Chemotherapy use among Medicare beneficiaries at the end of life

Background: Although many observers believe that cancer chemotherapy is overused at the end of life, there are no published data on this. Objective: To determine the frequency and duration of chemotherapy use in the last 6 months of life stratified by type of cancer, age, and sex. Design: Retrospective cohort analysis. Setting: Administrative databases from Massachusetts and California. Patients: All Medicare patients who died of cancer in Massachusetts and 5% of Medicare cancer decedents in California in 1996. Measurements: Use of intravenous chemotherapy agents, chemotherapy administration, or medical evaluation for chemotherapy from Medicare billing data for each patient in 30-day periods from the date of death backward. Results: In Massachusetts, 33% of cancer decedents older than 65 years of age received chemotherapy in the last 6 months of life, 23% in the last 3 months, and 9% in the last month. In California, the percentages were 26%, 20%, and 9%, respectively. Chemotherapy use greatly declined with age. Chemotherapy use was similar for patients with breast, colon, and ovarian cancer and those with cancer generally considered unresponsive to chemotherapy, such as pancreatic, hepatocellular, or renal-cell cancer or melanoma. Patients with types of cancer that are unresponsive to chemotherapy had shorter duration of chemotherapy use. Conclusion: Among patients who died of cancer, chemotherapy was used frequently in the last 3 months of life. The cancer's responsiveness to chemotherapy does not seem to influence whether dying patients receive chemotherapy at the end of life.