Health Data Standardization Improves Patient Care

To enable providers to make the most informed decisions about a patient’s care and reach “ready, reliable care” the Military Health System aims for, hundreds of data sources from disparate medical, dental, and readiness systems must be integrated. However, providers need consistent and standardized data to accurately diagnose and treat a patient’s medical condition. That’s not necessarily the case now, said Dr. Jesus Caban, the chief data scientist for Enterprise Intelligence and Data Solutions.
Soon, “if you have been diagnosed with sleep apnea, no matter where you receive care within the MHS, what data system is used to pull the information, or what analytical tools are employed to generate reports, the definition of sleep apnea will always be the same,” Caban said. Currently, different data systems may have different definitions for sleep apnea, and that could potentially affect a patient’s ability to get benefits for that condition as they pass through different organizations, such as moving from active duty to retirement as a veteran, according to Caban.

Common Data Model

To overcome that issue is a Common Data Model, which helps standardize medical vocabulary, Caban said. Standardization is one key component of the Defense Health Agency’s Strategic Plan for fiscal years 2023-2028.
Caban presented how the MHS is adopting a Common Data Model at the 2024 Healthcare Information and Management Systems Society Global Health Conference & Exhibition, in Orlando, Florida, on March 13.
“As part of the MHS stabilization effort, we see standardization of clinical practice guidelines, standardization in the electronic health record, standardization in the clinical workflows,” Caban said, “Now, we need to focus on standardization of data so everyone can count the same way.”
The first step in the process was to understand the vocabulary being used by industry and in academic settings, Caban said.
Among common data models used in health care, the Observational Medical Outcomes Partnership stands out as one of the most widely adopted across industry, academia, and government agencies.
In the early 2000s, the Food and Drug Administration spearheaded a public-private collaboration with pharmaceutical companies and health care providers to establish a common data model to standardize observational studies such as clinical trials. This collaboration led to the establishment of the OMOP common data model.
Stemming from that effort came the Observational Health Data Sciences and Informatics community. OHDSI is a “multi-stakeholder, interdisciplinary effort that standardizes vocabularies to create uniform analytics,” according to the OHDSI website.
“This open community has been providing guidance, recommendations, directions, mappings, and tools for health care organizations like the MHS to embrace a common data model,” Caban said.
OHDSI members include the Department of Defense, the Department of Veterans Affairs, FDA, and the National Institutes of Health. It has more than 2,000 collaborators across 74 countries and health records for about 810 million unique patients.
Research is another significant area where standardization will help. The DOD has numerous research efforts, many of which involve international collaboration. The MHS CDM will help streamline research by enabling faster integration of data across international partners and mapping of health data from diverse languages.

MHS GENESIS

The Program Executive Office Defense Healthcare Management Systems works to create interoperability and modernization of the DOD federal electronic record called MHS GENESIS.
On March 9, 2024, the DOD completed the deployment phase of MHS GENESIS across the global network of military hospitals and clinics. MHS GENESIS is the definitive and portable inpatient and outpatient medical record for service members, veterans, and their families across the continuum of care.
While the deployment phase of MHS GENESIS is complete, data gatherers still “face many challenges because there are inconsistencies” in medical and dental care reporting that necessitate ongoing optimization and enhancements, said Caban.
The sources EIDS PMO integrates include data from direct care, inpatient care, outpatient care, TRICARE, operational medicine, and ancillary applications, to name a few, he said. Added to that firehose of information are patient data from legacy medical records systems, and data from personnel and readiness systems.
For service members transitioning into or out of the military today, providers may want to look back 10 years ago in their medical and dental care records, according to Caban. But 10 years ago, the military was using the legacy records system called AHLTA.
These changes over time to the medical record pose challenges to a provider looking for whether a patient may have had a medical or dental condition in the past, Caban explained.
“Once you look at operational medicine, the systems run by our colleagues at JOMIS [the DHMS Joint Operational Medicine Information Systems], you have the added complexity of service members who may be treated by allied forces … for a short period of time and then transferred to U.S. military care,” Caban said.
Caban said EIDS plans to have 100% of MHS GENESIS and TRICARE data mapped by the end of the summer. That includes TRICARE inpatient and TRICARE outpatient data.
Benefits to Patients
Accelerating research and reducing the time from a clinical question to the answer is “one of the most important” benefits to patients, Caban said. Other benefits are interoperability and data scalability.
“As we work with other agencies, such as the VA, FDA, [and the Centers for Disease Control and Prevention], being able to have a common data model that we share … [produces] interoperability with those other federal agencies, and a key benefit is scalability as we bring more and more data sources forward. We can continue to scale up by adding more and more different datasets and databases and making sure they follow the common vocabulary,” Caban said.

What’s Next

The next phase in EIDS’ common data model effort is to work on its adoption and raise awareness throughout the MHS.
“Then, we will start doing a lot of user engagement sessions, training sessions to showcase the benefits of this, while at the same time adding other data sources because basically we started with two or three key data sources,” said Caban.
“Next year we’ll be working with our JOMIS colleagues to make sure the operational medicine data are included,” because “there’s some uniqueness in the DOD data—for example, deployments.” The OHDSI isn’t too familiar with operational medicine terminology.
The VA is going through a similar initiative using the Common Data Model. “We have been working very closely with the VA to make sure we know how they’re mapping the data; they know how we’re mapping the data; and we are mapping the data the same way or very similar way,” said Caban.