Being able to identify key risk factors and health indictors across the end-of-life continuum is critical to ensuring that hospice patients receive the right care, at the right time. VNS Health is able to accurately anticipate important transitional cues throughout a patient’s end-of-life journey, thanks to a suite of AI-driven predictive tools. These tools can place patients into certain risk profiles, in order to help facilitate earlier detection of hospice patients and better manage their disease trajectory.
In May 2022, the Centers for Medicare and Medicaid Services (CMS) introduced the Hospice Visits Last Days of Life (HVLDL) score into the Hospice Quality Reporting Program (HQRP). VNS Health’s executive-level objective for 2023 was to outpace both the New York State and national HVLDL average scores.
Together with the hospice operations team, VNS Health Business Intelligence and Analytics developed an evidence-based machine learning algorithm, designed to identify patients who were likely to expire within the next seven days. This data was then pulled into a dashboard that allows hospice teams to schedule the right amount of visits for these patients.
After a limited pilot period, the dashboard was soon made available to all of VNS Health’s hospice teams. Team members were encouraged to use the actionable insights provided by the dashboard to guide their recommendations for care assignments and scheduling visits.
Since its launch, the dashboard has calculated over 812,000 million risk scores for over 10,000 hospice patients. VNS Health achieved a 61% improvement in HVLDL quality scores. Read the VNS Health case study, Improving HVLDL Performance by 61% with Predictive Analysis, to learn more about the mortality risk dashboard and its dramatic impact on end-of-life quality outcomes.
To learn how your health care organization can use the HVLDL Mortality Risk Dashboard to improve quality scores and to enhance quality of life for individuals during the last 7 days of life, contact VNS Health at [email protected]
