Sick To Death > Chapter 2 > Trajectories of Illness Across Time
How could health-care planning, financing, and policy identify and target the vulnerable population of people who have serious, progressive, chronic conditions that cause disability and suffering and will end in death? Specifically, for patients who are sick enough that death would not be surprising, would health-care services tailored to their particular needs improve their lives? If so, how are we to define the group and fit appropriate services to it? And, finally, does the process of tailoring services to patients' needs show us useful, informative divisions among those who are sick enough to die?
To answer these questions, it is helpful first to understand how to split the whole population into groups, according to what people in those groups generally need from the health-care system across time (Lynn 2001). The conventional divisions are by disease and setting of care. But these do not work well for a population that often has multiple diseases and needs multiple settings of care. Initially categorizing the entire population into the following three groups, as shown in the pie chart in figure 5, serves as a useful framework for organizing health care (Lunney, Lynn, and Hogan 2002; Lunney et al. 2003):
Priorities for generally healthy people and people with stable or mild chronic conditions are straightforward: to stay healthy, prevent disease, or restore good health. In these stages of life, people need mostly prevention and acute-care services (routine visits to health-care providers for prevention and health maintenance, and emergency medical services).
The third group, however - patients who live with serious chronic disease that will worsen and cause death - has much more complicated needs and priorities. People in this part of their lives still want to live as well as possible and usually for as long as possible. They and their families have many goals: relief from symptoms; help with family burdens; reasonable costs; ongoing good quality of life; a sense of being in control and maintaining dignity; and an opportunity to come to peace with spiritual issues and relationships. These goals coexist with those of preventing illness and prolonging life, even when pursuing such goals requires aggressive or invasive treatments. Holding all these goals simultaneously contrasts with the usual thinking that people near death need comfort and life closure, but not treatment. Instead of a "transition from cure to care," people living with fatal chronic illnesses have multiple complex goals and priorities that evolve over time.
Tailoring services to match the time course - or trajectory - of needs would be overwhelming if each disease and care setting required its own protocol. Conveniently, three patterns, illustrated in the three graphs in figure 5, appear to encompass most of the ways people traverse their time of living with chronic, progressive, eventually fatal conditions (Lunney, Lynn, and Hogan 2002).
Most patients with malignancies maintain comfort and functioning for a substantial period, even for several years following evidence of an eventually fatal condition. However, once the illness becomes overwhelming, their status usually declines quite rapidly in the weeks and days immediately preceding death (Morris et al. 1986). Because of this rather predictable course, such patients often have a period when their prognosis is reliably shorter than six months and thus they can qualify for hospice enrollment. Turning down "curative" treatment, as Medicare's hospice benefit requires, is largely symbolic; usually cure would never have been possible. However, most patients facing progressive cancer want to slow the cancer's progress and often have good functioning until very late in the course.
Many cancer patients become seriously disabled only with overwhelming disease, when the patient has only a few weeks to live. Having hospice services available to support patients and their caregivers during that concentrated period works reasonably well (Institute of Medicine 2001). About half of all cancer patients in the Medicare program now take advantage of hospice services (U.S. General Accounting Office 2000); these individuals represent about two-thirds of all the Medicare patients who use hospice (Gage and Dao 2000). Some cancers don't follow this trajectory because they regularly cause long-term disability rather than a short terminal decline (prostate cancer in elderly men, for example), and some other conditions (such as severe strokes, and some AIDS cases) follow this trajectory along with cancers.
The second group of dying patients includes those with organ system failure, such as congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), cirrhosis, or kidney failure. They often live for a relatively long time following diagnosis and have long periods of staying in balance with the remaining organ system function. During those times, patients have few symptoms. But eventually some physiological stress overwhelms the body's reserves, leading to an exacerbation of serious symptoms.
Patients with organ system failure face an uncertain prognosis (Fox et al. 1999); most die suddenly from an unpredictable complication in the course of their chronic disease. The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT) found that most of these patients were stable and reasonably comfortable within a week of dying. The median heart failure patient still had a 50–50 chance to live six months, whether based on attending physicians' estimates or on a statistical model, on the day that turned out to be the patient's last full day alive (Lynn, Harrell, et al. 1997). In liver failure with cirrhosis, the median prognosis to survive six more months is approximately 4 in 10 within just two weeks before death (Roth et al. 2000). Death from emphysema is similarly difficult to predict, at least until the last few days (Claessens et al. 2000). The uncertainty of timing makes it appropriate to treat exacerbations as long as the patient is willing, even though treatment may fail and death follow. So the last phase of life for patients with this trajectory is much longer and indistinct, beginning at the point when the patient is first regularly limited by illness and continuing throughout recurrent exacerbations until death.
Close ongoing disease management has regularly proven to be effective in reducing exacerbations and improving survival span for patients with organ system failure (Rich et al. 1995; Bodenheimer et al. 2002). Many exacerbations arise from the patients' forgetting guidance about diet or medication, having insufficient funds to buy medications, or not understanding how to detect and treat early signs of an exacerbation. Programs that seek to prevent exacerbations and enhance quality of life generally emphasize self-care education, close symptom monitoring, around- the-clock on-call treatment of early symptoms, and consistent use of medications. Using quality-improvement methods, clinical teams have shown that adding advance care planning, mobilizing routine and urgent services to the home, and having deathbed care available at home yield further improvement in reducing exacerbations and costs and in tailoring services to patient and family preferences (Lynn, Schall, et al. 2000).
Because Medicare payments do not directly cover these services - though Medicare does cover hospitalization and emergency treatment - few people can get optimal care for serious organ system failure (Lynn, Wilkinson, and Etheredge 2001). Most patients following this trajectory are not homebound and so do not qualify for Medicare benefits for home care. In addition, these patients' prognoses are usually too uncertain for them to enroll in hospice programs. An individual physician or physician group could provide some of these services and be paid for office or home visits, but few physician practices have enough volume to support around-the-clock availability of a skilled person who knows the patient, has the patient's record, and could go to the home for timely evaluation and treatment. To justify the costs and ensure the skills, such a nursing service probably must enroll one hundred or more patients with similar organ system failure; thus implementation also requires limiting the number of provider organizations serving any one area. The United States does not have an established way to encourage concentrating services in a small number of providers in one area.
Those who escape cancer and organ system failure are likely to die in late old age with generalized frailty of all body systems or neurological failure (such as Alzheimer's and other dementia). Alzheimer's dementia will likely affect fourteen million individuals in 2050 (Alzheimer's Association 2001). The final course for these patients is more appropriately characterized as a slow dwindling away, rather than the terminal course of many cancers or the intermittent exacerbations of organ system failures.
These patients are likely to lose the ability to take care of themselves long before death. As a result, they ordinarily require intensive personal care throughout their period of dependency, which imposes substantial burdens on both paid and volunteer (usually family) caregivers. These patients usually experience a slow worsening of self-care, with occasional episodes of more serious infections, injuries, or other illnesses. Some patients may get a substantial illness that takes life abruptly, especially when all concerned have decided not to pursue aggressive treatment to prolong life. Others continue to lose ground slowly and eventually die. Medical personnel do not generally classify someone as "dying" from dementia, Parkinson's disease, or just the multiple limitations of advanced old age. These patients tend to have uncertain prognoses and are therefore often not eligible for hospice under current coverage mechanisms (Hanrahan et al. 1999). Until the very last stages of their illness, they are not home- bound and, consequently, do not qualify for Medicare benefits for home care. In the final stages of the disease, patients usually require very little skilled nursing, although they do require significant personal care, which Medicare or other health insurance does not cover.
Although these patients are difficult to fit into hospice or home-care nursing, an array of services has grown up to support the frail elderly and people with dementia. Care coordination services, Medicaid coverage of in-home and nursing-home personal care, Meals on Wheels, and other social service agencies fill some of the gap that would otherwise be left between conventional hospital and post-acute care on the one hand, and the real needs of these patients and their families on the other.