Cancer Control Research5R03CA139567-02
SCREENING MAMMOGRAPHY IN ELDERLY WOMEN USING LINKED BCSC - MEDICARE DATA
DESCRIPTION (provided by applicant): The incidence of breast cancer is highest (464.8 per 100,000 women) among women 75 to79 years of age. Despite efforts to promote screening and early diagnosis, approximately 48% of elderly women have metastatic disease and 9% have an unknown stage at presentation. The effectiveness of screening mammography in decreasing mortality among women 75 to 79 years of age was demonstrated by two recent randomized controlled trails from the Netherlands. Regular mammography use in this age group was associated with a 29.5% reduction in mortality from breast cancer, and the positive predictive value of mammography for breast cancer was 47% (versus 36% among women aged 50-69 years). Advancing age is associated with increased vulnerability to cancer and comorbidities. The presence of three or more comorbid conditions has been associated with a nearly fourfold higher rate of all-cause mortality at three years compared with women with primary breast cancer and no comorbidities. Several studies have used the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare program data to address mammography utilization, and breast cancer outcomes. Aside from the known risk of under ascertainment of comorbidities from Medicare claims data, the major limitation of these published studies is the difficulty in distinguishing screening from diagnostic mammograms. This distinction is important in certain situations, such as mislabeling a mammogram as screening instead of diagnostic when it is performed for evaluation of a breast complaint. Misclassification of mammograms leads to a bias in estimating screening mammography rates, breast cancer incidence, and outcomes, especially among elderly women with comorbidities who may not undergo regular screening mammography. This R03 research proposal is submitted as a complement to PI's NCI-funded K07 award "to examine the impact of comorbid conditions on breast cancer stage at diagnosis among elderly female Medicare beneficiaries" by utilizing the linked Breast Cancer Surveillance Consortium (BCSC) - Medicare databases. The linked BCSC - Medicare database has specific data elements (e.g., health history, screening history, radiologist's indication for mammography, hormone use, and presence of breast symptoms) to accurately differentiate screening from diagnostic mammograms. This proposal will answer several important questions about mammographic screening for breast cancer and stage at diagnosis among elderly women with comorbidities. The aim of this study is to estimate and compare the associations between specific comorbid conditions, and combinations of comorbid conditions, and (a) breast cancer stage at diagnosis and (b) utilization of screening mammography among elderly female Medicare beneficiaries without a prior history of breast cancer. This proposal is a collaborative project with BCSC investigators. The linked data are expected to be available in Fall 2008.This proposed project will be completed in 2 years (from 04.01.09 to 03.31.11).