Unity Finds Harmony in a Private HIE Infrastructure

Despite its name, Unity Health System, in Rochester, N.Y., is somewhat divided.

As the third largest healthcare system in Rochester, it is home to a 346-bed single hospital with multiple departments and clinics, three nursing homes, about 70 remote sites—40 of which are physician practices— a home care facility, and a regional lab that has international ties. Thus, when it began down the path toward electronic health record (EHR) adoption, it was faced with disjointed systems and schedules.

“Each unit got on the train at a different time,” explains John Glynn, the vice president and CIO at Unity. “Our ambulatory practice was the first to [adopt] an EHR in 2003, they went with NextGen. The hospitals said, ‘We got to get an EMR.’ Well NextGen doesn’t have a hospital EHR, so we picked Cerner. Then our long-term care facilities went with Answers on Demand. Then our home care business went with Misys, which is now Allscripts.”

With four different EHRs, Glynn says the hospital needed to create a longitudinal view of the patient’s record. Different departments, clinics, and sites were unable to share data with each other. For example, Glynn notes that if a patient population was coming out of the emergency department from one particular practice, they couldn’t understand why because the two systems didn’t talk to each other.

In 2010, the New York State Department of Health released the HEAL 17 (Health Care Efficiency and Affordability Law for New Yorkers) grant around data sharing and improved care coordination. Unity focused on diabetic patients in their application. When it won the money, it was through the $14 million HEAL grant that it began to build a coordinated infrastructure that it would later use in a larger scope.

Creating Care Coordination

In late October of 2011, Unity invested in technology from dbMotion that allowed it to create an integrated health information exchange (HIE). dbMotion, which is now a part of the Chicago-based Allscripts, provides a longitudinal clinical data repository with semantically normalized patient data, point-of-care tools, a physician portal, population tools, and an analytics gateway.

A year after that contract was signed, the HIE, branded u.Net Connect, was up-and-running. Not only were the four EHRs able to talk with each other, but because Unity hooked up with the Rochester regional health information (RHIO) and connected public HIE to private HIE, it was able to gather data from the other two health systems in the area.

“From that creation of this semantic, interoperable data that dbMotion gave us, we’ve now been able to push that into a clinical data warehouse and start doing some population health management around that data,” Glynn says.

From that creation of this semantic, interoperable data that dbMotion gave us, we’ve now been able to push that into a clinical data warehouse and start doing some population health management around that data

Currently, Unity is creating dashboards and registries for this analysis itself and giving it to care managers. Glynn says the health system has gone from focusing on just diabetics to also looking at congestive heart failure patients.

The technology allows Unity to add new systems to the integrated HIE over time. This will come in handy when Unity merges with Rochester General Health System, a move that was announced in 2013 and is slated to go into effect later this year. “Plugging a fifth EHR into this into this infrastructure really becomes a great bridge strategy for us,” Glynn says.

Moving the Needle for Diabetics

Glynn says Unity has seen consequential results with its diabetic population health push. Among the 3300 patients involved with the program, hemoglobin a1c levels are averaging below 9 percent, a reduction of 15 percent for Unity. That happened over the course of 65 days, he adds, when the average time is typically 12 to 18 months.

John Glynn

“The satisfaction from a physician perspective, just within the four walls of Unity, has been huge,” Glynn says.

That’s not to say that this has been “easy” work. As Glynn notes, “It’s not plug and play.” Getting vendors on board with this kind of integration was something that took time. The project leaders also had to put the data into proper vocabulary to ensure there would be trust from the physician side. “There’s a lot of work you have to do to create that integration of the clinical data to ensure it’s semantically interoperable in a way that makes sense for them,” Glynn notes.

The satisfaction from a physician perspective, just within the four walls of Unity, has been huge

Unity plans on doing more population health analysis in the future with additional dashboards and registries. It will also add a capability within the HIE that allows additional patient data to be pushed out to the clinician proactively, rather than them having to search for it.

In Glynn’s eyes, this kind of interoperable technology makes the argument against “rip and replace” and the idea of a single EHR vendor across the enterprise. “I think in a day when capital and IT resources are thin, to spend the money to and toward a rip-and-replace strategy may not make the most sense. The HIE technologies are a great alternative strategy. We’ve shown that it can be done,” Glynn says.