Hackensack Meridian Health is testing the utilization of predictive analytics technology to improve timely hospice transitions.
A group of hospice providers and data scientists at the New Jersey-based health system have developed artificial intelligence (AI) technology aimed at helping its clinical teams to better identify when a patient is nearing the end of life or could benefit from serious illness care.
A main goal is to develop a statistical model that helps provide benchmarks for clinicians to know when to have serious illness and end-of-life care conversations with patients and families, according to Dr. Lauren Koniaris, vice president and chief medical informatics officer at the health system.
“Think of this as a ‘check engine’ light,” Koniaris told Bloomberg. “It’s a gentle nudge to help us take the best care of our patients.”
The predictive analytics tool calculates the likelihood that a patient will die within six months and become eligible to receive hospice care. The technology flags declines in patient conditions through the health system’s electronic health records that prompt physicians to hold discussions around hospice and palliative care.
For example, if a patient’s likelihood of death is found to be 70% or higher, clinical teams receive notifications for hospice evaluations. A likelihood of 30%–69% triggers a palliative care consultation.
Dubbed Serious Illness Care Connect, the software is being tested among roughly 150 physicians across Hackensack Meridian Health’s service region.
The health system offers hospice, palliative, primary and pediatric care, behavioral and home health, among other services. Hackensack Meridian Health operates three academic medical centers, 18 hospitals and more than 500 patient care locations. The nonprofit health system employs more than 35,000 team members, including roughly 7,000 physicians.
A larger pilot phase is in the works for 2024. A challenge the health system is working to address includes assessing whether the predictive analytics software disproportionately recommends hospice and palliative conversations among particular patient populations, according to Sameer Sethi, senior vice president and chief data analytics officer at Hackensack Meridian Health.
As of May 2023, the health system’s analysis of the software showed the model had no specific biases, Sethi stated.
Hackensack Meridian Health is among a slew of other hospice and palliative providers leveraging technology to grow utilization.
Predictive analytics systems have gained momentum in hospice and palliative care as a way to improve access among patients in need earlier in their illness trajectories. Providers have increasingly used these systems to build their census and ensure that patients and families receive the right care at the right time.