Identifying potential indicators of the quality of end-of-life cancer care from administrative data.

PURPOSE: To explore potential indicators of the quality of end-of-life services for cancer patients that could be monitored using existing administrative data. METHODS: Quality indicators were identified and assessed by literature review for proposed indicators, focus groups with cancer patients and family members to assess candidate indicators and generate new ideas, and an expert panel ranking the meaningfulness and importance of each potential indicator using a modified Delphi approach. RESULTS: There were three major concepts of poor quality of end-of-life cancer care that could be examined using currently-available administrative data (such as Medicare claims): institution of new anticancer therapies or continuation of ongoing treatments very near death; a high number of emergency room visits, inpatient hospital admissions, or intensive care unit days near the end of life; and a high proportion of patients never enrolled in hospice, only admitted in the last few days of life, or dying in an acute-care setting. Concepts such as access to psychosocial and other multidisciplinary services and pain and symptom control are important and may eventually be feasible, but they cannot currently be applied in most data systems. Indicators based on limiting the use of treatments with low probability of benefit or indicators based on economic efficiency were not acceptable to patients, family members, or physicians. CONCLUSION: Several promising claims-based quality indicators were identified that, if found to be valid and reliable within data systems, could be useful in identifying health-care systems in need of improving end-of-life services.