"The high cost of dying" revisited.

This review of the literature of the past decade on medical care costs at the end of life finds that the data do not support the often-voiced hypothesis that the rise in medical care costs is due largely to the disproportionate use of high-technology medical care by persons who die. It also shows that although the intensity of care, as indicated by hospital expenditures, declines with age, any savings on hospital costs of very old decedents are offset by nursing-home costs. Studies of hospice care and advance directives are reviewed for their effectiveness in reducing end-of-life costs, but these strategies are not promising at this time, largely because of the difficulty of predicting when an individual patient will die. It is suggested that curbing the rise in medical care costs will require basic changes in the physician-patient relationship and in our attitude to death.