By Ken Barbe, CEO, IMAT Solutions
LinkedIn: Ken Barbe
LinkedIn: IMAT Solutions
Healthcare has made significant progress in making patient information more accessible, but a recent congressional report highlights persistent interoperability challenges.
According to the Congressional Research Service (CRS), 100 percent of nonfederal acute care hospitals and 91 percent of office-based physicians reported using certified electronic health records (EHRs) in 2024. Yet barriers remain, including uneven health IT adoption, information blocking, gaps in the Trusted Exchange Framework and Common Agreement (TEFCA), and governance challenges.
These findings are particularly relevant as healthcare organizations accelerate AI adoption. Beyond exchanging information across systems, organizations must ensure their data is complete, accurate and consistent enough to support increasingly sophisticated applications.
Interoperability Progress Has Not Eliminated Data Gaps
While healthcare has made progress in connecting systems and establishing data exchange standards, significant gaps remain. The CRS report found that only 19 percent of surveyed behavioral health facilities participated in a health information exchange (HIE), while public health agencies continue to face funding and technical challenges.
Even when information is exchanged, inconsistent terminology, coding structures and patient identifiers can limit its usefulness. Duplicate records, missing information and conflicting details often require additional review, creating inefficiencies in care coordination, quality reporting and population health management.
These challenges also raise questions about whether healthcare data is sufficiently reliable to support AI applications.
Why AI Raises the Stakes for Data Quality
AI is creating new opportunities to improve clinical and administrative workflows, support predictive analytics and help healthcare organizations make better use of information.
However, the effectiveness of these applications depends heavily on the quality of the data supporting them.
An AI application analyzing patient populations, for example, may draw information from EHRs, claims systems, laboratory results and other sources. If those records are incomplete, inconsistent or incorrectly matched, the application may generate unreliable conclusions or overlook important patterns.
The challenge becomes more significant as organizations expand AI across multiple departments and use cases. Poor data quality that previously required manual reconciliation can become a larger operational problem when the same information is used repeatedly across automated processes.
AI can accelerate the analysis of healthcare information, but it cannot automatically resolve every weakness in the underlying data. This makes data readiness an essential consideration for organizations seeking to move AI initiatives beyond isolated pilots and into everyday operations.
Building an AI-Ready Data Foundation
Healthcare organizations can begin addressing these challenges by treating data quality as a core component of their interoperability strategies.
That requires looking beyond whether information can move between systems and evaluating what happens after it arrives. Several capabilities are particularly important.
- Data aggregation: Bringing together clinical, claims and other healthcare information from disparate sources to establish a more comprehensive view of patients and populations.
- Normalization: Applying consistent terminology, formats and coding structures so information from different systems can be interpreted and analyzed together.
- Validation: Evaluating data for accuracy, completeness and consistency before it is used in reporting, analytics or AI applications.
- Patient identity reconciliation: Identifying and resolving duplicate or mismatched patient records to improve confidence in the information being analyzed.
- Data governance: Establishing clear processes for monitoring data quality, managing access and maintaining accountability as information moves across systems and applications.
These capabilities establish a more reliable foundation for AI while improving operational efficiency. By creating shared processes for cleaning, reconciling and preparing data, organizations can reduce manual effort and support a wider range of clinical, operational and analytical applications.
Making AI Readiness Part of the Interoperability Strategy
Healthcare organizations should not view AI readiness as a separate initiative that begins only after an AI application has been selected.
Many of the capabilities required to support AI are the same ones needed to improve quality reporting, population health management, care coordination and value-based care.
For example, an accountable care organization working to identify gaps in care needs timely, accurate information from multiple providers and systems. A health plan preparing for digital quality measurement depends on consistent clinical and claims data. An HIE supporting regional data exchange must address differences in how participating organizations collect and structure information.
Each of these activities benefits from the same underlying investments in aggregation, normalization, validation and governance. A practical starting point is to assess the current data environment, identify weaknesses and prioritize improvements based on the organization’s operational and strategic needs.
Healthcare leaders should also establish measurable indicators of progress, including data completeness, error rates, processing time and the amount of manual intervention required to prepare information for use.
These measures can help organizations determine whether interoperability investments are producing improvements in data usability rather than simply increasing the volume of information exchanged.
The Next Measure of Interoperability Success
The latest congressional findings reinforce the need to address persistent interoperability challenges, including uneven participation, governance gaps and information blocking. However, healthcare organizations must also focus on the quality and usability of information after it has been exchanged.
As AI adoption accelerates, complete, standardized and trustworthy data will be essential to supporting reliable applications, improving workflows and maximizing interoperability investments.
Ultimately, interoperability success should be measured not only by how effectively information moves between systems, but also by how well it supports better decisions, greater efficiency and improved patient care.