Diabetes in elderly patients with breast cancer in the United States: An analysis of data from the Surveillance, Epidemiology, and End Results (SEER)–Medicare linked database

Abstract

11050 Background: Despite increases in both breast cancer and diabetes, little is known about their co-occurrence. With evidence that cancer and diabetes can reflect related processes such as obesity, understanding the prevalence of diabetes in patients with breast cancer is critical. Methods: Using the SEER-Medicare linked database, 52,977 women newly diagnosed with breast cancer at ages 66+ years during 1998–2002 were identified, and a random sample of Medicare beneficiaries without cancer, matched (1:1) on sex and residence county were selected as a comparison cohort. The occurrence of diabetes was defined as ≥1 hospital or ≥2 physician service claims, and the date of cancer diagnosis served as the index date for the breast cancer patient and her matched cancer-free individual. For both cohorts, diabetes prevalence was measured for the ≥12 months prior to the index date, and incidence was estimated for non-diabetic patients during three time periods after the index date (3 months, 12 months, and overall), with follow-up through 2005. Prevalence and incidence rates (per 1,000 person-years) and their 95% confidence intervals (CI) are standardized to the age and race distribution of the breast cancer group. Results: Rates were higher in breast cancer patients for all time periods, particularly shortly after diagnosis. This pattern of heightened risk immediately following cancer diagnosis held when rates were stratified by age, stage, and race/ethnicity.For both cohorts, diabetes incidence was typically two times greater in Blacks and Hispanics compared to Whites. Conclusions: Older women with breast cancer are at increased risk for diabetes, suggesting that there may be shared biologic pathways such as insulin, INSR, IGF-1, and IGF-1R. [Table: see text] [Table: see text]