Quality end-of-life care: where do we go from here?

THE RIGOROUS STUDY by Emanuel and coworkers reported in this issue of the Journal underlines two critical points in relation to end-oflife care. First, the various "pieces of the puzzle" at the end of life—for example, pain control, advance care planning, and palliative care—should be integrated into a coherent conceptual framework of quality end-of-life care. Second, the appropriate evidence base for this framework is derived from patients' perspectives. After exploring these two points, we make two further observations about next steps in improving quality endof-life care: we need to develop a quality improvement strategy at the organizational and health system level, and we need to focus on quality end-of-life care not only in developed countries but also throughout the world. The first point has now been made by at least five groups—a consortium of organizations that published a position paper in Journal of the American Geriatric Society, the Institute of Medicine Committee on End-of-life Care, our research group, Steinhauser et al., and Emanuel and Emanuel both in this article and in an earlier publication. The underlying assumption is that to make gains in quality of end-of-life care, "lumping" (addressing the problem of end-of-life care as an integrated whole) is better than "splitting" (addressing the various "pieces of the puzzle" in isolation). The extent to which this assumption proves true remains to be seen. However, it is obvious that the quality gains related to "splitting," which is the predominant research approach in end-of-life care, has led to disappointing results in the quest to improve end-of-life care. The "nail in the coffin" of the "splitting" strategy was probably the SUPPORT trial. The second point has been put forth by three of these groups: our research group, Steinhauser and colleagues, and Emanuel and colleagues. We conducted in-depth qualitative interviews of patients receiving dialysis, people with human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS), and residents of a long-term care facility to identify five elements of quality end-of-life care: adequate pain and symptom control, avoiding inappropriate use of lifesustaining treatments, achieving a sense of control, relieving burdens on loved ones, and strengthening relationships with loved ones. Steinhauser et al. used qualitative focus groups and interviews with providers, bereaved family members, and patients with cancer and HIV to identify six elements of quality end-of-life care: pain and symptom management, clear decision making, preparation for death, completion, contributing to others , and affirmation of the whole person. Emanuel and Emanuel used factor analysis of survey data from terminally ill patients, the majority with cancer, to identify eight elements of quality end-of-life care: patient-clinician relationship, social connectedness, caregiving needs, psychological distress, spirituality/religiousness, personal acceptance, sense of purpose, and clinician communication. Significantly, the survey questions in the Emanuel and Emanuel study also arose initially in focus groups with patients and providers. Because the frameworks are based on