What do we know? So what?

Most patients are dying in institutions

There is concern that inpatient facilities will not be able to cope with the large increase in deaths in the years to come. Despite documented preferences for home death (25-27), the majority of deaths from terminal illness still occur in hospital. This happens mostly when there is no guarantee of 24-hour support at home or back-up from specialized personnel (28).

In the United Kingdom, often considered the birthplace of modern hospice and palliative care, the percentage of home deaths fell from 31.1% in 1974 to 18.1% in 2003. If the trend continues, it is anticipated that less than one in 10 (9.6%) will die at home in 2030. Such a decline in home deaths would correspond to an increase in institutional deaths of 20.3% (29). In the United States, of the more than 1.4 million deaths in 1997 available for complete analysis, 52.8% occurred in hospital, 23.6% occurred in nursing homes and only 23.6% occurred at home (30).

Further analysis suggests that the opportunity for home death is disproportionately high among Caucasian people, and those with higher socioeconomic status (31, 32), who are married (27, 29, 33-34), who are suffering from cancer (27, 30) and living in a racially homogeneous area (35) in both the United States and other Western industrialized nations.

Observational studies have also shown that expressed preferences (36), poor functional status, intensive home care support, living with relatives and extended family support are associated with more frequent deaths at home (37).

Meanwhile, a lack of 24-hour home care services and poor coordination between health and social care services have been blamed for an increase in hospital deaths in spite of patients' preferences for death at home (38-39).

Care at the end of life is improving slowly

Over the past 25 years, the field of palliative medicine has developed in response to the needs of dying patients and their families such that in many countries it is recognized as a specialty or a sub-specialty.

In the management of non-cancer chronic diseases near the end of life, there is clear evidence that the management of symptoms often has to go hand in hand with the continued management of the underlying illness.

Having demonstrated its effectiveness in improving assessment, documentation and care for patients in the last 24 hours to the final days, the Liverpool Care Pathway is now increasingly adopted in many parts of the world (40, 41). There is also good evidence that advance care planning led by skilled facilitators who engage key decision-makers directly over multiple sessions leads to increased utilization of advance directives (42, 43) and a better death experience at the end of life (44, 45).

It is increasingly recognized that in the management of patients with heart failure, there are reduced re-admissions and improved continuity of care with multi-component interventions (46). Although effect sizes are small, there is weak to moderate evidence suggesting that comprehensive and individually targeted interventions can relieve caregiver burden and improve satisfaction mainly in patients with dementia (47, 48). These interventions involve multi-disciplinary collaboration, address needs across care settings and over time, and facilitate communication by personal and technological means.

Despite pockets of excellence and the growing knowledge base, widespread adoption of the principles of palliative care and the dissemination of such knowledge is still lacking (49, 51). Only in recent years have major organ- or disease-specific textbooks paid much attention to supportive and palliative care of end organ failure (52, 53). Palliative care education is still not an essential component of many medical and nursing schools (54, 55). When offered, teaching tends to be fragmented, ad hoc and lacking in coordination. Most teaching is hospital-based and little attention is given to home care, hospice and nursing home care. As a result, many health professionals still find themselves standing by helplessly as patients suffer and families fret.

Besides textbook revision and improvements in health care curricula, the way forward should include the development of palliative care leaders and faculty, creating standards of care and certification, and promoting clinical programs as venues for education and enhanced educational resources for end-of-life care (56, 57, 58).

People die differently

Knowledge and skills alone are inadequate in the provision of good palliative and supportive care.

In their 1965 book, Awareness of Dying, Glaser and Strauss first described the different types of interaction that occur between the dying and those around them (59). They described the contexts of interaction based on the different degree of awareness of the dying phase: closed awareness, suspected awareness, mutual pretense awareness and open awareness. The impact of each type of awareness context upon the interplay between patients and personnel is profound, for people guide their discourse and actions according to who knows what and with what certainty.

The authors subsequently went on to describe the various patterns of dying in a book, Time for Dying (60), which provided the beginnings of our understanding of the different trajectories of dying. These patterns of (a) sudden and abrupt death, (b) gradual decline followed by period of more rapid decline, such as seen in cancer, (c) the entry-reentry deaths against a background of steady decline of many chronic illnesses and (d) the progressive frailty followed by death, were subsequently illustrated in a study of Medicare beneficiaries (61) and by Lunney et al (62) in their cohort study of four US regions. One of the challenges faced by those interested in supportive and palliative care for people living with multiple chronic diseases is that the majority of hospice and palliative care programs were designed to support a cancer trajectory. In cancer, there is usually
a period of overall slow decline until anti-cancer treatments are stopped, followed by a relatively rapid decline in function towards the end of life. These expected deaths carerlikely to have a fairly predictable terminal phase, where there is time to anticipate palliative needs and plan for end-of-life care. It may also largely match public expectation of dying. However, this does not necessarily serve the needs of those dying with other trajectories.

With the recognition of different terminal trajectories, Joanne Lynn, in a Hastings Centre Report (63), raised the notion of mass customization, in order to meet the needs of the terminally ill. Mass customization aims to define manageable populations with similar needs to then engineer services that match the size of the population and its predictable needs.

It is estimated that about 20% of Americans will die following a course of gradual decline followed by a more rapid period of deterioration. This course, followed by most major cancers, requires excellent medical care during the long period of good function, followed by supportive and palliative care for patient and family during the period of rapid decline.

Other conditions, such as chronic heart failure and chronic obstructive lung disease, tend to follow a course of slow decline punctuated by serious exacerbations, with death occurring rather suddenly (the entry-reentry trajectory). It is estimated that about 25% of Americans follow this course. Those living with this trajectory usually benefit from (a) chronic disease management to reduce the likelihood of exacerbations and to sustain all possible function, (b) rapid intervention at the first sign of exacerbation, preferably in the home rather than the hospital and (c) good advance care planning to direct care in the event of overwhelming exacerbations.

Approximately 40% of Americans are estimated to follow the trajectory of long-term dwindling of function with death following physiological challenges such as those triggered by influenza, urinary tract infection, pneumonia or a broken hip. Half of these patients lose cognitive function. Those following this trajectory tend to benefit from supportive care over the years, requiring assistance with everyday activities and long-term interventions to promote optimal levels of comfort for patients and family caregivers.

One of the main drawbacks of this approach is its foundation on the assumption that the vast majority of patients will have the right things done for them at the right time because it is built into the system and part of the expected pattern, while downplaying the fact that good care could arise from prudent choices by individual patients, and their health professionals and caregivers.

Dying is a multidimensional experience

Dying is not just a physical demise, and health professionals must strive to identify and meet the multidimensional needs of people with progressive disease. Centuries ago, spiritual care dominated end of life care. Although palliative care set out 40 years ago to address the suffering of total pain, including lack of personal integrity and inner peace, spiritual distress at the end of life has remained relatively unexplored (64), although it is accepted that quality of life is modified by all dimensions of personhood (65).

Defining and assessing spiritual needs, however, is problematic. A useful definition is that proposed by the US Institute of Medicine, which states: spiritual needs are the needs and expectations that human beings have to find meaning and purpose in life; such needs may be specifically religious but even people who have no religious faith or who are not members of an organized religion have belief systems relating to meaning and purpose (66).

Spiritual issues are frequently very significant for people living and dying with lung cancer and heart failure (67). A secondary analysis of in-depth serial interviews suggests that there might be typical patterns of social, psychological and spiritual needs towards the end of life (68-70). In lung cancer, the social trajectory mirrored physical decline while the psychological and spiritual wellbeing decreased together at four key transitions: at diagnosis, after getting home after initial treatment, during disease progression, and in the terminal stage. In advanced heart failure, social and psychological decline both tended to track the physical decline while spiritual distress exhibited background fluctuations.

Knowledge about these patterns can improve the ability of health professionals to anticipate and share with patients when they are likely to be distressed. Explanations for patients and their carers about when practical, emotional and existential issues might be expected to occur, and the services available, can empower them and their carers, and this can be very reassuring for all.

This holistic view, considering each dimension of need, may lessen the multi-specialist approach, and moderate the current technological imperative with care focused on interventions to prolong life, with sometimes overzealous and futile treatment. Considering these different trajectories would bring spiritual assessment and care into focus, highlighting that many patients have spiritual issues from diagnosis of cancer or chronic life threatening illness, not just at the very end of life.

The implication of this is that spiritual support should be available for patients from diagnosis, sooner rather than later. A patient-centered approach that supports people in their own worldview while allowing for expression of fear, doubt and anxiety may help patients in their search for meaning and purpose, and prevent spiritual concerns amounti g to disabling spiritual distress. Some questions that could be asked of people living with multiple chronic diseases, and of their loved ones, in order to facilitate such an approach are (71):

- What is the most important issue in your life right now?

- What helps you keep going?

- How do you see the future?

- What is your greatest worry or concern?

- Are there ever times when you feel down?

- If things got worse, where would you like to be cared for?

Allowing patients to raise spiritual and religious issues may be therapeutic, as may the use of a gentle prompt, such as: You seem fine today, but do you ever feel down or a bit low? This may allow them to reveal their personally felt narrative, rather than the public account they may tend to offer, as patients often have competing narratives in their minds. Patients may sometimes ask us about our own beliefs. In such cases, it might be useful to acknowledge the question, reflecting it back to the patients to ask them about their beliefs. This is because they may just be looking for an opportunity to express their own feelings and needs.

Supportive and palliative care save money

There is evidence that enrollment into hospice and palliative care services saves money. In the US there has been an increase in Medicare dependents choosing hospice benefits, from 27% in 2000 to 40% in 2005. An independent study from Duke University in 2007 showed that hospices provide compassionate care for those reaching the end of life and save Medicare an average of $2,300 per patient, amounting to savings of more than $2 billion last year (72).

A 2008 paper also reported that hospice enrollment results in substantial savings in government expenditures (22 percent) among all short-stay (< or = 90 days) dying residents of nursing homes. For long-stay (> 90 days) dying residents, hospice care led to some savings (8%) among cancer residents while it was cost-neutral among dementia residents, while adding some cost (10%) for residents with a diagnosis other than cancer or dementia (73). In hospitals, matched patients who received palliative care resulted in highly significant cost savings for the hospital compared with those who did not (74). The savings were primarily through reduced hospital stay, an increase in the death-at-home option, and a lower use of hospital emergency rooms by complex cases. In Spain, palliative care led to significant cost savings with greater efficiency and no compromise of patient care (75).