UNC Health uses Epic to pull in other EHR data to create better comprehensive patient view

Using the company's "Happy Together" tool, UNC was able to document 12 percent of A1c test completions and 8.1 percent of flu shot completions from external data sources.
By Bill Siwicki
09:37 AM
Share
Epic EHR UNC Health

Caldwell Memorial Hospital in Lenoir is among the hospitals in the UNC Health Care system. Credit: Google Maps

It’s very difficult for physicians and other frontline caregivers to get complete pictures of patients’ healthcare, of course, since patients so often receive care at multiple locations whose IT systems may not be interoperable.

But the University of North Carolina Health Care is making great strides in this area, culling data from external sources and pulling it into its Epic electronic health record by using a special technology design philosophy from Epic called Happy Together.

For example, as of Jan. 31, 2018, for the diabetic A1c test, 5,120 external completions out of 42,619 total completions were pulled into UNC’s EHR; that’s 12 percent of completions due to external data. Patients who have diabetes or prediabetes should get Hemoglobin A1c testing at regular intervals. Of the patients who have this completed or are up to date, 5,120 are due to external data out of 42,619 total completions.

[Also: UNC Health reaches Stage 7 with advanced EMR, analytics capabilities]

As of the same date, for flu vaccines, 15,004 external completions out of 185,021 total completions were pulled into UNC’s EHR; that’s 8.1 percent of the completions due to external data. For the current flu season, if one looks at all patients eligible for this health maintenance topic and one looks at the patients who have gotten their flu shot this season, 15,004 patients had an external flu shot or entered it via My UNC Chart out of 185,021 patients who had their flu shot completed.

“We know that many patients receive care scattered across multiple providers and health systems,” said Robb Malone, vice president of practice quality, innovation and population health services at UNC Health Care. "Coordinating care across the continuum is necessary in order to sustain the health and well-being of the patients we serve.”

The availability of data from multiple health systems directly in the patient’s chart improves UNC’s ability to care for its patients and helps reduce redundant or unnecessary testing, which prevents inconvenience and/or pain for the patient; saves healthcare dollars for patients, insurers and care providers; and saves time for both the patient and provider, Malone added.

“The availability of A1c data from multiple health systems is a great example and a major development in the management of diabetes,” Malone said. “Using the additional data now made available in the patient’s chart, providers can coordinate care more proactively and ensure we meet the needs of our patients based on their full clinical history.”

Happy Together is a design philosophy. It refers to how Epic blends natively documented data with externally documented data into a single, integrated picture, to improve patient care and care coordination. Data is woven together into a comprehensive narrative for clinicians, care managers, community members and patients, according to Epic.

Specifically, some EHRs treat outside data as separate standalone documents, very similar to an electronic version of a fax. Such an approach introduces usability challenges as clinicians and caregivers must scan through sometimes voluminous documents to get a full picture of a patient’s care.

Happy Together extracts the discrete facts from these documents and presents these facts interwoven with native documentation, removing duplicates and creating a single problem list, medication list and more, according to Epic.

But the integration goes beyond the view. For example, calculating a patient’s clinical risk needs to take into account the patient’s full clinical picture, which is difficult to do when the patient receives care across many organizations.

Epic’s risk scoring algorithms can calculate based on non-native documentation, like the number of emergency department visits the patient has had – regardless of whether the emergency departments use the same EHR – and information about their clinical conditions that might not be known to the local system, Epic said.

“Integration of the data can be separated into two parts,” said Mike Plesh, director of information technology in UNC Health Care’s information services division. “The configuration and the connectivity. The configuration was the complex part, where the Epic@UNC EHR teams had to map the individual data fields so they match the mappings that are to be used across the other Epic organizations.”

This was achieved through a workgroup consisting of all Epic partners across North Carolina to create the common values to be used in the data exchange. Epic sees Happy Together as a philosophy because it requires collaboration and agreement across many organizations. The connectivity was the less complex part of the project because it uses the already existing Care Everywhere model that allows for the sharing of patients across Epic organizations, Plesh explained.

North Carolina’s health information exchange, NC HealthConnex, had a role to play, too. 

“UNC Health Care is a foundational partner with NC HealthConnex and has had the opportunity to improve how data is shared with them and all Epic organizations across North Carolina,” Plesh explained. “One of the features UNC Health Care and Epic helped collaborate on with NC HealthConnex was a way to determine if an organization is using Epic as the EHR and, if so, to not return other Epic data back to another participating Epic client.”

The reason for this is because the data is already being shared through the Epic-to-Epic Care Everywhere functionality and the organizations did not want duplicated data returned. Where the organizations get the benefit from NC HealthConnex is by being able to integrate non-Epic organizations’ data into their Care Everywhere screens and providers can consume it in the same manner as if the data came from an Epic organization.

UNC Health Care learned a variety of lessons along the way during this hefty project, and shares some with the industry at large.

“Data mapping is complicated but really important,” said Annie Whitney, program manager of population health solutions on UNC Health Care’s practice quality, innovation and population health services team. “In order for this to work, trading partners have to work together on data configuration and mapping so the data shared is usable. We’ve also reached out to the NC HealthConnex about data mapping.”

Feedback also is an important lesson learned, according to Whitney. Clinicians like the integrated views, and seeing visits and notes from external organizations in-line with UNC information; this saves time and prevents them from having to go back and forth into different screens, she said.

Twitter: @SiwickiHealthIT
Email the writer: bill.siwicki@himssmedia.com