Interoperability woes still plague healthcare in 2026. Here's what to do about it.

How interoperability brings holistic visibility to health systems

Achieving interoperability requires a holistic view of data across systems and solutions, according to Padman.

Ideally, when a patient goes for cancer imaging, biopsy results would get shared via a messaging standard like Health Level 7 (HL7) with a pathologist, a radiologist, an oncologist and primary care.

"In a seamlessly connected, interoperable environment, you really have all these experts all working together with access to information from the entire trajectory of the patient's care, but the reality is that even in a very advanced health system we are working with, that is not the case," Padman said.

Pramila Srinivasan, Ph.D., CEO of CharmHealth, a healthcare technology solutions provider, sees APIs as key to enabling interoperability.

"Building a strong, standards-based API program is essential," Srinivasan said. "Open APIs -- especially those built on FHIR -- give practices flexibility and reduce the friction of connecting third-party tools."

Although APIs can help, connecting mobile solutions to EHRs and patient records to decision-support applications, health systems still need to gain a full view of the workflow, whether it includes technology, people, or both, Padman explained.

So, health systems must train workers on how to share data using HL7, the SMART on FHIR API and FHIR, she said.

Healthcare interoperability challenges include governance, semantic standardization of complex medical concepts and business processes that regulate the exchange of data, according to Anjum Khurshid, M.D., Ph.D., associate professor of population medicine at the Harvard Medical School & Harvard Pilgrim Health Care Institute.

"Lack of interoperable systems will be a serious barrier to our health systems achieving that vision of an accountable, transparent, safe and efficient delivery system," Khurshid said.

For instance, physicians need to access data from other departments to complete their summary reports. Padman described how the community services division of one healthcare organization aimed to improve women and children's health services as well as infant mortality and health outcomes. To achieve these improvements, it needed to access health information, including a full view of costs, charges and claims, from its parent organization to complete assessments of its program from quarter to quarter as well as year to year.

"Organizations really need to remove these barriers that exist between the parent and the subsidiary organizations in terms of providing visibility," Padman said.

Strategies to overcome interoperability issues

Here are some ways health systems can bridge silos in data access.

Improve data quality: Improving data quality and understanding the process of data collection from the time the ED front desk or a kiosk checks in a patient can improve interoperability, according to Padman.

"You can put in checks and balances to make sure that the quality of the data that's collected at source via the technologies is as accurate and the quality is as good as possible when the human puts in the data," Padman explained. "You need to have technologies and solutions in place to do monitoring, sampling, periodic evaluations and so on, and this approach needs to happen at every point in the system where data gets collected. This goes back to understanding that full pipeline of how the patient's trajectory can be modeled."

Additionally, connected software tools strengthen data quality, according to Srinivasan. For example, the CharmHealthHub offers third-party tools such as the PatientPay billing platform, Zoho CRM and Medfinder medication database to provide key resources that connect to EHRs. Similar to the Epic Showroom, CharmHealth also has an app marketplace. It includes remote patient monitoring tools such as the Thinklabs One digital stethoscope and clinical documentation tool ScribeAI.

"When practices use more connected tools, we get a more complete and accurate picture of the patient journey," Srinivasan said.

Connect to health information exchanges (HIEs): HIEs are an integral component of interoperability, and most states have them now, Srinivasan said, noting that CharmHealth is integrating with as many of them as possible.

"They act as shared repositories of patient information, so a patient's history follows the patient rather than getting stuck in a single system," Srinivasan said.

Data fragmentation especially hampers the ability to train AI models because you cannot train a model on a single system's data.

"You need data that reflects many systems, which is closer to what an HIE provides," Srinivasan said.

Khurshid noted that HIEs were established in all states with public funds to address interoperability challenges.

"Today, HIEs are operational in most states and communities of this country and can easily play a key role in improving interoperability for AI, without the need to build new infrastructure," Khurshid said, citing his September 2024 study in NEJM AI.

Adoption of standards: Khurshid explained that full adoption of standards defined by the Assistant Secretary for Technology Policy within ONC and CMS, such as FHIR and the United States Core Data for Interoperability (USCDI), will be required to improve interoperability.

In addition, new frameworks and standards, like the Model Context Protocol (MCP), are emerging that can reduce the need for custom integrations, according to Srinivasan. She said protocols like MCP can help modern tools communicate with EHRs and other software tools more consistently.

 "With an interoperability layer like MCP in place, the model can plug into workflows immediately, and even non-developers can extend the ecosystem without having detailed knowledge of how APIs work," Srinivasan said.

In addition to achieving interoperability through technology such as data standards, interoperability will require cooperation among policymakers, entrepreneurs, patients, community and business leaders, Khurshid said.

Further, an interoperability strategy should include a coordinated population strategy as health systems build AI models, he suggested.

"Large health systems and health IT vendors are actively building these capabilities, but how do we make them available to everyone in every community, and not only to those who can afford to go to these select few locations?" Khurshid said.

He noted that while AI models could increase the rural-urban divide, the health IT industry could use existing shared data infrastructure to build tech tools that "serve both local and national goals of promoting interoperability and subsequent AI use."

Remember that as health systems address interoperability challenges, providing patients with easy access to their data will be essential, Khurshid underscored.

"Interoperability established around this concept of patient control and access will go a long way in helping overcome the nontechnical barriers to interoperability," he said. "This will also allow health IT leaders to pursue meaningful and impactful interoperability that contributes to improving desired patient and organizational outcomes."

Brian T. Horowitz started covering health IT news in 2010 and the tech beat overall in 1996.