Interview Judith E. Nelson, M.D.
A palliative care and critical care doctor, she is professor of medicine at the Mount Sinai School of Medicine and associate director of Mount Sinai Hospital's Intensive Care Unit. This is the edited transcript of an interview conducted on Nov. 21, 2009.
Has [working in critical care] turned out to be what you had imagined?
It's turned out to be every bit as compelling, but for many different reasons. I had no foresight about the complexity of the emotional issues and the weight of the decision making about the use of the resources. I had very little training to confront particularly the emotional issues and the family needs.
It may sound like a cliche, but I really saw the ICU as a place where people came to get their lives saved, and that I would work on doing that. I didn't expect to be able to do it all the time, but that was why I was there. And as time went by, it became clear to me, first unconsciously, that there were these people whose loved ones were very sick in the ICU, and the patients themselves, who came to this critical illness with their whole background of psychology and their family dynamics, and were then placed in this crisis situation and the most intense experience of their lives. And the most important decisions they had to make and that I had to make all had to be made on very short order. This is not something that I had any preparation for.
Talk about dying and death in America. What do you see?
When I grew up, nobody talked about death and dying. My father was a physician in a generation where patients were not told of their diagnosis, for the most part were not told that they were dying, and we didn't speak about dying or death in my home, and when I was in medical school, we really didn't speak about it in a direct way. We didn't speak about how we would handle it if we were faced with it. We didn't speak about how the patients and their families would approach it. We didn't speak about the impact on us when it was happening. That's why, when I came to realize it, it first came unconsciously. ...
Most of us say we want to die at home, yet most will die in a hospital, and many in an ICU. What drives that?
I described my own attraction to the field of intensive care, which had in significant part to do with the miraculousness of the technology and this knowledge base and skills base that would allow people who were hemorrhaging to death the ability to tide them through these most acute problems. I think the miracle of this is very attractive and seductive for everyone, and the reality of it is simply not evident outside the ICU itself for many, many people. The popular images of this on television are not the reality that I live on a day-to-day basis.
The burdens of intensive care can be very, very heavy, and the outcomes are often not good. So we have to face this extraordinarily difficult challenge of knowing when to use this miraculous technology and for how long and knowing when to try to preserve for people a peaceful and dignified process of dying. Walking that line is the very hardest part of my job, and constantly recalibrating myself from one side of it to the other.
How do you do that?
I think I start always from the premise that if I can save a life, return somebody to the condition they were in before the critical illness, that that is absolutely my top priority. So I always start from that default position.
What kinds of things can you do then to keep someone alive?
So we can make sure that people who cannot breathe on their own are supported by a respirator. We can support blood pressure when someone is in shock. We can resuscitate someone when they are hemorrhaging. We can often correct an abnormal cardiac rhythm. And, at least initially, we often have remarkable success -- certainly not always. But usually the initial therapy has some chance of being successful. So my default position, which I don't always default to, is to save a life, because the decision making about whether or not this is appropriate, whether it should be continued, can always come at any point, sooner or later after that. But then finding that point when the intensive care is not achieving the goals is the nub of the work, and recognizing when the therapy is becoming more burdensome than potentially beneficial, and going through that process in my own judgment, and helping a patient and a family through that process at the same time.
Is that evident to you one, two or three days in? What is the trajectory for that?
It's very variable. It may be immediately obvious in someone who has very, very advanced disease of a variety of kinds who has been very severely debilitated for a long time before the critical illness. It may be immediately obvious that the kind of care that we are best equipped to give in the intensive care unit is really not going to serve that person's most appropriate and realistic goals.
But in other situations, it's not immediately obvious, but it becomes obvious over a period of hours or days, sometimes weeks. And the way it becomes clearer is that we get a chance to see how well the intensive care therapies are working. We also get a chance to continue communicating with his family because the patient is often too ill to speak to us. We get a clearer sense of what this patient's values are, what their goals are, what's acceptable to them as a quality of life. And then we have to think about whether what is medically achievable is within the range of what that person's goals are.
Describe the movable line between prolonging death and extending life.
The distinction between prolonging a life and prolonging a death is a very key distinction, and it is a difficult line to draw, and I take that process more seriously than probably anything else I do in my job. I think the line gets drawn usually in a process in which we can see what the outcome is likely to be, and that outcome usually becomes clearer and clearer as time goes by, as we can evaluate how the intensive care therapies are working. It also becomes clearer and clearer to us whether this outcome that we believe we can achieve or not achieve is going to be acceptable to this patient and his family. It may be a death outcome, or it may be the inability to return to an independent life, which for some people is a worse outcome actually than death.
If it becomes clear what is likely to happen, and we share that information with the people who love this patient, and we decide together whether the expectation of that outcome is strong enough and that outcome is unacceptable enough, then continuing the care in the ICU is not serving that person's goal. Then it becomes prolonging something other than physiological function and life to achieve a goal that's acceptable to that patient.
What does prolonged death look like?
Although we can never be 100 percent certain until the moment of death that someone is dying, there are clinical situations where the odds are so overwhelming that someone can['t] survive the hospitalization in a condition that they would find acceptable, that we can see that outcome and compare that with the burden of the treatment. When it is virtually a foregone conclusion that that unacceptable outcome is going to occur, then using this technology to support the physiology of the patient doesn't make sense. And it is invasive, and it's burdensome.
It's a situation in which a person is completely dependent for all of their care on a nurse and a physician; where the patient cannot even attend to their most personal care and has to be cleaned from head to toe and every place in between by another individual; when they're not even awake. And our nurses do that in the most unbelievably respectful way, but still, it's a part of this experience. It is being attached to machines with constant noise from alarms and signals. It is being surrounded by electrical devices and monitors, with no control over any of your bodily functions, quite literally. And although we strive as hard as we possibly can to prevent discomfort, it's probably impossible to prevent it at every moment. So there are discomforts, and one hopes rare but occasional pain and other kinds of distress, fear, delirium. All these things are occurring for people.
In addition to that, you're in a bed that has side rails to protect you from falling, but also may make it more difficult for the people who love you to get close to you. Even if there's open visiting in an ICU, which some ICUs have and some don't, it is not a place where loved ones move about freely. They're uncomfortable and unhappy and fearful. And all of that is part of the surrounding. So it's a very disconnected, depersonalizing and occasionally even painful and frightening experience. I don't think anybody wants to die that way. I think most of us, not everybody, but most of us would be willing to go through it for a good outcome, but nobody wants to be like that if nothing good is going to come of it.
I think that over time, the expectations that people come into an intensive care unit with become informed by the reality of the intensive care unit. It's not like television. It's a place where many people are facing losses. The family in one room can hear the crying from the other rooms. They walk past room after room of critically ill people, and it can be a very frightening environment. And people faced with that day after day after day, and with loved ones who often can't communicate with them and who are tethered to machines and wires and so on, start to think about, what if this doesn't go well? How will we look back on this? And is it worth this? And is there a good enough chance that something we all want is going to come out of this that we are going to take the risk of continuing this for too long?
And is it life?
And is it life? Or is it a life that's acceptable to this person? And will it be a death that's acceptable for this person, if this person has to die? And what will happen if the person doesn't die but is permanently dependent on life supports, or doesn't die and is not permanently dependent on life supports, and there's nothing to discontinue, but the person lives on and on and on in a condition that wouldn't be acceptable to them?
How sick are your patients in the ICU? What brings them into the ICU?
I work in a medical intensive care unit [MICU], and the patients are some of the very sickest patients in any hospital anywhere. There's more death in the intensive care unit than anywhere else in the hospital. In this day and age, in 2009-2010, most patients in intensive care units are older adults, and they come to the ICU with pre-existing chronic illnesses, and the critical illness is superimposed on that. Many of them have every major organ involved in the critical illness. Working on any one piece of it affects every other piece of it. The dysfunction of any one piece of it compounds a dysfunction of any other piece of it. So these are people who would not be alive if this technology had not been invented. ... Some of them have overwhelming odds of dying, some of them have a better chance, but they all are facing a life-threatening situation. In this day and age, the push is to have people not stay in a hospital very long.
They're usually sedated, so it's all about them and it's not about them.
Probably most patients in our intensive care unit are sedated. We try to sedate them a little bit less than we used to for a variety of scientific reasons, medical reasons, but they still are mostly sedated. And it's very, very easy for a patient in the ICU to lose what's called their personhood, to stop being them, because the things that are them are not the things that we can see at that time. I'm often struck when a family has put pictures on the wall or brought some other thing that is important to this person -- a blanket or a pillow or a piece of clothing or something special -- into the room, or even just cards, because they're such a stark reminder that this is an individual person, and this is somebody's brother, mother, person of significance to someone. And it is very easy for that to be lost unless we have constant reminders of that.
The families are there, too, and they are very much part of the picture. And ideally, we can get a good sense of who the patients are and who they are as a person from their families. And that takes time and listening.
When patients are sedated, does it make treatment decisions even more difficult? Is it easier to do too much because of that?
It's a hugely challenging factor to try to make a decision [that] is reflective of the choices and values and preferences of this patient when we cannot elicit those directly. And I am very well aware that people may even take steps to write down their preferences in advance or think about them at some point in the past, and that when actually faced with a serious illness, they may feel somewhat differently about it. So it's very difficult, even when we have written evidence of somebody's wishes at a particular time, to understand what they would want at this moment. And I've seen that kind of movement in both directions, both toward and away from intensive care therapies. And the only way that I can see around it is to try to listen for as much information as I can about who this person is and what they would decide -- to the extent that it's their decision -- if they were there in a way that they could interact with us.
What can you do to keep someone alive in the ICU? What is "doing everything"?
"Everything" is a very complicated concept. "Everything" actually needs to get deconstructed and explored in terms of its meaning for a particular patient and a particular family and for a particular ICU team and what makes sense in that particular context.
But everything includes potentially a respirator, which is an invasive machine that is attached to a patient via a tube that's in their trachea, goes through their vocal cords, after which they can't speak while it's in that position. It can include medicines to raise someone's blood pressure, if they're in shock with a blood pressure that's too low to support their circulation. It can include treatments to support kidney function -- dialysis and other similar treatments. It can include massive fluid and blood transfusions for somebody who is in shock or hemorrhaging. It can include a whole variety of cardiac therapies. And many of the medicines and things that we do are administered through special intravenous catheters that are placed in large veins, often in the neck. Patients have catheters in their arteries that monitor their blood pressure minute to minute. And all those monitors are hooked up to electrical screens, where the vital signs and other key physiologic information is portrayed on a screen. And that screen often dominates the room for everyone.
The ICU represents the bright and dark side of modern medicine. There's an incredible complexity to it.
I think that's right. People ask me what I do for a living, and they say, "I don't know how you could do it." It is my routine, and I am used to it in a certain way. On the other hand, a day doesn't go by, and usually not an hour, when the intensity of the intensive care unit and the power -- both the bright side and the dark side -- are not very, very apparent to me. So it's a very emotional job, although at the same time I recognize that I need to make decisions that are well informed and not carried away.
Nobody wants to die. Nobody wants to die. That's how we carry forward as a species. And at the same time, nobody wants to die badly. And that is my job. My job is to try to prevent people from dying if there's a possible way to do it that will preserve a quality of life that's acceptable to them, but if they can't go on, to try to make the death a good death. And most people do not want to live in an extremely compromised condition, particularly if they are not cognitively aware or if they are very, very severely functionally dependent -- which isn't to say that people aren't willing to live with extraordinary disability and make good lives. It's really quite remarkable what people are willing to live with. And if that's OK with them, my job is to try to accomplish that for them.
What kinds of things have surprised you? What has changed your idea of what could be a meaningful life?
I take care of a number of patients who have very serious chronic illnesses, people who have diseases that have caused them to lose parts of their body, make it impossible for them to walk, confine them to a very narrow range of places that they can go, people who have difficulty breathing. And yet most people are willing to accommodate. Most people are willing to redesign life to reduce it down to the essence for them and what it is that's absolutely most important as long as they can interact with the people they care about and do some of the things that are most important to them. Clearly there are lines that can be crossed, and it's very important for us to see that and not use intensive care to restore someone to a life that's not going to be acceptable to them. For example, many people don't want to be in a facility if they're going to be severely cognitively impaired and totally dependent for their care. So I have to balance that. ...
But that's not easy?
It's not really easy. The risk is that somebody ends up in a position that they absolutely do not want to be in, and they are very vulnerable and voiceless, and cannot extricate themselves from that position. And many people are in that position, too, people who are cognitively impaired, coming into a critical illness, who can't speak and have no one to speak for them, and I don't know what's right, and I don't want to make a decision for them. ...
There is a case on both sides, in some situations. I suppose there's always a case on both sides. These are decisions about whether people are going to be alive or not alive, and if they are alive, if they're going to be living with dependence on life supports or not, or in a quality of life that may or may not be acceptable to them.
How do you honor a patient's wishes when the patient or family don't know what those wishes are?
There's a big problem with communication. And I really feel that if we work on that very, very hard, that it is much easier for all of us to make decisions. It's easier for patients and their families, and it's easier for us. And that means that we really have to be talking with these patients and their families as much as we possibly can, and not just giving information to them, which has to be given in a way that they can absorb it and understand it, in lay terms. But it's a mutual exchange of information, and we've got to listen. And the research shows what we just know on a day-to-day basis and [from] our own experience, that doctors are very good at talking, and most of them are not as good at listening. So we have to work very hard on explaining to the best of our knowledge what we think is going to happen, and revisiting and revisiting it, and listening to try to understand who this person is, who usually can't communicate to us, and whether this, as it's now evolved, is potentially something acceptable as a life. That's the process.
There's a domino effect: One decision leads to consequences. A doctor might know the consequences far more than a family member might.
I think that most families facing these decisions really aren't fully informed -- not that anybody is deliberately concealing information from them. But for one thing, people in a crisis situation, in an intensive care unit, a family, it's very hard for people to absorb information. They're very distressed. It's very hard to hear. You could talk at people and not get the information across, so you need to be very careful and skillful about how the information gets communicated. And you need to be very much aware of the emotional background of the people coming into the situation. There's a cognitive channel, and there's an emotional channel. And if you're only tuning into the cognitive channel, you're just missing everything that there is to be seen and heard on the emotional channel.
So I think you really need to be very sensitive and very measured and paced about delivering information. And I think if you do it that way, that over a period of time that varies from patient to patient, family to family, people can get a better sense of what the implications are at different decision points. I don't think that's happening yet most of the time.
So that, for example, somebody who is unable to come off a mechanical ventilator in the ICU, and gets a tracheotomy placed to move the connection of the ventilator to the neck, and then go on for further efforts to take them off the respirator, that at least half of the time are not going to succeed, the research shows that those people don't understand that moment and what it means in terms of the future. Those people believe that, for the most part, that having a tracheotomy is a good thing; that it means that the doctors are going to continue to try to take the patient off the ventilator. And what they don't understand is: That moment when the patient needs that tracheotomy because they haven't come off the ventilator is a very serious prognostic indicator for the future, meaning it tells us that we need to be very, very concerned about what the outcome of this patient is going to be. So that moment really needs to be captured in a extensive communication. And it's not that at that moment I expect or want a family to decide, "Let's not do the tracheotomy." It's just that I want them to know then that the risks are high, so that when, two to three weeks later, things are not going well, they will at least feel that they were alerted that things might not go well, as opposed to having expectations that were never modulated to be more realistic. ...
So I think we need to give people that information and the time to absorb the situation as it evolves. Otherwise, the expectations are just dashed in a moment. And that's when the anger and the guilt and the regret are so deep and so intense. And I think we can avoid that to a large extent in the intensive care unit, by anticipating the possible outcomes with the people who have to make the decisions with us. ...
What kind of decisions will family members make?
Common decisions that come up in the intensive care unit are: Are we going to use or continue certain kinds of intensive care therapies? So, are we going to use a respirator? Or, if the patient is already on the respirator, are we going to continue the patient on the respirator? Are we going to use some other form of organ support, like a dialysis treatment? Or, if the patient is already on dialysis treatment, are we going to continue that treatment? Those are some of the decisions that have to get made.
... People who are asked to make those choices just in that stark way may lose the fact that it's the underlying illness that results in the patient's poor outcome and not something that they did. I think we really need to be sensitive to that and to help them more. It's really not fair to simply offer a choice among many options to a family that is not in a position to really exercise that choice in an informed way and will potentially carry a long-term burden of making a choice that they think was solely up to them and affected their loved one's outcome in a pivotal way.
How prepared are families? What do they want to know from you about that?
In my experience, most patients and families, and even some clinicians, may not be aware of what happens to a patient who lives through the acute period of critical illness but doesn't actually recover. So we have a "survival without recovery" situation that is very grave. And we have people continuing on the respirator, and often on other life supports, who may never come off. And these are the broken survivors of intensive care. And the better intensive care gets, the more of these broken survivors we have, and we have to contend with them. This is our responsibility as intensive care physicians. We cannot just discharge them from the intensive care unit and congratulate ourselves that these people survived the ICU, because they and their families are living with all of the shattered pieces.
Research shows that these people have a lot of distressing physical and emotional symptoms. And it's a worse disease than cancer in terms of the mortality. Most of these people die within six months after they leave the ICU or hospital, and very few of them, particularly in the older-adult group, ever go home in a good, functional condition. ...
Talk about the complications for family members, the difference between starting and stopping a treatment.
The legal experts and the ethicists tell us that forgoing life-sustaining treatment is a decision that can be made, whether it involves withholding a therapy, not giving it at all or withdrawing it after it's given. Yet in my clinical experience, for almost everybody involved -- the families, the patients if they're involved in this decision, and the clinicians -- it feels much more difficult to stop something that's already been started. For the families in particular, I think it's very, very important to try to make it clear to them that withdrawing a therapy is allowing a death to occur that is naturally going to occur, and is not their fault, and is not the proximate result of their decision.
Many people feel that making a decision to discontinue an intensive care therapy is either euthanasia or murder. So I think this perspective on the role of the underlying disease and illness in leading to the death is very important to convey and to help people to realize.
It's an extraordinary responsibility and power. Most families look to you as guide.
... The truth of it is that all of these outcomes are mostly preordained by the underlying illness and the disease. The availability of the therapies has created this fiction that we can orchestrate this, one way or the other, when the truth of it is that for all of this magnificent technology, the underlying illness and the medical condition of the patient are far and away the most important factors in determining the outcome. But it feels like, when you have the technology available, that your decisions to use or not use it are like the decisions to allow life or not allow life. And that's not a position that any of us wants to be in.
So I really agonize over these decisions, and I guess I err on the side of using the technology for as long as it seems to me that there's some reasonable prospect that something good may come of it, as opposed to foreshortening the care. But in the end, we need to allow a natural death sometimes, and we don't have the technology to stop it. And that's not our fault.
Is there even more anguish because it's possible you could be wrong?
The uncertainty is the most disturbing part of the decision making. Critical care medicine has some of the most sophisticated predictive models of any field of medicine. You can assemble a huge number of factors related to a patient and other things and plug them into a predictive equation. But in the end, you really don't have anything that tells you what's going to happen to this individual that you're caring for, and so you have to live with that uncertainty, and you have to do the best that you can. I think you have to acknowledge it in the discussions that you have with the patients and their families, and actually there's evidence that families and patients know of this uncertainty and understand it, and understand that we have to work within the context of it, and that all we can do is give them our very best guess, and structure the process in a way that we don't make monumental decisions until things become pretty clear. And that may take a little time.
Should families be deciding these things? Is it well-intended policy gone wrong?
I think that in trying to give patients and families a voice in their care, which they should have, that we ended up saddling them with extraordinarily difficult decisions that burden them and cause them to struggle with decision making without information when they're in an emotional crisis. I think we as clinicians need to accept our responsibility here, and we cannot escape by laying out a menu of options to people who are really not in a position to choose. We have to explain these things, in a way. And there are certain decisions that are simply medical decisions, that relate to the burdens and benefits of therapy. And that's my professional responsibility. It's not up to anybody to make a decision that a treatment that is more harmful than is helpful should not be given. That's not allowed. And so I have to look at every situation and take those responsibilities on myself, and to share the ones that are less clear, where we all need to weigh burdens and benefits, and to be sure that all the decisions are informed by the values and goals of this individual, and so to listen to those families. So I think we have burdened people with decision making that they experience as being very distressing at times.
And again, this is very variable. There's research about this. There's research about the variation in families' wishes to receive recommendations from physicians about medical decision making, about the extent to which they want to play a role in the decision making, and of course there's great international variation in this. There are countries in which the doctors still, by and large, make all of the decisions. And we have the other extreme. I think we need to find a better resolution of the competing concerns here: The concern to make sure that patients get to exercise their autonomy; and the concerns that patients and their families don't get burdened with medical decisions that they can't possibly be in a position to make, and that only cause them distress.
How do most of your patients die in the ICU?
Like most ICUs across the country at this point, most patients die after some kind of decision to limit life-sustaining treatments. We're so good at keeping things going that unless there's a decision to limit something, it takes a lot longer for a death to occur. That represents a big change over the last couple of decades.
There are many deaths that still occur on full life-sustaining treatment and deaths that are preceded by attempts to perform resuscitation, further resuscitation, even after weeks of ICU resuscitation. And cardiopulmonary resuscitation -- CPR -- in the ICU is not like it seems on television. In fact, there's a wonderful article in the New England Journal of Medicine about TV CPR compared to real CPR. Real CPR has many people surrounding the bedside of the dying person, usually not anyone they know or anyone they love. Usually families are not there. Some ICUs will now allow families to be present. People have their gowns removed because they have different devices attached to them. They may receive electrical shocks to try to jump-start their hearts again, and they may receive chest compressions, where somebody is pressing repeatedly on their chest, and medications are administered. And that's a different kind of death.
Again, if the resuscitation could actually bring somebody back to a quality of life that they would want, then it's worth doing. In an ICU, when something like that happens after a prolonged trial of intensive care, that is virtually never the case, or very, very rarely. ...
Extubation, the scheduling of death: It's a new kind of death in terms of the control. It's being scheduled.
One of the things that we discuss with people when there are decisions being made about discontinuing life support is when that might occur. And it's an unusual situation where we have a little bit more control over timing, although again, we discontinue the life support and the life goes on for some period of time. And that is not my decision.
But if people, for example, want to be sure that there's a loved one from out of town who would really like to be able to kiss this patient before they die, then we can adjust and accommodate the timing to some extent. It is very odd to plan things in that way, but we have so little control over everything that if we can act in some way to make it easier for the patient and their family, and more in line with what their hope is for a good death, then I think we should use that ability to manage the situation as best we can. But I'm always careful to tell people that there are a lot of surprises, and we never quite know what's going to happen. And in some ways that's very reassuring.
At this point, how much of the idea of a good death is for the patient, and how much for the family?
A lot of the time patients are not conscious. I think there are a number of players in this drama, and it has a profound impact on all of us. Certainly for this particular individual, for their family, it's a moment that's remembered forever, so I do want the family to feel that this scene is as close to what can be achieved in this set of circumstances as it can be. That includes a lot of personal care to the patient, like brushing their hair, trying to remove tape and residuum of tape to make them seem as if they're in their bed as opposed to in a cubicle in a hospital.
I think it is very important for the family. I can't know how the patients experience it, but I'm going to presume that it's important to them. It is very certainly important for those of us who care for these patients in the ICU. I've come to appreciate more and more what it means for our nurses, who may or may not be part of the decision-making process about the withdrawal of the treatment, but who are implementing every part of this care plan right through and beyond the end. It's the nurses who come afterward to care for patients after their death and ready them for their move out of the ICU.
How difficult is it for you to do something that a family wants when you don't believe in it?
This is a very hard part of the job. There are two really important principles that I bring to this situation, one of which is that I've got to be true to the patient and I've got to be true to my responsibility to provide the best possible care for the patient. The second is that I've got to support that family, and if we don't see things in the same way, I've got to do everything I can possibly do to get us to a shared perspective.
So I need to understand why we may not agree initially and try to work it out in such a way that we can have some shared goals and something that's acceptable. And if we can't, and if the family really feels strongly about something -- and this is not so common -- but if they feel very strongly about something that I might not recommend, I still feel I need to support that family as long as it's not harmful to the patient, because I can't imagine anything worse than feeling that you need to make a certain decision, and then feeling ashamed of it somehow or other in the eye of your loved one's physician. I think to lose that support from the clinician is very devastating for families.
What about differing views of other physicians in the MICU? How do those tensions play out?
As the patients get more and more ill, the complexity of the disease processes increases, and more and more organs become involved, and that may mean that more and more clinicians become involved. The ICU is a place that is filled with teams of clinicians with different specialties and different points of view, all trying to contribute to the care of this particular individual, and it's completely predictable that there would be some differences of opinions about how things should go, and different styles of practice. In some ways that's a good thing, because it's good to get different perspectives, because you really shouldn't be too sure. On the other hand, it can become very confusing, certainly for the families and the patients, who may get conflicting information from different teams, and also for the clinicians themselves. It can also operate to diffuse the responsibility for the care to the point where the ship doesn't have a captain.
This is a huge issue in complex illnesses and in an era of specialization. I think one of the important jobs of the ICU doctor is to try to integrate the input of all these various specialists in a way that is coherent both to the ICU team, to the rest of the clinicians who are involved, and to this patient and family. Again, this is often just an issue of really good communication and having people in the same room and having a dialogue.
I think the challenges of finding shared goals with patients and families are the same as the challenges with clinicians, and they're the same as the ones that we have in every relationship in our lives. I've actually learned a lot in dealing with patients and families, other clinicians.
Can you talk about that shift when a patient nears death?
I think that for a cancer patient or a patient who might receive a transplant or the many other ways in which people can become very ill, it may be that the physician who has provided extraordinary treatments all the way along has a great deal of personal difficulty in letting it go. ... When this comes up, I think it's really very important not to shut somebody, like a clinician, out of the decision-making process. Clearly, if we dismiss this kind of perspective as completely unrealistic or destructive, it's a big problem not just for them but also for these patients and their families, for whom these people have been the physician from the beginning of the illness and who have done enormous amount of good. So I think all of us have to continue to participate in the decision-making process, talk it out, and most of the time all of us will come around to a shared perspective at some point. Again, I think that because I'm not usually involved over a long period of time with these patients and their families, it may be a little bit easier for me to take a fresh look at the situation and to see when the ongoing treatments are not likely to achieve the goals that these patients and families want.
With Norman Smellie, you and Dr. [Keren] Osman agreed that his death was near, yet she had been in a position where the patient's wishes were confusing. What happened?
Mr. Smellie had had a bone marrow transplant. ... [He] had been in the hospital for months actually before he came to the intensive care unit. During that period of time, he was cared for by his oncology and transplant physicians, very meticulously trying to track down any possible explanation for his symptoms and his difficulties. But things continued to deteriorate, and he eventually ended up in the intensive care unit with very severe pain that had been persistent despite many, many efforts, and seemed to be related to conditions that were not reversible. He simply wasn't getting back on his feet, in any event. He was really bed-bound, tremendously bad symptoms, constant difficulty with his bowels and incapacitated -- and eventually increasingly confused, which made the situation even more difficult, because at that point, when it seemed like the circumstances for the context of the decisions were changing, he was less able to participate directly in the decision making.
So as things began to worsen, everyone was trying to get a better understanding of what Norman Smellie would be saying if he could speak to us clearly at this particular time, and would he still feel the same way in the face of the irreversibility of the situation.
... In this family there was, at least initially, a difference of opinion about what the patient would want in this set of circumstances. His health care proxy, who was his sister, had spoken to him about these issues at the time she was appointed as health care proxy, which was during the hospitalization. It was at that time that he had said, "I'm a fighter, and let's go." His girlfriend, who was also by his side throughout this process, had subsequent conversations with him in which he expressed increasing distress about his situation, increasing dissatisfaction with the quality of his life, the pain, and then as time went by, his increasing awareness of the irreversibility of the problem. And he indicated to his girlfriend that he would not want to have extraordinary measures if he became critically ill or had a cardiac arrest. So there was this conflicting set of perspectives at the [time] that these decisions had to get made. ...
What was so striking to me -- and we had many meetings with his family, and I think they ended up being very productive -- [was that] I think that his family members came to a consensus, and we got a better understanding through them of him.
... A decision was finally made after these many months in the hospital and his continued deterioration that he would not be placed on a respirator and that he would not have CPR if his heart stopped after all that. His goal of care would exclusively be his comfort. ... And he died within 36 hours of leaving the ICU, cared for by his loved ones, who were very grateful for the entire process, even though grieving a loss.
The disagreement between Jima, Norman's girlfriend, and the health care proxy Phyllis, Norman's sister. How common is that?
We do have disagreements within families, as we do between others. Usually they are best resolved by focusing the family members on what the patient would say if the patient was in the situation. In Norm Smellie's case, the issue was what he would say, and they had had different conversations with him. They were able to come to a resolution of their perspectives to find a common ground, and that was actually that the medical situation had really changed since he expressed his first views about it, and they came to an agreement that, faced with the new reality, that he would have shifted his perspective.
But usually those conflicts, when they arise, reflect the different perspectives of the family members on their own perception of the situation rather than their views of what the patient would want. So we can usually get them resolved if we can focus where we should focus, which is, how can we best achieve what we believe the patient would want? ... I think there's a lot of ambivalence in patients, in families and in physicians.
Norm was a physician's assistant, and Jima was a nurse practitioner. Did it make the decision harder, knowing what each decision might mean?
I think he was better informed, but in the end, I think we're all patients when we are in the patient's role, so I think that Norm had the same hopes and fears as other patients. I think he had more information, but I think he ended up with the same hopes and fears and internal struggles and ambivalences.
Tell me about Diana Reid.
I cared for a patient some months ago who was a very young woman, in her early 30s, who had scleroderma, a very bad disease which resulted in changes in her skin, changes in many organs of her body, and eventually the loss of parts of all of her fingers and an inability to eat and difficulty with her breathing. But the illness was relentlessly progressive, and she had more and more difficulty and came to the hospital. ... Initially the ICU team was able to get her back to where she had been before she came to the ICU, but still with this terrible chronic disease. And she went back to another part of the hospital, still struggling with these problems, and then deteriorated again with terrible difficulty breathing, facing going back on a respirator, really gasping for air. [She] came back to the ICU in that condition.
I found myself face to face with her, having never met her before. There had not been an extensive discussion before then about these issues. And she was really so ill and so urgently in need of intensive support, if she was going to be having it, that it was virtually impossible to have a conversation with her.
Her surrogate decision maker, her sister, was not on the ICU. She was at home, fairly far away. And all I could really do was the best I could in that moment, which was to arrive at an understanding with that [in mind], because the situation was so emergent, and that we would start respirator treatment. We would put her on the ventilator, we would give her intensive care therapy, and when her sister arrived, we would discuss it further; that she could always come off it if that seemed appropriate. And she indicated to me that if she wasn't getting better, that she just did not want to have a prolonged course on the respirator. So with that understanding, she went on the respirator and other intensive care therapies, sedated and unable to communicate.
Over the next days, we all struggled to understand what would be the right thing. She was not getting better quickly, and even if she was able to get through this critical illness, which was not possible, she was facing all these other issues. We had many discussions with her sister and her mother. And in the end, she remained on the respirator. A decision was made by her family not to attempt resuscitation if her heart stopped. I'm virtually certain again that that would not have succeeded, and they did not want to put her through that.
After the meeting we filmed with you, the sister said she wanted everything done.
We had a number of conversations with her family. ... Her sister agonized over this decision as a health care proxy. She expressed the feeling that if she made a decision to discontinue the respirator or any other therapy that she would be killing her sister, and she just couldn't take that step, so she agonized about it. And it was very difficult for all the caregivers.
What happens in end-of-life conversations? What do you say to families?
[People] don't come to the ICU for the end of life. That's not why they come. They come there because they would die if they didn't come there, and the idea is that some extraordinary effort is going to get made so that they won't die. And yet many of them do. So the conversation in the ICU is a different one than the conversation with someone who is on a downhill course over a longer period of time, who has not come with the expectation of some miraculous technologic[al] intervention.
The conversations, wherever they occur and in whatever context, are always difficult. And people, whether they're coming to an ICU or not, often have very high expectations of what medicine can achieve, even if they have a long, chronic, difficult illness that is inexorably going downhill with a few spiky improvements in between. So we are having these conversations in the context of at least a possibility that the death might not occur. And that is the biggest challenge of all.
What I try to do is to make sure that people are alerted to the possibility of death so that this process of comprehending that possibility can be set in motion, hopefully enough in advance that the person has time to absorb it a little bit and process it.
Are you generally direct? Do you say the words "dying" and "death"?
Often I'm not sure of what's going to happen, so what I would say more often is: "I'm very worried that things will not go well. I fear that this person may not be able to leave this intensive care unit alive, or that this person will be, if they do survive this illness and the ICU, will remain dependent on life supports or will never recover to an independent life." If someone is more obviously dying, then I would say that. I do try to use the word "dying" or "death," because it just is avoided so often, and yet it's in the air somewhere, and there is something very useful about having it out in the open if that's appropriate, if death is a possibility.
We have to be very careful whenever we have these conversations to get an idea of what the person on the other side of this conversation is prepared to discuss. I try to ask whether it would be OK to talk about what I think is going to happen or what has happened or what is happening, and would they like to hear about that, and would it be appropriate for me to discuss it.
In most cases, people want that information, but not always. I think it's really up to them. I don't want to withhold information if people want it, but I do not want to assault them with information if they don't.
What do families generally want to ask in those conversations?
I think first and foremost, people want to understand what's going on. They are in a situation that is completely foreign. Many physicians and nurses feel like the ICU is a foreign place. There's research in France on intensive care unit families, showing that they do not comprehend the most basic information about what treatments the patient is receiving, what is the most important organ involved in the patient's illness, and whether there's a risk that the patient might die.
Then I think people have a fundamental question: "Is this treatment worth the burdens of using it? Why are you doing this to my loved one? Does it make sense? And will it achieve something that we want?" They may not ask the question in that way, but I think that's the question that has to be answered.
How do they usually ask it?
Sometimes people are asking for minute-to-minute information and not asking about the big picture. When we give the minute-to-minute information, we need to frame it in a larger picture so that people can put these pieces together in some way that makes sense. ... And some people are asking for the big picture.
... Interestingly, people don't ask as much as I might have thought. I don't think they necessarily expect to get information, and they're hesitant to bother somebody to get the information, and they're maybe afraid of some of the answers, and I think they're just overwhelmed. So I think we have to be proactive about these conversations.
In the big scheme of things, do those conversations matter?
I think those conversations are absolutely everything. I get letters from the families of patients who die, a lot, thanking me for the care, and what people are grateful for, whether the person lives through the critical illness or dies from the critical illness, is a sense that we took time to explain what was happening, why we were doing what we were doing, and to try to understand who their loved one was, and how this care relates to their loved one's goals and values and preferences.
Broken survivor population is increasing. What are the consequences for society?
This is a serious national health problem. The estimates are that at any given time, 100,000 people are chronically critically ill, on ventilators, in various kinds of facilities -- respiratory care units; some of them stay in ICUs; long-term acute care hospitals -- all over the country. Most of them are older adults. So there will be more and more as the population continues to age and as we continue to offer more aggressive treatments to older adults. It's estimated that the cost of caring for these people is in the $20 to $25 billion range annually.
This is a most serious problem, and it is really interesting that it has not gotten a lot of attention. It needs to get a lot more attention, both from policy-makers but also from scientists, from clinicians in the intensive care units, who need to understand it very, very clearly and explain it to patients and families if that's appropriate. Our house staff need to be trained about it. We need to know where our people are going after we do what we do in the ICU. That's one of the most important ways that we can measure the value of what we're doing in the ICU.