Home Care Tech Report: Data Analysis Tool Identifies Home Care Patients Ready for Hospice

In the past year, Great Lakes Caring, the largest healthcare at home and hospice provider in the state of Michigan, has significantly reduced its number of 7-day hospice stays while increasing its number of stays lasting 14 days or more. CEO William Deary attributes this change to a new software tool implemented in mid-2014. The agency’s corresponding uptick in revenue and cash flow has more than paid for the cost of the new system.

Using proprietary algorithms and sophisticated math,Medalogix Bridgecompares each patient’s status with key data points from thousands of patients and informs clinicians when the time has come to transition a patient from home healthcare to hospice. It is then up to the nurse’s and social worker’s professional judgment to determine the right moment to broach the subject with the patient and family, taking into consideration emotional and psychological readiness to hear the clinical rationale. Deary delivers the same message to his staff. “People are rarely familiar with hospice,” he explained. “They do not know that more appropriate care is possible there and that it does not necessarily imply ‘giving up.’ We let them know that some people are actually discharged from hospice from time to time.”

“We expect extraordinary effort from our nurses, such as our policy that all documentation be completed in the home,” Deary told us, “so our nurses are well within reason to expect extraordinary performance from us. That is why we implement innovative systems likeMedalogix Bridge.” He added that the system helps nurses focus on the “top 10” home health patients most appropriate for hospice, rather than feeling as though they have to have the hospice conversation with every patient. “Medalogix grabs the data directly from Homecare Homebase,” he added, “so there is no extra work required of our nurses.”

When a nurse and then a social worker fail to convince the patient and family that it is time for hospice, Great Lakes Caring refers them to the palliative care physician on staff.

According to Medalogix CEO Dan Hogan, Bridge analytics are highly accurate. The predictive model’s validity has been peer reviewed by data scientists at a prestigious university in Nashville, Tennessee and is proven robust and highly effective at identifying patients who could benefit from hospice care. “The math we use is sophisticated and precise,” he told HCTR in a recent interview from his Nashville office. “When Bridge reports, ‘past patients with the same characteristics (age, co-morbidities, symptoms, etc.) as this patient have succumbed to their illness within 180 days 95% of time,’ our clients can clearly see the tool’s accuracy.”

He related the story of a product demonstration with a prospective customer. A nurse on staff submitted her father as a sample patient but, when Bridge declared he was nearing the end and was a candidate for hospice, she became angry and voted against her agency adopting the system. “Her father passed away two months later,” Hogan reported.

Like the clinician in Hogan’s story, many of the Great Lakes Caring nurses were cautious before coming to trust the advice of a machine. Clinical judgment is subjective by nature and most nurses are understandably reluctant to believe that mathematics can ever be more accurate than a professional who talks to and touches the patient. Little by little, however, the extraordinarily high accuracy rate thatBridgecontinually demonstrated eventually brought the Great Lakes staff around. “They do not see this software system as their replacement,” Deary was quick to point out, “but as an objective, analytical confirmation of their clinical judgment. Most often, a nurse’s response toa Bridgerecommendation is, ‘just as I thought.'”

Medalogix is already integrated with Homecare Homebase, the EMR Great Lakes Caring uses, but Hogan told us there will be many more partnerships in the near future, both forBridge and for the company’s other flagship product,Touch.He describesTouchas a predictive analytics engine that help providers identify patients most at-risk of an exacerbation that would land them back in the hospital. A case study of theTouchdata analysis product and what it has done for a large Ohio agency is available on theMedalogix web site.

©2015 by Rowan Consulting Associates, Inc., Colorado Springs, CO. All rights reserved. This article originally appeared in Tim Rowan’s Home Care Technology Report.homecaretechreport.comOne copy may be printed for personal use; further reproduction by permission only.editor@homecaretechreport.com