UnityPoint Health has established multiple predictive analytics models to curb readmissions by assessing factors such as length of hospital stay, readmissions "heat zones," and no-show appointment risk, Mike Miliard writes for Healthcare Finance News.
According to Miliard, the health system successfully cut its risk-adjusted readmission index at a pilot hospital by 40 percent over a three-year period. That success has propelled the system to build out a more holistic approach toward readmissions by going beyond simply targeting at-risk patients in need of intervention and using analytics to inform the best intervention for each patient.
As Ben Cleveland, a data scientist at UnityPoint, put it, "How can we use analytics to inform the intervention after [patients at high-risk of readmission] leave? How can we assess the likelihood of success for those interventions?"
How the models work
To address those long-term, holistic factors, the health system created a data analytics model that calculated not only a patient's overall risk of 30-day readmission, but also the patient's daily risk for readmission each of the 30 days post-discharge. "What we found is that some patients were at much greater risk of coming back early on after their stay, and then others tended to be more at risk later on in that 30-day timeline," said Cleveland.
According to Cleveland, the health system "developed a risk heat map over that 30-day timeline that visually depicts a patient's risk very quickly." The map shows "heat zones" of the specific days within the 30-day timespan a patient is most at risk for readmission, Miliard reports, which allows the staff at UnityPoint to more efficiently schedule interventions.
UnityPoint also developed a freestanding no-show appointment model that can be used in conjunction with the readmission risk map to help clinicians customize strategies around each patient's individual risks, Cleveland said.
"If clinicians and care teams decide to schedule follow-ups, we'll actually compute the no-show risk," he said. "So you've identified their readmission risk over time, you've planned two follow-up appointments with their [primary care provider]—but it turns out [the patient has] a high risk of not showing up for those, so you have to augment your strategy a little more to ensure interventions actually happen."
According to Cleveland, UnityPoint also created a "freestanding length-of-stay model that predicts how long a patient will be in the hospital," which Cleveland said has helped with discharge planning and resource allocation.
Implementation is key
While the tools themselves are useful, the key factor for success is careful implementation, Cleveland and Rhiannon Harms, UnityPoint's executive director of strategic analytics, said.
Harms said, "We recognize that there's training and education we need to provide as analytics professionals to our clinical and business leaders on how to use predictive analytics. It's a different way of approaching the problem than looking in the rear-view mirror with descriptive analytics. We've really tried to do some proactive work on the analytics competency and training piece."
As part of those efforts, Harms said their training has focused "on the use case and working to be a partner to our clinical and business leaders, to enable them to drive better results," adding, "It is for us about developing better solutions with the end workflow and end result in mind, and doing that in partnership with those who provide care to our patients" (Miliard, Healthcare Finance News, 8/23).
Primer: The CFO's guide to readmission penalties
The Hospital Readmissions Reduction Program reduces hospital inpatient reimbursement up to 3% based on performance relative to national averages across five conditions.
We've analyzed the latest CMS data to share how hospitals did, how excess readmission ratios can be used to calculate how many readmissions you are 'over', and the conditions that drive the largest penalties.