involves four basic steps: (1) planning a change, that is, developing specific steps (e.g., data collection strategy, selection of study population) and time-table to reach the objectives; (2) doing or implementing the planned change and collecting data on outcomes and processes, including unexpected problems and results; (3) studying the change by analyzing the data, comparing results to expectations, and summarizing lessons; and (4) acting as guided by the analysis, including revising the original plan, expanding it, or identifying a new strategy. Although investments in these strategies have not necessarily been documented by controlled studies to be as successful as promised (see, e.g., Shortell et al., 1998), they have a recognized place in efforts to improve the performance of the American health care system (see, e.g., IOM, 2001b) including the delivery of consistent, effective, and compassionate palliative, end-of-life, and bereavement care (Lynn et al., 2000).
A wealth of ideas for systematic quality improvement in end-of-life care have been generated by an initiative cosponsored by the Institute for Healthcare Improvement and the Center to Improve Care of the Dying. Leaders of the initiative subsequently discussed these ideas in Improving Care for the End of Life: A Sourcebook for Health Care Managers and Clinicians (Lynn et al., 2000; Web site at http://www.medicaring.org/educate/navigate/qi.html).
One chapter of the sourcebook is devoted to reports on strategies tried by a number of health care organizations to reduce fragmentation of care, improve continuity and accountability, and increase satisfaction with care. Specific, measurable objectives for these projects included reducing patient transfers (especially when death is near), providing patients and families with one or two central contact people, coordinating hospice and oncology units in a managed care system to provide a continuum of care and promote consistency in pain management, establishing reliable procedures for professionals to communicate with each other about a patient’s care, and establishing and delivering on specific promises (e.g., that a known person will respond to a request for help within 30 minutes, day or night).
Some of the programs described in the sourcebook sought to coordinate care across multiple sites and providers following the model Program of All-Inclusive Care for the Elderly (PACE) that was pioneered by On-Lok Senior Health Services in California. (PACE also provided the model for the Medicaid-based pediatric palliative care demonstration projects described earlier in this report.) Other projects were focused on narrower issues, for example, developing reliable, easy ways for advance care plans to be available, recognized, and honored by providers in all settings and circum-