End-of-life care was the subject of several studies that appeared in a recent issue of the Archives of Internal Medicine. Although the studies looked at end-of-life (EOL) care from different vantage points, a common theme emerged: The importance of ongoing communication and a stable relationship between doctors and patients is the basis of prudent EOL decisions.

“These reports help shed light on an important issue: the differences between certain populations in end-of-life care, including lower rates of hospice use among certain populations,” said Otis W. Brawley, MD, chief medical officer of the American Cancer Society. “But the issue here may not be about race as much as it is about economics and culture.”

For patients with cancer, their families, and their health care providers, end-of-life care presents numerous challenges. A study by Anthony Back, MD, and colleagues, was a longitudinal, qualitative study of patients, family caregivers, physicians, and nurses drawn from a community-based sample (Archives of Internal Medicine 2009;169:474–479). A total of 31 physicians were recruited to participate in the study, and they, in turn, recruited 55 patients with incurable cancer or obstructive pulmonary disease who were not expected to survive for another year, 36 family caregivers, and 25 nurses. All participants were interviewed at enrollment, 4 to 6 months later, and at 12 months (38% of patients died during the average 6 months of follow-up).

Before their deaths, patients worried about physician abandonment, and this was related to the loss of continuity between the physician and patient. Patients quoted in the published article expressed concern that when death approached, a relative stranger would take over their care. Patients feared this form of abandonment because they placed great value on the professional expertise of their primary physician and also because they felt the loss of their personal relationship and familiarity with specific doctors and nurses. To illustrate, a quote in the article described how a patient asked a nurse to come along to an unexpected hospital admission and to visit the person in the hospital. Family members were also displeased that the doctor was out of the picture once the patient entered hospice.

For their part, physicians were aware that some patients would equate cessation of chemotherapy, for example, with abandonment. As a result, some physicians engaged in reassurance and continuity strategies, such as having the patient continue to come in for office visits. Family members noted the importance of closure with the physician after the patient's death, sometimes with a telephone call or a short letter. Physicians noted several problems with attaining closure with each family, especially their time to make necessary telephone calls. One physician, who was quoted in the article, reported having 10 to 20 patients in hospice at any given time.

Back et al concluded that 2 distinct elements contribute to ameliorating the patient' sense of abandonment at the end of life: (1) continuity, both of the primary physician's expertise and the patient-physician relationship, and (2) closure of the therapeutic relationship.

A second study, by Baohui Zhang, MS, and colleagues, looked at health care costs in the last week of life as part of the Coping with Cancer longitudinal multi-institutional study sponsored by the National Institute of Mental Health and the National Cancer Institute (Archives of Internal Medicine 2009;169:480–488). Participants included 627 patients with advanced cancer and an estimated life expectancy of less than 6 months, who were interviewed at the start of the study and followed to death. Costs accrued in the last week of life associated with hospitalization, stays in the intensive care unit (ICU), and life-sustaining care such as mechanical ventilation and resuscitation were assessed.

Less than one-third (31.2%) of participants reported having a conversation about EOL issues with his or her physician. Compared with patients who did not engage in EOL discussions, those who did discuss these issues with their physician had lower functional status, more comorbid conditions, greater symptom burden, and were more likely to want to know their life expectancy, to acknowledge that their disease was terminal, and to report not wanting to spend time in the ICU. Because these differences were likely to influence clinical and economic outcomes, the researchers used propensity-score matching to compare 124 participants who reported EOL conversations with an equal number of participants who did not, and who were similar regarding clinical and demographic characteristics associated with EOL conversations. Propensity matching effectively removed the bias that could be associated with a patient's report of an EOL discussion. When costs associated with the last week of life were assessed, mean aggregate costs for patients who discussed EOL care with physicians were $1,876 (standard error of the mean [SE], $177) compared with $2,917 (SE, $285) in patients who did not have these conversations. The difference between the 2 groups was $1,041 or 35.7%. Lower costs were attributed to more limited use of ICU care and interventions such as mechanical ventilation. Aggressive procedures did not confer a longer survival time. In addition to higher costs, lack of EOL discussions was associated with a worse quality of death, according to their informal caregivers, typically a spouse.

Zhang et al cautioned against drawing conclusions about a causal relation between EOL discussions and cost differences during the last week of life, as this was an observational study. They estimate, however, that if the proportion of patients with advanced cancer in the United States having EOL discussions with their physicians was increased to 50%, then the yearly associated cost savings could be in the range of $76 million. They concluded that better communication between advanced cancer patients and their physicians could result in lower costs at the end of life, as well as a better quality of life—and death. Among the strategies they suggested for making progress on this front are physician reimbursement for EOL discussions, more physician education about EOL conversations, expanded palliative care programs, and collaborative management of advanced cancer patients by oncologists and palliative care specialists.

A third study in this issue, by Amresh Hanchate, PhD, and colleagues, examined racial and ethnic differences in end-of-life-care costs (Archives of Internal Medicine 2009;169:493–501). By using a random, stratified sample from a Medicare population of deceased persons, the investigators used regression analysis to examine race and ethnic differences in Medicare expenditures according to age, sex, cause of death, total morbidity, geography, life-prolonging interventions, and hospice use. Compared with whites, costs in the last 6 months of life were 32% higher in blacks ($26,704) and 57% more in Hispanics ($31,702). The analysis determined that between 55% and 68% of the differences in end-of-life expenditures between whites and others was due to differences in demographic factors, total morbidity, socioeconomic indicators, and, most of all, geography. Costs were higher in areas where medical costs overall were higher. Costs were lowest—and racial and ethnic differences were smallest—in several states in the southeastern United States.

Whites used hospice more often (26%) than blacks (20%) or Hispanics (23%). Use of the ICU and intensive procedures such as gastrotomies, resuscitation, cardiac conversion, and mechanical ventilation were significantly higher among black and Hispanic patients than among whites, which accounted for the balance of the difference in costs compared with whites. The investigators identified the greater use of intensive, life-sustaining interventions among blacks and Hispanics as an area for future research, especially because these interventions contrast so sharply with health care expenditures among minorities at other stages of life. Blacks and Hispanics, for example, receive less chemotherapy and cardiac catheterization than do whites.

Furthermore, the authors made the interesting observation that blacks receive lower quality primary care and make less use of primary preventive services than whites. If, as Zhang et al propose, on the basis of their findings, patients who have EOL conversations with their physicians have lower health care costs during the final week of life, lack of consistent contact with a primary care physician may drive up costs. In fact, Hanchate et al did find, in an earlier study, that patients in their study who had more visits with a primary care physician in the latter months of life had lower costs and used the hospital less frequently near the end of life (Journal of General Internal Medicine 2008;23:1330–1335). In addition, enhancing culturally competent communication about EOL treatment options may prove an effective way to ensure that minority cancer patients are aware of the incurable nature of their disease and are realistic about the risks and benefits of life-prolonging treatments. The result would be more informed and, therefore, better EOL decisions for all groups of cancer patients regardless of their status as belonging to either a minority or majority population.

“People who do not have adequate access to health care often do not have an ongoing relationship with a specific physician; they are more likely to use emergency rooms or other acute care centers. Without a regular source of care, there's less opportunity to build a trusting relationship. And trust, we are growing to learn, can have a profound impact on medical care,” commented Dr. Brawley. 1

thumbnail image

Illustration 1. Patients with terminal cancer who are near the end of life may fear physician abandonment when there is not ongoing communication between patient and physician. Credit: Silvia Jansen/

Download figure to PowerPoint