trials is almost always many years under 65. For example, the median age of women with metastatic breast cancer participating in two randomized trials that evaluated pamidronate in preventing bone complications was 57 years (Hillner et al., 2000). This is in marked contrast to the 1988–1992 age-adjusted incidence rates of 72.8 per 100,000 for women under 65 and 445.4 per 100,000 for women over 65 (Kosary et al., 2000). The data available for policy are built on a population that is at least a decade younger than the population actually facing these illnesses. This is important because the financial and family resources, physiological reserves, and comorbidities of younger persons are dramatically different than they are in older persons. The burgeoning size of our over-65 population demands that research deal with this target population, especially when the data are used for Medicare policy.
Other considerations make the generalizability of published studies quite problematic. Most studies are carried out in academic centers, which serve populations quite different from random samples. Most require patients who are able to travel and to consent, which eliminates many with cognitive disability or severe poverty. Most also require that the patient have no other serious or life-threatening disease.
As an alternative to aggressive, often futile, expensive therapy requiring repeated hospitalizations, increasing numbers of terminally ill cancer patients are enrolling in hospice or other home care. By opting for hospice or extensive home care, patients commonly remain at home among loved ones, receiving care that is focused on their symptoms, emotions, spiritual concerns, and family.
However, the long-held assumption that hospice care can contain costs at the end of life is being challenged on a number of fronts. These supposed benefits were among the reasons for hospice becoming a benefit in the Medicare program in 1983. Early research appeared to confirm this (Brooks, 1989a, 1989b; Brooks and Smyth-Staruch, 1984; Kidder, 1992; Mor and Masterson-Allen, 1990). Kidder estimated that during the first three years of the hospice benefit program, Medicare saved $1.26 for every dollar spent on Part A expenditures. Mor and colleagues examined data from the National Hospice Study to assess the time in the hospice program that would be the most informative period for which to evaluate associated costs. Enrollment periods of one to three months tend to yield the most savings (Brooks and Smyth-Staruch, 1984; Kidder, 1992). However, when the time period extended six months and beyond, the savings were not substantial. The National Hospice Organization commissioned a study that found that hospice saved $1.52 for every $1 spent by Medicare (Lewin-VH1, 1995).