Over the past several decades, numerous clinical trials have consistently demonstrated that chemotherapy is efficacious in both premenopausal and postmenopausal women with breast cancer. However, little is known about the use of chemotherapy and its effectiveness in the community because of a lack of large population-based data. While the Surveillance, Epidemiology and End Results (SEER) tumor registries provide data on radiation therapy and surgical treatment for breast cancer, they do not provide information on chemotherapy. They are not required to collect this information, and the medical records in oncologists' offices, where chemotherapy is likely to be administered, may not be routinely reviewed for data on the primary course of cancer-directed therapy. Medicare claims are a potential source of national data on chemotherapy use for breast cancer cases aged 65 and older. However, no study has examined the validity of this information. If Medicare claims provide accurate and complete data on chemotherapy, their utility for breast cancer research would be considerably enhanced, allowing for population-based analyses of the current uses of chemotherapy as well as effectiveness studies in the community. Therefore, we propose to examine the utility of information on chemotherapy from Medicare claims data for women aged 65 and older who have been diagnosed with breast cancer. Our approach is to identify women aged 65 years or older who have been diagnosed with breast cancer in the New Mexico Tumor Registry (a SEER registry), that can be linked with Medicare claims data by unique identifiers. We will contract with the registry to abstract the medical records on chemotherapy administration for eligible patients in all facilities including medical oncologists' offices where patients may have received chemotherapy. We will then compare information on the use of chemotherapy from medical chart review with information obtained from Medicare claims data.
Our aims are 1) to determine whether Medicare data can identify women who received chemotherapy for breast cancer; 2) to determine whether Medicare data can differentiate among the specific regimens of chemotherapy; and 3) to determine whether the number of claims can be used to estimate the number of cycles of chemotherapy. Based on this information, we will characterize the nature and extent of error in using Medicare claims for chemotherapy effectiveness research among older women diagnosed with breast cancer. We will then use the Medicare-SEER linked data for all SEER areas to examine patterns and outcomes of chemotherapy among older women with breast cancer. Information generated from the medical chart reviews will also allow us to examine patterns of chemotherapy care independent of the Medicare claims data or independent of those parts of the claims that we find generate invalid estimates.
Showing the most recent 10 out of 12 publications
