Innovations in End-of-Life Care
an international journal of leaders in end-of-life care

An Interview with Robin L. Fainsinger, MBchB, CCFP

How do you define "quality of life"?

Now that is a question I think about a lot. We all define quality of life with the same motherhood statements. Essentially, we have come up with the same broad domains—the physical, psychosocial, spiritual, existential. When we define quality of life in palliative care, we're talking about maximizing patient and family comfort across these domains. Quality of life is a complex mix of those elements, with many nuances within that broad framework.

The patient and family are always the unit of care. We talk about function and autonomy—the sense of control that people require. We talk about advance care planning. We talk about patient and family satisfaction, well-being, and the perception of care received. We talk about the family burden and the need to relieve that burden, while at the same time strengthening relationships among family members. And then, we need to remember bereavement. On the health care provider side, we talk about provider continuity and skill, which the health care system needs to provide. Quality of life includes all of these elements.

It has been suggested that we need to be careful about our definitions because when we, as health care professionals, talk about these domains, we can get stuck with a narrow interpretation that perhaps serves our administrative or financial purposes. Whereas, what families and patients define as quality of life may not necessarily be the same sorts of things we are attempting to measure to justify our existence to administrators and the people that finance our programs. These multiple perspectives complicate the definition of quality of life.

The thing that bothers me most about how we define this concept is this: What we measure in numbers to report to the people to whom we justify the existence of our palliative care programs is relatively simple, compared to what we would need to measure to prove to ourselves that we really are providing quality of life to patients and families. One factor that makes this question complex and that disturbs me is this: When should we measure quality of life? Do we measure it at the point of entry to the program? Do we measure it at some kind of defined midpoint? How do we measure it toward the end of life, when patients may no longer be able to report on their quality of life? We certainly can't get any measurements from patients after death.

We can get measurements from families, but their perceptions are by definition colored by their experience—"I'm exhausted now. My relative's going to pass away in the next few days, and this has been a horribly exhausting experience for me." The fact that we may have provided exemplary 24-hour physical and psychosocial support will not necessarily change the fact that the caregiver is exhausted. So then, if you ask a family member a week into bereavement, he or she may still be exhausted. If you ask the person three months or six months later, he or she may be feeling a lot better about things. Or, at that point, the memory of the relative's death may have focused on the fact that although you provided exemplary care for six months, in the last six hours of life a phone call was missed, or a physician or nurse didn't show up at the right time. And so deciding what makes up this concept becomes complex. How to accurately measure quality of life is very troubling. And, for whom are we measuring it?

Things become more complicated with defining and measuring quality of life when we care for patients who die at home. We would certainly do this with the motivation to improve the quality of life for our patients. However, we need to examine how we measure the financial, physical, and psychological cost to the quality of life of the families who provide this care in often difficult circumstances.

You're raising many interesting and thorny issues. What do you do on the ground, since you're running these programs?

At the end of the day, I think we do what we practically can. And we do it at different levels for different purposes. There are clinical, teaching, research, and administrative purposes for measuring the concept of quality of life. We look at the clinical aspects so as to ensure that we are providing the best quality of life to patients. We look at quality of life for teaching purposes because we learn things that help us teach some of the aspects of physical and psychosocial support for patients and families. In terms of our administrative purposes, it becomes easier to measure quality of life in a more global way in order to prove that the program is meeting its goals.

Let me review what those initial goals were. Dr. Eduardo Bruera designed the Edmonton program with the vision of developing a comprehensive program to increase the access of patients with terminal cancer to palliative care services. And that was the key. This was what we were interested in. But the second key goal was to decrease the number of cancer patients who die in acute care facilities. Dr. Bruera and others focused particularly on cancer-related deaths simply because—for many of us—that is the largest population we look after. And from a statistical point of view, it is easier to measure and report on this group because it is a little easier to define. This second goal gave us a financial rationale for the program. If our program could generate cost savings, we could justify both why the health care system needed us, and why we needed the funds to develop the hospices and provide community palliative care. The other goal of the program was to increase participation of family physicians in the care of terminally ill patients.

An example of an administrative use of the concept of quality of life is that ever since the program was established, we have been able to show that we've met these goals. The number of patients with cancer-related deaths who had access to the palliative care program has gone from around 20–25 percent of patients who saw a palliative consult team before the program, to around 85 percent of advanced cancer patients who now see a palliative care consult team at some point in their final illness. If patients were in the tertiary unit, that contact would be a lot more intensive. If the patient's care had been relatively straightforward—i.e., the patient was seen in an acute care hospital, went home, and was seen by the family physician—the contact would have been shorter. But that's the bottom line—more patients suffering from cancer now have access to palliative care.

By measuring the number of inpatient days for cancer-related deaths both before the program was instituted and then afterward, we showed that we were able to decrease the number of inpatient days for cancer-related deaths from around 25,000 to 7,000. When you calculate that cost savings, depending on what you pick as the cost of an acute hospital bed, it is a considerable amount of money.

Let's use a patient example to flesh out how the program works. Take Mrs. Smith. She's at home, suffering from advanced emphysema, on oxygen, and not getting out of bed. She's got a weak heart. In the United States, if she had a family caregiver and an astute physician, she'd be a candidate for the Medicare Hospice Benefit. Without a family caregiver, she would need to be able to afford supplemental home care to stay at home. What would be available to Mrs. Smith in Edmonton, if she didn't have someone who could care for her 24 hours a day, but she wanted to remain at home?

No program is perfect, and you have just hit upon our Achilles heel. Patients in Mrs. Smith's situation have very little chance of remaining at home for the rest of their lives, unless they have enough money to pay for a 24-hour caregiver. Home care can provide a lot of services in the home, and we, the palliative care program, can provide all the symptomatic and supportive advice possible. But if there is no one to provide physical care to a bedridden, confused patient, it is impossible for that person to remain at home in most circumstances. Home care can provide about 4 to 5 hours of a personal care attendant in the home, which means there's still another 19 or 20 hours a day that the family must fill in, or services have to be bought. In this regard, the situation is quite similar to that in the United States. There are exceptional circumstances; if someone clearly looks like he or she will pass away in the next couple of days, sometimes the home care services are able to put in 24-hour care for that short period. But for the person who has become bedridden for a number of days or weeks at the end of his or her life, the services don't exist. That is why we have the hospices.

So that person would be a candidate for the inpatient hospice?

That person would absolutely be a candidate for hospice. The reality is that from a simply administrative, economic point of view, it's much—I hate to use the word "cheaper"—but it makes more economic sense to have that person in an inpatient hospice unit where you have one nurse looking after 6 or 7 patients, rather than 6 or 7 nurses you'd require if those patients were each at home.

Is the Edmonton Palliative Care Program financed through the Canadian national health system?

All the services I've described, including the home care services—although limited—are covered. There's a nominal charge for being in one of the continuing palliative care units—$15 a day—which, for most people, is way less than they would get in pension money in any event. So for most people, that nominal cost is not a hardship.

There are some gaps in any system. For example, although it is a nominal cost, if the patient is relatively young—not in the seniors group—and has a family and a mortgage, then that nominal cost can be a burden. Also, there are hidden costs with being at home, in terms of some dressings, or the tubing of pain pumps, or perhaps some sort of technical thing that may not be covered 100 percent by insurance.

It also sounds like the hospitals—or hospices, since you have various entities—are not in financial competition with one another. Is that correct?

There's absolutely no financial competition among any of these institutions. All of the hospices have plenty of work. There's no financial or profit margin one must prove to justify the institution's existence. What we have to prove to justify our existence is that we are doing what we say we are doing. Patients across the region have access to palliative care, and we are looking after them in the most appropriate setting for their needs, which relates to patients' and families' quality of life.

Can you say more about the communication across the different places in which patients are cared for, so that as patients move across settings, the staff in each is aware of what health care providers in other settings are doing?

That's exactly why it's set up with this web. The programs are interconnected, so that we could attempt to have a seamless flow of patients from one place to another. The physicians are part of the same structure. We meet in the same places. We have the same common educational rounds, and sit on the same administrative committees. We answer to the same authority within this complex web. So, if the patient goes from the Royal Alexandra Hospital to the tertiary palliative care unit, or to a hospice, or to the home, or to the Cross Cancer Institute, we know whom we need to contact to ensure that the palliative care services we started continue. In addition, we use the same assessment tools to measure symptoms, cognition, and pain syndromes. This gives us a "common language" in the transfer of information.

Let's return to the nitty gritty world of patients—how do you take this amorphous and multifaceted term, "quality of life," and apply it in practice?

From a bedside point of view, we have a number of outcome measures that we utilize throughout the city, either on a daily basis or two or three times a week, depending on the setting of the patient. Probably, the most useful tool I would point to is the Edmonton Symptom Assessments Scale (ESAS),* which is a scale with two versions: a numerical and a visual analogue scale, on a number of domains, from the physical to the psychosocial. The domains it assesses are pain, tiredness, nausea, depression, anxiety, drowsiness, appetite, feeling of well-being, and shortness of breath. There is also a space for "other problems" elicited from the patient. In the numerical version, on a scale from 0 to 10, the person indicates how much of a problem that item is at this point in time, with 0 being no problem, and 10 being the worst possible. In the visual analogue, the patient sees a continuum and makes a mark on that line to indicate how much of a problem that domain is.

Now, there are many quality of life scales. There's a confusing array of different ways we can measure symptoms. What I like about the ESAS is that it is extremely short. It takes the least time of any scale I'm aware of. It's the most straightforward and simple way for people to—on a repetitive basis—tell you how they're feeling right now. Using this same scale over time, we have a way of following the progress of people's symptoms.

Some of the so-called physical symptoms are perhaps the easiest to track. People have sometimes asked, "What do you mean by depression? And what do you mean by best feeling of well-being?" I would point to some of the research where, for example, people like Harvey Chochinov have shown that one of the best screening tools for depression is a single question: "Are you depressed?"1 And essentially, that's what we're doing with this question. Robin Cohen showed that asking people many questions about quality of life also correlated with a single question that essentially was a global question about their well-being. For example, the feeling of well-being visual analogue: "How are you, given all these problems? Tell me how you think you're feeling right now." That's essentially what we ask people when we say, "How are you?" Most of the time when we do that in a polite social situation, we do not take the time to listen to the answer. But we are being serious when we ask this question of our patients. So it is a simple tool, and it is easy to administer. We use it as a standard across the city, in all settings. And all of this information goes into an extremely large database of information collected in the home, in hospices, in the consulting services, and in the tertiary unit.

In this way, we do have a simple, clinical quality of life measurement that we can follow over time, both to assess the result of our interventions and to use clinically with patients. We use this tool to open a conversation with patients about their current assessment of their quality of life. When a patient indicates that, for example, his or her anxiety is 8 out of 10, and yet we had not realized that this person was anxious, the ESAS gives us a wonderful clinical opportunity to say, "Well, I noticed that you assessed your anxiety in this way. Let's talk about that a little bit more; what are you anxious about?" So from a bedside point of view, it is clinically useful.

From a teaching point of view, it is also a useful learning tool. We can show students how we picked up that a patient is experiencing pain, intervene to address the pain, and then observe with gratification as the pain visual analogue disappears, down to close to zero with appropriate treatment.

We also use other assessment tools, including a staging system for cancer pain. We follow measures of cognition, and we use screening questions for past history of alcohol abuse.2

We attempt to improve the quality of life for our patients by using quality of life assessments to identify the patients with the most problematic issues. We want to ensure that we then look after those patients in the component of our program most suited to their needs. For example, the comprehensive, well-resourced tertiary palliative care unit would obviously be the best setting for patients with the most complex physical and psychosocial issues. Use of our quality of life assessments helps us identify these patients and then track the impact we are having on them as we continue to treat them and monitor their progress. At the same time, if administrators were to challenge us to demonstrate that we are using our resources appropriately, we can use this information for that purpose as well.

Are you able to use the information you get from these tools, in terms of alleviating these symptoms or problems other than pain?

From a day-to-day clinical point of view, we find the tool very useful in terms of picking up things we may otherwise have missed. It allows us to respond—to ask patients and families to clarify what it is they are trying to tell us, and to monitor things on a day-to-day basis. We have also used these tools in many research publications. And we use them on a monthly basis with the administrative authorities to demonstrate that we are doing what we set out to do.

How do you address the needs of families, in terms of quality of life?

We don't have a quality of life tool that we use on a regular basis to measure what we are doing for families, which is probably an area we should be looking at, in fact. One of the practical problems of measuring quality of life is that even some of the simple things I have talked about are difficult to do. Many programs balk at the idea of measuring anything. It's seen as being intrusive at a time when patients and families are having some difficulty.

How do you respond to those who feel that assessment is burdensome to dying patients: "I don't want you doing any assessments. This patient is dying. I don't want any research or measuring of this patient. I just want you to care for this patient's clinical symptoms and psychosocial needs, but no tools, or assessment, or research"?

Our answer is to call attention to the paradox here. Just for a moment, think about the fact that one of the criticisms of pain and other symptom management is that health care providers do not ask about the patient's experience. What we're doing is, we are asking. And we are asking a lot. As a result, most people appreciate the fact that they are being assessed and that staff members are taking the time to ask them about these symptoms. But we have a wide, diverse world of patients and families, and not everyone tolerates a lot of questions.

One of the ways to respond to this resistance is to explain what it is that we are attempting to do, and why we are asking patients questions about their ability to concentrate and remember things. We explain our concern about how medications may affect the person, and our need to understand a wide range of different symptoms so that we can discuss those symptoms with them and attempt to respond to the symptoms. Most people who have some initial negative reaction will then understand the value of the tool. So, the vast majority of people will then either be happy because it wasn't a problem in the beginning, or because they have understood the explanation. There will be the odd individuals that just refuse, and we have to respect people's right to be cared for the way they want, and acknowledge that we will not have measurement outcomes in a few situations.

We do have another tool called the Edmonton Comfort Assessment Form (ECAF). One of the problems with the ESAS is that, ideally, it's meant to be filled in by the patients. But as patients deteriorate, it becomes more and more difficult to have them complete the tool, and we then end up doing proxy assessments with staff or families completing them. We noticed that when families complete the ESAS as proxies for patients, they report pain as being the most difficult problem, often in a patient where the pain had been previously well-controlled. So, the family is filling in that the pain is a horrible problem now that the patient is semi-comatose and unresponsive. What we have done is ask the family to explain what are they seeing that makes them believe the patient is in pain.

What we found is that people see many things—including agitated confusion, or the fact that the patient appears to be breathing relatively fast, or that the patient is not eating—whatever the family observes, they have a huge tendency to label these observations as pain.

This observation on our part has been a great learning opportunity about the gap in the care that we're providing to families. We've perhaps missed and have not explained well enough to families how we understand the symptoms they are observing. And what is significant is that we thought we were doing it, but, obviously, for some families, not well enough. We would meet family members weeks, or perhaps months, after their relative had died, and we would remember that the relative became unresponsive, and appeared to pass away with no discomfort. But when we met with family members, they would tell us, "Oh, it was awful! My mother had a horrible time." And yet, we hadn't recollected the passing in at all the same way. It was this dissonance between staff and family perspectives that led to the ECAF.

Part of this perspective gap is due to the distress that families experience when a loved one dies, no matter how good the care is. Some of my frustration about measuring quality of life derives from this tension. At the end of someone's life, I think we have to be realistic. We have to remember that at the end of the day, death is a traumatic experience for families. If we were to tell families, "You have the good luck to be part of our palliative care program, and we are just a wonderful palliative care program. We are so wonderful, we are going to grant you one wish. What would your wish be?" Many people, if they honestly searched in their hearts, would say, "If you are that wonderful, my wish is that you make my relative not die." And that's what we have to remember when we try to show outcomes in quality of life.

Other thoughts on addressing quality of life in an integrated palliative care program?

To a large extent, it is confusing, complicated, and difficult. We need to be realistic about what it is that we are trying to do. The global dimensions—for example, access to palliative care and the right location for the right patient—are easier to measure. The micro, day-to-day assessments of quality of life for patients who are dying are shifting sands. We have to be rather humble about our ability to accurately measure and represent them.

Post a Comment

* The ESAS is available online, see Resources and Tools page. [Return to Featured Innovation]

References

1. Chochinov HM, Wilson KG, Enns M, Lander S. Are you depressed? Screening for depression in the terminally ill. American Journal of Psychiatry. 1997;154:674-76. [Return to Featured Innovation]

2. CAGE is a mnemonic to remember the key words-Cut down, Anger, Guilty, Eye-Opener-in asking patients the four questions relating to their past use of alcohol. [Return to Featured Innovation]