Background and objective: Chronic myeloid leukaemia (CML) affects approximately 3000-5000 Americans each year, with the American Cancer Society expecting 4600 new cases in 2004. The incidence of CML increases with age; median age at diagnosis is 67 years. Long-term data on the economic burden associated with CML among the elderly are sparse. To fill this void, our study uses population-based data to evaluate longer-term treatment patterns, outcomes and costs among elderly Medicare beneficiaries following their diagnosis of CML. Patients and methods: This retrospective cohort analysis used linked data from the Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute and Medicare claims. Study cohorts included 567 patients aged greater than or equal to65 years newly diagnosed with CML between 1991 and 1993 in a SEER registry and followed for 5 years or until death, whichever occurred first. In addition, 567 control patients without CML matched I : I by age and sex (average age 78 years, 52% male) were included. The costs of care were based on total Medicare payments (in constant 1998 dollars). Groups were generally similar in terms of nonmatched variables. Results: Five years following diagnosis, 13% of CML patients were still alive versus 68% of the control patients (median survival: 14 months vs >60 months, respectively). The average total Medicare payments were $US35 785 for CML patients versus $US21 161 for control subjects (monthly payments: $US1688 vs $ US428, respectively; p < 0.001). Approximately 25% of CML patients underwent Medicare-covered cancer treatment (11% outpatient chemotherapy, 6% inpatient chemotherapy and 8% outpatient interferon-a therapy). Younger patients and those diagnosed in the later years were most likely to be treated. Costs for treated patients were higher, and they lived longer. Conclusion: Our findings suggest low treatment rates, and substantial excess mortality and costs associated with CML among the elderly. The recent introduction of imatinib has dramatically changed the treatment of CML, which is likely to cause important changes to the economic burden of CML. Our results can be used as a baseline for evaluating the impact of such new therapies as data from clinical trials become available. Further work is needed to characterise this disease and the complex factors that influence treatment decisions and associated health outcomes in elderly patients.