Breast Cancer Surveillance Consortium: Working together to advance breast cancer research
Home   |   Data   |   Statistics   |   Tools   |   Collaborations   |   Work with Us   |   Publications   |   About   |   Links

Comparative Effectiveness of Breast Imaging Strategies in Community

Grant number: RC2 - ARRA
PI Name: Miglioretti D
Title: Comparative Effectiveness of Breast Imaging Strategies in Community

Film-screen mammography is the only screening test that clinical trials associate with reduced breast cancer mortality; however, controversy persists about how often screening should occur and the ages during which screening should be performed for subgroups of the U.S. population. Recently, advanced imaging technologies such as digital mammography and breast magnetic resonance imaging (MRI) are spreading into clinical practice despite limited evidence of their effectiveness in the community, where most women receive care.

The Breast Cancer Surveillance Consortium (BCSC) is uniquely positioned to quickly examine the comparative effectiveness of conventional and new breast imaging technologies and determine how screening strategies could be personalized for optimal delivery based on patient demographic and risk factor information. The BCSC has collected prospective data on breast imaging in community practices in five diverse states for 15 years. This ongoing expansive research resource includes high-quality longitudinal data with long-term follow-up on more than 2 million women and 95,000 breast cancers. It includes detailed information throughout the cancer care continuum: from cancer screening and diagnosis through treatment, surveillance, and death. It is the only resource of this size with risk factor information (including mammographic breast density) that is also linked to:

  1. mammography examinations accurately identified as either screening or diagnostic and as either film-screen or digital;
  2. cancer registry and pathology databases to determine both benign and malignant outcomes;
  3. Medicare and Group Health managed care claims data to obtain comorbidities and healthcare utilization; and
  4. death data to ascertain vital status and cause of death.

We propose to use and expand the BCSC infrastructure and network of breast imaging facilities to conduct comparative effectiveness studies of breast imaging strategies. We will do this by extending current data collection to include new imaging technologies and by establishing new collaborations with health services researchers, junior investigators, and modelers from the Cancer Intervention and Surveillance Modeling Network (CISNET). A consortium of National Cancer Institute-sponsored investigators, CISNET uses modeling to improve the understanding of how various cancer control interventions affect the U.S. population. Such models are invaluable when evidence from observational studies is insufficient to inform clinical practice recommendations and it is not feasible to conduct clinical trials large enough to test different combinations of imaging modalities. Through this Grand Opportunity, we propose additional data collection and collaborations that will increase our capacity to provide evidence on how to deliver the most effective breast cancer screening. Our specific aims are to:

Aim 1: Use existing BCSC screening data with long-term follow-up (overall and breast cancer mortality) and newly linked BCSC-Medicare data to compare the clinical effectiveness of 1- vs. 2- year mammography screening intervals for subgroups of women by age, race/ethnicity, comorbidities, breast density, and body mass index (BMI). Clinical effectiveness will be measured by the balance of favorable cancer outcomes (early stage at diagnosis and survival) and potential harms (cumulative false-positive and benign biopsy rates).

Aim 2: Extend results from a current BCSC comparative effectiveness study of breast cancer screening with digital versus film-screen mammography by using newly linked BCSC-Medicare data combined with Group Health managed care claims data to compare downstream healthcare utilization and costs between the 2 screening modalities for subgroups of the U.S. population >65 years by age, race/ethnicity, breast density, BMI, and comorbidities and build the capacity to compare the relative costs of services provided in fee for service and managed care settings.

Aim 3: Use BCSC data and results from Aims 1 and 2 in four CISNET breast cancer simulation models. These models will: compare the clinical and cost-effectiveness of three breast cancer screening strategies (film mammography, digital mammography, and triage to mammography type based on age or breast density) with different intervals and starting and stopping ages and explore how patient factors (age, race/ethnicity, BMI, breast density) affect ideal screening strategies.

Aim 4: Position the BCSC to evaluate the comparative effectiveness of new imaging technologies for breast cancer detection in community practice. We will implement a prospective data collection system specifically designed to capture clinically and scientifically relevant data on new technologies across our network of practice sites, initially by expanding data collection to include breast MRI. We will use these data to evaluate patterns of breast MRI use in community practice by patient characteristics and cancer risk factors.

Aim 5: Expand the BCSC Statistical Coordinating Center’s capacity to design and conduct comparative effectiveness research in community settings by developing epidemiologic and statistical methods focused on sampling and design strategies for prospective observational and community-based randomized trials.