Key points are not available for this paper at this time.
PURPOSE: To determine the incidence and complications of myelodysplastic syndromes (MDS) among Medicare beneficiaries. METHODS: Retrospective review of 2003 Medicare Standard Analytic Files utilizing International Classification of Diseases for Oncology ninth edition CM code 238. 7 to identify new MDS patients, with 3-year follow-up. RESULTS: Among 1, 394, 343 individuals in Medicare Standard Analytic Files age > or = 65 years, 162 per 100, 000 were coded as newly diagnosed MDS during 2003 yielding a calculated 45, 000 new cases in the United States Medicare > or = 65 years population. Patients with MDS were older (median age, 77 years), and over-represented by males. Among patients with MDS diagnosed during first quarter of 2003, 73. 2% suffered cardiac-related events during 3-year follow-up, which exceeded the Medicare population (54. 5%; P <. 01) even when age adjusted (odds ratio, 2. 10; P <. 01). Significant increases in prevalence of diabetes (40. 0% v 33. 1%), dyspnea (49. 4% v 28. 5%), hepatic diseases (0. 8% v 0. 2%), and infections (sepsis: 22. 5% v 6. 1%) were noted in MDS (all P <. 01) compared with the Medicare population. Patients with MDS requiring RBC transfusions had greater prevalence of these comorbidities. Acute myeloid leukemia developed within 3 years in 9. 6%, with increased transformation among transfused (24. 6%; P <. 001). The 3-year Kaplan-Meier age-adjusted survival for MDS was 60. 0%, which was significantly lower than the Medicare population (84. 7%; hazard ratio, 3. 08; P <. 001), and mortality was further increased among transfused MDS (P <. 01). In 2003, median payment for MDS was 16, 181, compared to 1, 575 for the Medicare population (P <. 001). CONCLUSION: MDS is a common hematologic malignancy of the elderly, which places patients at risk for comorbid conditions. Transfusion dependency identifies patients with MDS at additional increased risk of organ impairment and shortened survival.
Goldberg et al. (Tue,) studied this question.