Grady Health System, one of the nation's largest public academic health systems and a safety net provider serving the Atlanta community, is using Jvion's solution to reduce the risk of re-hospitalization within their patient population.
Grady Health System (GHS) is one of the nation's largest public academic health systems and a critical safety net provider serving the Atlanta community. GHS provides quality, cost-effective, and patient-focused care for some of the most at-need, complex patients. Through Grady Memorial Hospital six community health centers, and leading specialty centers, GHS is a vital and integral part of Atlanta and the region.
GHS is committed to "becoming the leading public academic healthcare system in the United States." Part of this commitment rests on driving quality outcomes for each individual patient. To help support GHS's goal, the system initiated a project leveraging predictive analytics to reduce Congestive Heart Failure (CHF) patient readmissions. This portion of the patient population is at increased risk of complications and mortality. According to a study published by the American Heart Association, close to a quarter of CHF patients discharged from an acute care facility are readmitted within 30 days.
Within GHS's patient population, there are many socioeconomic factors that pose challenges to preventing readmissions. The organization knew they needed an advanced, machine learning solution that could account for the thousands of variables impacting a patient's risk of a readmission. GHS needed something accurate, affordable, and effective that could be used by our practitioners at the point of care.
GHS contracted with Jvion to initially help lower rates of CHF patient readmissions. In the seven months since GHS deployed the predictive solution, GHS has been better able to target and support this critical patient group. Glenn Hilburn, Vice President, Clinical Systems for GHS said, "With the predictions delivered by the Jvion tool, our practitioners are better able to zero in on high-risk patients. This has helped us optimize our resources and shift care to individuals with CHF who carry with them the greatest likelihood of a readmission."
The kind of patient-centered machine learning that drives Jvion's solution is designed to become more accurate as new data are added to the tool. The solution in place at GHS has consistently identified almost all of the at-risk patients. And the system will continue to drive greater accuracy and coverage with every new patient entered into our predictive machine.
With CMS (the Centers for Medicare & Medicaid) driving higher penalties and broader adoption of value-based models, having accurate predictions that enable prevention will be essential to ensuring sustainability. "We have to be able to pin point those individuals who are most at-risk so that we can better align our clinical and community resources. And in the complex social and economic environment in which we deliver care, we require the kind of predictive computing power delivered through patient-centered machine learning solutions like Jvion's to reduce suffering and help us do more for our patients," said Hilburn.
Going forward, GHS plans to expand the program to include all cause readmissions for the system's full patient population. The health system is also looking to Jvion's community prediction platform to help identify and serve rising risk individuals while they are still in the outpatient/community setting.