Duo teams to help simplify digitization of EHR

The introduction of the electronic health record was hailed as the solution to the American medical crisis. However, in reality it has resulted in physicians spending long hours completing notes after office hours have ended.

“Today, the electronic health record is touted as being one of the major drivers of physician burnout,” said John Windle, M.D., professor of cardiovascular medicine, and the Richard and Mary Holland Distinguished Chair of Cardiovascular Science at UNMC.

It turns out that simply digitizing the paper record doesn’t work, especially when computers don’t talk to other computers, Dr. Windle said.

In an effort to simplify the digitization of one aspect of the electronic health record, Dr. Windle and his son, Tom Windle, clinical informatics research analyst in the research information technology office at UNMC, have focused on a proposed medication data model.

Their work is supported by a six-month $250,000 grant from The Pew Charitable Trusts and is part of the Common Healthcare Data Interoperability Project, a collaboration with the Duke Clinical Research Institute.

The principle of their project is to create a structure in which individual drugs and classes of drugs are normalized across all specialties.

The two reviewed the data dictionaries of 40 specialties to see what each required and wanted for their registries, analyzed existing standards and compared what was different and similar among each.

Currently, each registry digitally encodes medications differently, said Tom Windle.

For example, some registries code the cholesterol lowering medication atorvastatin as a specific statin, while another might group it with other statin medications as one class with one code. Or it might be classified with different codes based upon whether the patient is in the hospital or not.

“To further complicate things, many electronic health record vendors have not adopted a formal classification system,” Tom Windle said.

After reviewing all the registries, Windle said they made recommendations on how the existing standards and terminologies could be harnessed in order to have interoperability and be something that vendors of electronic health record databases like EPIC could use when they create their databases.

“It would streamline provider workflow and make things more efficient and less frustrating across specialties for the physician,” he said.

At the Pew Charitable Trusts Conference held last month in Washington D.C., their work was presented to specialty societies that develop registries as well as members of the Food and Drug Administration and the Office of the National Coordinator for Health Information Technology and Medicare.

There, they discussed how this new data model would be implemented, key technical challenges and how it would be received by federal agencies such as the Center for Medicare and Medicaid Services.

“We have to get computers talking to, and understanding each other. We are close to getting medications that mean the same thing — whether it is EPIC or Cerner, an orthopedic, OB, or cardiology registry,” said Dr. Windle. “We hope to validate this work within the next 12 months.”