It seems easy to conceptualize quality measures for various purposes, but in practice, we are at an early stage in measuring the quality of end-of-life care. Substantial normative and empirical research is still needed to develop and validate a conceptual model of quality end-of-life care, to develop and test the psychometric properties of proposed measurement tools, and to demonstrate the tools’ effectiveness in multisite studies before they can be used nationally. This chapter describes what is known and what still needs to be done to develop widely applicable quality indicators for end-of-life care for cancer patients. The following questions guide the discussion:

1. What is currently known about the quality of care for dying cancer patients?

2. What are the proposed definitions and conceptual models for quality of care of dying persons and their family?

3. What can we currently measure with existing nationally collected data?

4. What do we want to be able to say in the future?

5. What research is needed?

Few studies have characterized the experience for dying cancer patients and their families. From a national perspective, only one ongoing data collection effort routinely characterizes dying from cancer, and a second occasional survey was carried out six times between the early 1960s and 1993. The National Center for Health Statistics (NCHS) compiles data from all death certificates nationwide and publishes annual summaries that include cause of death, place of death, and other demographic information.

The other survey that has characterized aspects of dying is the NMFBS, last carried out in 1993. The 1993 survey represents a 1 percent sample of all deaths at age 15 and older. Unlike the mortality followback surveys in the United Kingdom (Addington Hall and McCarthy, 1995a, 1995b), the NMFBS does not characterize the quality of the dying experience (e.g., pain management, satisfaction). Rather, the U.S. survey collects information on socio-demographics, use of alcohol and medications, lifestyle, health care resource utilization, and difficulties with functioning (Lentzner et al., 1992; NCHS, 1998).

What can be learned from the available data? In 1998, 538,947 people died of cancer in the United States. Five types of cancer account for 70 percent of those deaths (Figure 3-1; NCHS Web site). Over the past decade, there has been a trend toward more cancer patients dying at home (Figure 3-2;