, at least, she would not want what you're asking us to do, so let's talk a little bit more about what your reasons are for thinking she would want this.” So there would be a way of sort of confronting the surrogate with evidence, some evidence, that indicates that they might not be quite right about what their family member would want. Is that, would that be okay?

DS: Yeah, I mean I think then you get into the logistics of who is going to bring that up, and how are they going to bring it up, but you know it might be that the physician or even a clinical ethicist would meet and this would be a segue into that conversation. Because those conversations are hard to start without being confrontational. Right so, you're sitting down, you're thinking in the back your head we just don't know that you're making the decision that this patient would have made. How do you address that? And maybe one way you can address that is to say “you know we have this other kind of evidence, what do you think about that, and is your mother like this group of people?”

TT: I mean so “the evidence we have suggests that people somewhat like your mother would want this. Rather that, so let's talk about that, would do you think about that?”

DS: Yeah, it would just have to be done very carefully. But it could potentially be something to be brought up in those kinds of conversations, which are OK. Well then let's segue from there into what are the debates about P.P.P. You know, what might people say against it, why might people not want to use it, sort of maybe on theoretical grounds, but also really practical grounds. So what do we know about the detractors from this.

TT: Well also, so one of the, one of the points that people make is that the patient preference prediction is not—the process is not advanced, it's not the creation of an advance directive. Now when someone creates an advance directive, they go through an entirely different process than we use when we’re applying or developing the P.P.P. So in the P.P.P. we might discover that African-Americans are more likely to want aggressive care in the end of life, say, but when an African-American completes their advance directive they don't choose certain options because they're African-American. They don't say “oh, well because I'm black I think I'll do this.” They choose it for much more personal reasons, the same kind of reasons, largely, that other people will have. And so you can't think of the P.P.P. as representing or somehow replicating the process of developing an advance directive. I mean the other disanalogy is that certainly when we're appointing a durable power of attorney, selecting a family member to represent us, we are not just, we are acting as an act of trust because they’re our family, because they love us. We trust them and we also care about them, you know, we want them to have some role in making decisions about us because we're concerned about how difficult it may be for them and recognize perhaps that that difficulty may justifiably lead them to make decisions that wouldn’t be exactly what we would want, but we also care about them so when we create the advance directive we're kind of doing two things: we're entrusting our family to make decisions for us because we love and care for them, we're also trying to hoping that they will also take into consideration what we ourselves may have said we wanted, but we recognize that there may be a tension between those two things, but we're going to trust them to figure out how to resolve that tension. So the P.P.P. doesn't do any of that. So how can you replace an advance directive?

DS: And maybe it shouldn't. So I think we said in the beginning, though, that you know this is actually most helpful as an aid in the decision making process and not as a replacement for an advance directive, or as any way potentially overriding advance directives. But you're right that this, you know, the granularity or the particularity of an advance directive is probably going to be more helpful than sort of generalizations about sociodemographic groups. So in that sense, it's, I don't think it's meant to replace the advanced directive. And the logic of it is certainly different, so that's something we'll have to think about and again there's all these methodological problems that you mentioned at the beginning of how exactly we get sort of representation and what that will look like. So it's something to think about as the P.P.P. kind of thinks about what it's going to ask and how it's going to ask it.

TT: I mean the other, I'm not sure this is a philosophical problem, but it is an ethical and practical problem, is that when you have family at the scene—I'm not talking about the unbefriended patient who has nobody—but someone who has family, whether or not they have a group, or proper attorney, but you have them at the scene. You're making a decision in real time. The P.P.P. is going to be created by posing hypothetical scenarios to people. Those scenarios are going to necessarily be somewhat simplified and stripped. There's no way they're going to correspond exactly to whatever the medical circumstances might be of a particular person. So you lose that contextual relevance when you're trying to apply the P.P.P. You don't really know what that person would say about this much more complicated situation. You might know what they would say about that hypothetical scenario that you used when you created the P.P.P., but they might make a very different decision given the complexity of the circumstances that they're in now. We have no way of knowing that. So there's a danger, then, that we apply the sort of simple-minded algorithm of the P.P.P. the circumstances are much too complicated for it to really handle and so we're thrown back, then, on relying on the family. So if we're going to do that, anyway, why use the P.P.P.? [laughs]

DS: Right well that's a good question. You haven't answered your own question? [laughs]

TT: [laughs] No.

DS: You know, I don't think it sort of foundationally under my P.P.P. in such a way that then it's useless, but that these are considerations for you know how it gets applied and how we think through even how to construct to the kinds of surveys that would give us the data that we think is most important in these scenarios. We also think that you know we're generally talking potentially to healthy people who can't imagine what it's like to be quite that ill. It's very difficult for any of us to imagine what it would be like to be so incapacitated we couldn't make our own decisions. So we're asking healthy people who have their full capacity what kinds of decisions they would make in a circumstance they've never experienced or probably have never experienced. And we know from a lot of data that people just aren't great at predicting how bad or how good something might be so we tend to over predict how bad a circumstance will be, when in reality we might be able to cope with it. So we just know that people aren’t great at predicting their future states about health and what it will mean to them. And, you know, that again I don't think undermines everything about P.P.P., but we just have to keep in mind that in reality if people could speak for themselves in these incapacitated states they might say—they might change their minds about what they would want there. It’s just hard to know.

TT: Yeah well, to begin with they wouldn’t be incapacitated.

DS: Well that's right. So it's kind of impossible to know.

TT: And these difficulties apply to the use of advance directives, too. I mean we can’t, none of us can foresee the future with complete clarity or know exactly what we in fact would want at some later time because our preferences and attitudes and all kinds of other things change in between. So I don't, you know, this is a problem for the P.P.P., but it seems to be equally a problem for advance directives, so it’s just one of the inherent limitations that we have to work under.

DS: That's right and it's been debated amongst bioethicists and philosophers for decades. Something called precedent autonomy. How could you possibly predict what you would want, what your preferences would be, what your values would be, if your circumstances totally changed? Might you not change your mind? And so this has been-

TT: We don't have a way of testing the hypothesis in the case of incapacitated people.

DS: We do not. All we have is evidence of how people do change their minds, when they do still have capacity, but they become severely disabled. And we know that people are much better, they say their quality of life is better, than they would have predicted. So we have that kind of evidence but we can't necessarily apply that directly to the incapacitated patient because by definition they can't tell us what they want.

TT: So I think, you know, some of this is kind of cautionary—that we have to be really cautious about using the P.P.P. or giving the P.P.P. an authority that reaches well beyond what its capacities are. You know, so, thinking of P.P.P.s as advisory, perhaps, rather than decisional or dispositive that they will somehow settle disputes or override family or other surrogate decision makers probably is reaching too far.

DS: Probably, and I think we also need to think about, you know, is this just information, is this somehow a default so it—could it be used as a weak default. You know, we have this information about someone like your loved one and here it is and we think that that's probably right, but if you oppose that surrogate or family member, okay, that's fine. Or might it be a stronger default? This is what we're going to do unless you give us a really compelling reason or a narrative about your loved one that would override what the P.P.P. seems to suggest. And so a weak default versus a strong default can be another way we sort of try to parse how the P.P.P. would be used in a particular situation. And of course that weak default is just going to be a little bit easier to justify than a strong default or we demand compelling reasons not to use the P.P.P. as the decision maker unless the family can give us a good reason.

TT: And of course a stronger default is likely to escalate what may already be a adversarial situation in the intensive care unit.

DS: Well absolutely, I mean we've experienced it can get very difficult, hostile even sometimes, when the clinicians are recommending one thing, the family wants another thing, and they hit an impasse and it gets—it can get heated. So, you know, using other kinds of evidence about your loved one against what you say you know about them might not help alleviate those complicated situations between families and clinicians.

TT: I mean it just occurs to me that one of the things that supports a show, and I think that you mentioned this earlier on, is that the physicians don’t really look to advance directives. So, that you can have an advance directive on the on the chart, or in the record, but it doesn’t really shape the plan of care until it suits the physician. It's, you know, views about what's possible. But we could be more prescriptive about this and expect—we should be expecting, actually—that when there's advance directive there should be something in the chart where the physician indicates whether and how that advance directive is going to be used in the plan of care. And maybe you could do something like that with a P.P.P. so that when you've got evidence via the P.P.P. about what someone's preferences might be, the expectation would be that the physician would have to say something about how that information was going to be used in developing a plan of care.

DS: That would be a really positive outcome of the P.P.P. because, as you and I both know, having these goals of care conversations with families is not only difficult, it very rarely happens. So, you know, the plan of care just goes forward. It doesn't happen until there's really nothing else to do and then the physician wants to have a conversation about, you know, what treatments and what's working and not working and what the goals of those treatments are, but really they should be having them much sooner than that. So that, you know, this other kind of evidence can be brought to bear on the plan going forward rather than just at the impasse.

TT: Right exactly.

DS: So some people might also worry that in the implementation of P.P.P. we're falling into more stereotypes. Right so, you know, might we not be sort of reinforcing potentially even oppressions generated by a society that that make it so that some people get more care, some people get less care, based on, you know, their position in society. So do we worry about now reproducing injustices in our society by stereotyping certain groups, saying “no African-Americans would want this” or “women would of a certain age would want this.” Might that worry us, and I think it could worry us.

TT: Well I think, so I think part of part of the answer goes to the methodology. So we would want to have a survey tool that gathers not just little isolated pieces of information but lots of information so that you could see how having being low income, for example, the effect that that might have on your treatment choices is mitigated by something else about you so that the picture that gets painted from the P.P.P., it's not single minded. It's complex and rich. You know, I think the other response I want to make is that it's not—if in fact it turns out that part of the reason we're going we think that we should withdraw care from this, you know, poor African-American woman without a college education, is because of the data that P.P.P. provides, and that gives us confidence if that's what she would want—that's not oppression. That's respect.

DS: Right and it might even potentially undermine some of our stereotypes. What if through this information we find out that groups that we thought wanted certain things actually don't want those things, or do want things we didn't think they wanted. So it might help us together to sort of subvert some of those stereotypes by gathering lots of information. So at least a potential there, but I think it's something that needs to be continually thought of, like you said, as we as we think about methodology for the P.P.P. Well and you mentioned this Tom, but I think perhaps the strongest debate occurring amongst bioethicist is really about surrogates and about accuracy and autonomy and, you know, we like you said don't necessarily choose surrogates to make decisions for us simply because we think they'll be accurate about choosing what we would choose. We also choose them because, well, we love them and we make decisions often with them, and so our decisions affect each other. And so they’re more complicated reasons than simply, you know, they choose what I would choose than for the reason we choose surrogates. And potentially P.P.P. is giving us more accurate information and yet, you know, if you asked me “would you want your loved ones to be able to override your own decisions?” I might say yes in a lot of circumstances and the P.P.P. simply can't account for that.

TT: Right and, in fact, there is some evidence that when people are asked whether or not they will give their families leeway to make a decision, other than the one that they themselves wanted, typically a majority will say yes I do. I think that, you know, that leeway does—when I say I give someone leeway I don't mean that they don't have to pay any attention to what I said I wanted. You know, it's not that I think that when I say my family has leeway I'm not giving them permission or saying you should always override my decision, and pay no attention to it or pay no attention to what I would want. I recognize that that's only one of the things that they might need to or want to take into account and, I’m giving them permission to make that judgment, not permission to ignore me.

DS: Yeah, I think that's right. So, you know, I still want you to pay attention to the things I always said I wanted, but you know, most of us would say we recognize that in difficult situations we would want to give allowance to our loved ones to struggle, to maybe think through and take more time so not overriding everything we ever wanted, but certainly taking other considerations into mind as they do that. Which again is why the P.P.P. is probably most useful in aid to surrogate decision-making and not as the chief way we decide for patients. Well I think that's all I had to say about the P.P.P., for now.

TT: It has been fun!

DS: It has been fun! And I hope everyone keeps up on this. We are launching our own study, it's only in the beginning stages. So we will hopefully report back once we have some more information.

TT: Yeah, stay tuned.

Liz McDaniel: Thank you for joining us today. Please visit us online at bioethics.msu.edu, and follow us on Twitter at @MSUBioethics. This episode of No Easy Answers in Bioethics was produced and edited by Liz McDaniel.