for the issues surrounding death are but one subset of the general problem of how to take account of culture in the health care system. Next, I will review the key issue of how we talk about and understand cultural difference. What are these categories into which we divide our pluralistic society? I will next address the way in which existing bioethics practices surrounding death have failed to consider cultural difference. The latter topic is the focus of my current research. Using the methodological strategies of medical anthropology, I am studying how California patients from different cultural backgrounds make end-of-life decisions.1 Implications of cultural diversity on individual decisionmaking and resource allocation will be discussed very briefly. The paper will conclude with recommendations for the IOM panel regarding research, clinical practice, and teaching about cultural diversity and end-of-life decisionmaking.
Although beyond the scope of this brief paper, it is vital to remember that culture is not simply an inconvenient barrier to a rational, scientifically based health care system or a feature of ethnic "others." Deeply embedded cultural values are apparent in the way American medicine has approached the care of the dying, particularly practices that have separated terminal care from mainstream practice, denied the existence of the dying patient, and assumed that death was simply one of many medical problems open to a technological solution (see Callahan, 1993; Muller and Koenig, 1988). Professionals can be usefully thought of as operating within a culture influenced by widespread cultural assumptions and practices as well as by training and experience. There also appears to be a clear line dividing the medical model from the patient's model of death and illness leading to death (Churchill, 1979). And differences in professional culture between nurses and physicians may lead to varying understandings of patients' prognoses, as Anspach has demonstrated for the neonatal intensive care unit (1987). Western European health professionals respond with astonishment when told about the American practice of requiring a "Do Not Resuscitate" order for an elderly person dying in a long-term care facility. Cultural "points of view," based on specific contexts of meaning, do not apply only to patients and families from backgrounds unfamiliar to their health care providers.
In an increasingly plural society, cultural diversity among American health care workers is another important consideration. It would be inappropriate to structure the debate as if providers uniformly express white middle-class culture in opposition to the "ethnic otherness" of patients. Particularly in large urban areas, the diversity of professionals and staff workers in hospitals, nursing homes and long-term care facilities is adding a major area of complexity to decisionmaking about tube feeding, treatment of acute illness, and withdrawal of therapeutic interventions (CAHA, 1993). Many health workers providing direct patient care do not share the