BOSTON — Beth Israel Deaconess Medical Center and Lahey Health, along with three other hospitals, are currently in the midst of a mega-merger to bring the organizations under one umbrella. But as with most providers, they’re operating with a wide range of EHR vendors and versions.

There are three instances of Epic, three different versions of Meditech, athenahealth and Cerner, and the assumption is that we’ll rip and replace these versions and put in one monolith EHR, explained John Halamka, MD, Beth Israel Deaconess CIO at HL7’s DevDays on Tuesday.

[Also: Next-gen data management: From FHIR APIs to AI to genomics, tech is changing fast]

“That might be true if I had $2 or $3 billion,” or if I had the nerve to uproot the current EHR workflows, said Halamka. “But I don’t want to do that. So instead, what we’ve said is FHIR. Why don’t we just create a suite of apps, device neutral, then enable providers to have the appropriate workflow and experience, regardless of the transaction in the EHR.”

“This will enable me for the next five years or so to keep the EHRs in place and give people a unified experience,” he added.

But while the use of FHIR will allow the organization to save costs and fuel interoperability, there are still a few challenges in the U.S. to making full use of it, explained Halamka. “People think FHIR does more than it does at the moment. The trajectory is excellent, but we need to temper expectations.”

One of those issues is the lack of a patient wide matching system, where people are assigned a unique identifier. Halamka doesn’t think the industry can expect any changes in that under the current administration, nor any “major change in national identification habits.”

“We need to see some kind of service to track patient identity -- demographics alone aren’t sufficient,” said Halamka. Instead, a service could break down patients by their first name, a relative’s name and other things that can help us figure out who you are.

“We’ve created ways of saving that data and taking it place-to-place, but we’ve never had a way to get there,” he continued. “We told doctors they had to drive, but we didn’t build any roads. And we’re going to hold them accountable if they don’t get there.”

Another issue is that data sharing policies vary by state. So, for example, New Hampshire has different policies for the type of data that can be shared and its purpose, which differs from neighboring Massachusetts, he explained. “How are you supposed to coordinate care?”

While FHIR improves these issues, there are still 50 different states and six territories, “so 56 different policies,” said Halamka. “As we move from fee-for-service we have an imperative to improve care quality. But today, we can’t survive in a risk-based environment unless we share data.”

The quality and source of data also prove problematic, as the industry lacks standards, he explained. Consider consumer data, what should providers do with that data and how should it be interpreted?

“I think, the answer is that we need metadata, around the data, and we hope, maybe as a society we develop rules,” Halamka said. For example, FitBit is allowed to have a variable heart rate, but with a pacemaker, it’s not allowed.

“But please recognize, all of the underlying data you work with will have its flaws,” he continued. “Consider the source of data. There needs to be a solution to this mismatch problem because you can’t update everyone simultaneously. We need to accept the fact we’re going to have a lot of variation.”

And change won’t come from policymakers. “The government used to drive the agenda -- but I really think the private sector economy is where the action is … the change is going to come from you.”

Twitter: @JessieFDavis
Email the writer: jessica.davis@himssmedia.com