Visual analytics support for the HEiDi virtual physician COVID-19 deployment
Drs. Tamara Munzner and Kendall Ho were award DSI funding for their project, "Visual Analytics Support for the HEiDi Virtual PHysician COVID-19 Deployment." This project will leverage advances in data visualization and analytics to optimze the delivery of telehealth care to patients stricken with COVID-19. The outcomes will help health system experts to gain a holistic snapshot of the current care system and expedite analysis and decision-making. Ultimately, this will allow the health care system to respond rapidly and deftly to current and future health-related scenarios such as a pandemic outbreak. Summary of project follows below.
In the current extreme situation of the COVID-19 pandemic, health systems face unprecedented medical and social challenges that data science can help address. Dr. Kendall Ho’s group in Emergency Medicine has spearheaded the HealthLink BC Emergency iDoctor-in-assistance (HEiDi) project to augment the 811 service delivering health care guidance to the public through telephone access to nursing advice by integrating virtual physicians (VPs) into the triage process, to help balance the enormous increase in load due to this crisis. This project is being deployed in extreme haste, rolling out within only three weeks what would normally take many months or even years, with operational and strategic concerns being addressed simultaneously. The new data being gathered through this project needs to be analyzed in the context of existing health system data including a) service utilization and call data including follow-up call-backs, b) patient metadata and health system usage outcome data, c) VP and nurse shift scheduling data, d) other system and administrative data, and e) health economics data. Even as the project is being deployed, Ho’s group is developing assessment criteria to establish its efficacy considering the dual goals of high-quality patient outcomes and satisfaction, and sustainable cost to the health care system of delivering care.
Dr. Tamara Munzner’s group has extensive experience in visual analytics (VA), building tools for human-in-the-loop decision-making in complex and heterogeneous data environments. Visual analytics approaches allow human analysts to comprehend the rich and nuanced nature of the full data landscape beyond the bare-bones descriptive statistics that summarize only the largest-scale trends. Her group has extensive experience in the methodology of conducting design studies, a user-centered and problem-driven design methodology in visual analytics, and in collaboration with experts in the genomics, biology, and medical domains.
This collaborative effort between Munzner and Ho will help the HEiDi project answer their driving data-centric questions including which information is needed for daily reports, how many doctors are needed in what optimal daily shift coverage for upcoming service days and follow-ups, which system usage outcomes change when virtual physicians interact with patients and with advising nurses, and which overload and triage issues remain severe. They will develop visual analytics workflows to represent as much information as possible visually in a way that expedites analysis and decision-making by health system experts, and their communication with many other stakeholders including clinicians, patients, and policymakers. The overall goal of this new collaboration is to provide visual analytics support for the HEiDi system, to help health system experts observe and improve the system even as it is built, in terms of virtual physician impact on patient experience, while adapting to the specific requirements of the current COVID-19 crisis.
In summary, this project will help health system experts to observe and improve the clinical pathway of 811 service on a strategic and operational level, to foster the effectiveness and efficiency of VPs, and to improve patient experience. It will also gain an initial understanding of stakeholders and this very interesting ecosystem, as a basis for a systematic and methodology-driven data science research multi-year project. Data science support for the 811 ecosystem will both address today’s urgent problems and serve to develop methods and tools that can be used in future extreme situations.