During a clinical encounter, providers are often distracted from having conversations with their patients about their health status and disease risk, given the need to document clinical data in the electronic health record (EHR). The promise of the EHR is that the data entered by providers during a clinical encounter can be used for real-time patient monitoring and shared-decision making, and for secondary use by researchers. EHRs have the ability to bring data to life, to make clinical data relevant for patients and providers, and to allow for a more efficient and patient-centered clinical encounter. To do so effectively, we cannot lose sight of the importance of stakeholder engagement when developing EHR-based tools and algorithms. This talk will describe the process for creating an EHR-based tool for addressing cardiovascular health in the primary care setting. The tool was designed to automatically pull data from a variety of clinical sources, and its use resulted in improvements in cardiovascular health among patients. In the context of this project, we demonstrated 1) use of an evidence-based cardiovascular health algorithm at the point of care; 2) usability of the tool by providers; 3) improvements in cardiovascular health among patients with access to the tool; and 4) informatics can be considered an interventional discipline. The result of this work encouraged us to consider other settings in which this tool could be implemented and evaluated. Since cancer survivors are at increased risk for cardiovascular disease due to having poorer cardiovascular health compared to the general population as well as their potential for receipt of cardiotoxic cancer therapies, we refined the aforementioned tool with the input of oncology providers. This talk will enumerate the lessons learned from the process of adapting an existing EHR tool for use among a new population of patients and providers. Next steps for this work will include modifying existing cardiovascular disease risk scores for validation among cancer survivors. Current algorithms tend to underestimate cardiovascular disease risk among survivors since they take into account traditional cardiovascular risk factors but not factors that represent the unique experience of cancer survivors, including receipt of cancer therapies that may increase the risk for cardiovascular disease in this population. It is critical for algorithms to learn from new evidence and evolve to better predict risk among patients. This talk will outline best practices for the creation of algorithms that can complement EHR-based tools and serve as a resource for shared decision-making for disease prevention and management.

Citation Format: Randi E. Foraker. EHR-based tools and algorithms for managing cardiovascular risk among survivors [abstract]. In: Proceedings of the AACR Special Conference on Modernizing Population Sciences in the Digital Age; 2019 Feb 19-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(9 Suppl):Abstract nr IA19.