charts

Projects

OncoShare

Project Description

What determines whether efficacy—the best results obtainable in a controlled clinical trial—translates to effectiveness in clinical practice, where wide variation in access to care, physician recommendations, and patient preferences alter outcomes? Such a question can only be addressed through the integration and analysis of data on patient characteristics, practice variations and treatment outcomes.

Through generous support from the Richard and Susan Levy Gift Fund, we have developed an informatics infrastructure, called OncoShare, to improve our understanding of the uncharted territory of real-world breast cancer care. OncoShare has been undertaken in collaboration with breast cancer specialists, biostatisticians, and policy makers at Stanford Hospital & Clinics (SHC) and the community-based Palo Alto Medical Foundation (PAMF) and with technical support from the Stanford Translational Research Integrated Database Environment (STRIDE) and the PAMF Research Institute. OncoShare continuously collects all available electronic health record data related to patients who received breast cancer care at SHC and PAMF since January 2000 and integrates data on these patients from local and statewide cancer registries. The OncoShare data resource currently contains detailed clinical information on over 15,000 patients.

OncoShare serves as a successful example of how academic and community-based researchers can create a rich, robust, and secure data-sharing framework that allows collaborative investigation of why one breast cancer treatment is chosen over another and how optimal care can be determined. Our informatics research addresses two challenges in temporal representation and reasoning: (1) how to develop a standardized yet flexible schema to integrate a wide variety of longitudinal breast cancer care data and (2) how to recognize intelligently which treatment options patients were offered and given using both structured data and clinical narratives.

Related Publications

BMIR-2011-1461
Comparison of semantic similarity measures for application specific ontology pruning
, A. K. Das
First IEEE Conference on Healthcare Informatics, Imaging, and Systems Biology, San Jose, CA, IEEE Press. Published in 2011
BMIR-2011-1460
Alignment and clustering of breast cancer patients by longitudinal treatment history
, A. K. Das
2011 AMIA Annual Symposium, Washington, DC. In Press in 2011

Stanford School of Medicine