The volume of medical data for physicians to sort through is growing at an exponential rate and is projected to double every 73 days by 2020. Among healthcare executives, consensus is emerging that AI and other analytic tools will be essential to transform this expanding mountain of data into actionable insights that improve quality metrics and patient care. At the Center for Cancer and Blood Disorders (CCBD) in Fort Worth Texas, clinicians have turned to the Jvion Machine to harness CCBD’s vast clinical data to power improved oncology care while reducing costs under the Oncology Care Model (OCM).
The CCBD has found the Jvion Machine to be particularly effective at lowering risk for patients in value-based arrangements. At any given time, the CCBD is managing 1000-1100 patients who need case management oversight under the value-based care model. Barry Russo, CEO of CCBD, noted it would be impossible to hire enough case managers to focus on all these patients. With the Jvion Machine, case managers can prioritize care for the highest risk individuals, leveraging the Machine’s literature-backed interventions tailored to the needs of each patient.
To identify high-risk patients, the Jvion Machine analyzed all of CCBD’s clinical data and integrated it with socioeconomic data provided by government agencies and third-party sources. Socioeconomic risk factors, including lower education levels, lower income, living alone, not owning a home, among others, can signal more complex care needs within patient populations. Linking socioeconomic and clinical factors is the key to the Jvion Machine’s precision in identifying both risk and personalized interventions.
Together with CCBD, Jvion stratified patient risk across seven key clinical areas: 30-day mortality, 30-day pain management, 6-month depression risk, 6-month risk for deterioration, 30-day avoidable admission, 30-day emergency department visit, and 90-day readmission.
Jvion’s AI-enabled prescriptive analytics flagged patients who traditionally lay outside of the highest risk stratum and would otherwise never be identified as at risk of morbidity or mortality. Jvion’s enhanced approach allowed case managers to focus on the patients truly at highest risk and with the greatest likelihood of responding favorably to timely clinical intervention.
While CCBD is still validating the seven vectors internally and optimizing Jvion’s recommendations, the Jvion Machine can already automatically refer patients for pain management. Meanwhile, the psychology team is proactively calling patients flagged by the system for depression risk. CCBD social workers are also proactively calling “socially challenged patients”, who may need help filing for financial assistance. Russo noted that the provider’s OCM score improves with each patient that files for help and is approved.
In the future, CCBD hopes to leverage the Jvion Machine to identify inpatient fall risk. In other settings, Jvion’s fall risk vector has been attributed to a 75% reduction in resources required to prevent falls when compared to the performance of the status quo Morse Falls Scale, and a projected reduction of 20% of all inpatient falls. CCBD also hopes to apply Jvion to the outpatient arena so that patients can be automatically referred to CCBD’s pre-hab program if they are at risk of future injury. Under the OCM, rehab costs are attributable to CCBD and are higher than costs for inpatient admissions, so CCBD hopes to leverage Jvion’s AI-enabled prescriptive analytics to prevent injuries that would require rehabilitation.
1Densen P. Challenges and opportunities in medical education. Trans Am Clin Climatol Assoc. 2011;122:48-58.