For people 65 years of age and older, COVID-19 hospitalization rates in the U.S. have been as high as 1,245.7 per 100,000 population, straining the resources and capacity of health systems. ChristianaCare needed to effectively deploy its care management resources, identifying patients with COVID-19 who were most at risk of severe illness and hospitalization. By leveraging its analytics platform and the predictive power of the Healthcare.AI™ solution to provide COVID-19 risk prediction, the organization was able to provide targeted interventions to those most likely to benefit and help patients avoid unnecessary hospitalization.
In the U.S., COVID-19 hospitalization rates for patients 65 years of age and older continue to rise, stretching the resources and capacity of health systems.1 Predictive analytics can help healthcare organizations identify patients at risk for COVID-19 hospitalization, potentially preventing complications, and proactively address resource constraints.2
COVID-19 challenged ChristianaCare to make the most effective use of their care management team. It was challenging to identify patients with COVID-19 who were at risk of severe illness. ChristianaCare was using simple rule-based methods to prioritized outreach and provide the interventions required for high-risk patients with COVID-19 to avoid hospitalization, which was critical due to limited available resources in the wake of the pandemic, such as care management personnel, portable pulse oximeters, home care staff, and inpatient beds.
ChristianaCare needed a solution that could expeditiously and equitably identify high-risk patients with COVID-19, who have an increased likelihood of hospitalization, enabling the organization to prioritize and deliver targeted interventions to quickly mitigate adverse outcomes.
ChristianaCare leveraged the Health Catalyst® Data Operating System (DOS™) platform and a robust suite of analytics applications, including data science tools and predictive analytics, to identify patients with COVID-19 who were at risk for hospitalization. The organization first identified patients 18 years and older who tested positive for COVID-19 within the last 24 hours and who had not been admitted to the hospital. ChristianaCare then created a machine-learning model to predict the likelihood of hospitalization.
The machine-learning model uses 15 different factors to predict hospitalization and assign a risk score to each patient. The patient list and risk scores are pushed to ChristianaCare’s CareVio™ care coordination team dashboard. Care coordinators reach out to and assess patients with the highest risk scores and enroll patients in the appropriate care management program, enabling optimal allocation of resources and effective management of COVID-19.
ChristianaCare can use the data platform to evaluate differences in hospital admission, ICU admission, ventilator utilization, and length of stay. The organization can also assess the final discharge disposition difference between patients identified as high risk who were enrolled in the care coordination and those who were not enrolled.
ChristianaCare’s COVID-19 hospitalization risk-prediction model effectively enabled the accurate identification of patients at the highest risk, allowing the organization to make data-informed decisions about resource allocation and enroll patients in the appropriate care management program, positively impacting patient outcomes. Results include:
We quickly recognized that we could use the Health Catalyst data science tools to provide clinicians the insight required to direct clinical resources for COVID-19. Our care coordination teams rapidly integrated predictive analytics into their workflow, reducing hospitalizations from COVID-19, and avoiding nearly $2M in costs.
- Ed Ewen, MD, Director, Clinical Data and Analytics, Center for Strategic Information Management
ChristianaCare will continue to expand its use of data science and predictive analytics to improve the care provided to its patients and the effectiveness of its operations.
