In January 2024, the Centers for Medicare and Medicaid Services updated The CMS Interoperability and Patient Access Act. The new revision outlines requirements and specifications for what information medical providers need to provide, as well as how it should be formatted to ensure API security and data compliance. This is towards the goal of improving interoperability for the healthcare market.

According to a new report from Coherent Market Research, the global healthcare API market is primed to explode in the next ten years. Analysts predict that the industry could expand from $242.7 million in 2025 to $372.5 million in 2032. They also report an 18% rise in subscription-based API deployments across various healthcare sectors. There is clearly a huge driver for interoperability in healthcare APIs.

With that in mind, we’ve put together some best practices for improving interoperability to help make your healthcare APIs as efficient, secure, and compliant with the latest guidelines.

1. Master the HL7 Guidelines

Healthcare interoperability isn’t an arbitrary thing. Worldwide, organizations dictate data sharing standards and how to implement them. The most recent set of guidelines is called Health Level 7 (HL7), which specifies everything from how your healthcare data should be stored to how security and privacy should be approached. It outlines the Fast Healthcare Interoperability Resources (FHIR) specification.

Taking the time to read through the HL7 guidelines will remove all confusion and uncertainty about how to format, secure, and share your healthcare data, guaranteeing your healthcare products comply with the latest guidelines and best practices. For example, HL7 dictates that clinical and administrative information be organized into resources, which are standardized building blocks like Patient, Observation, Medication, and Appointment. These resources can then be combined, extended, or constrained using profiles and extensions to meet specific organizational or use case requirements while still adhering to the core FHIR model. Finally, HL7 demands that data be represented using JSON, XML, or RDF, making it available for a wide range of tools.

2. Follow REST and Implement Security

FHIR uses a RESTful API model to exchange data using standard GET, POST, PUT, and DELETE HTTP verbs. It also specifies that an API’s capabilities be summarized using CapabilityStatement resources. StructureDefinition and Implementation Guides further guarantee that resources can be used throughout an organization, further emphasizing interoperability.

As far as security is concerned, HL7 insists that developers use either OAuth 2.0 or the SMART on FHIR framework to ensure secure authorization and authentication. If a developer is using a legacy system, they may need to either use Transport Layer Security (TLS) or IPsec to make sure that transmissions are secure.

3. Use the Interoperability Maturity Model

There are different levels of data interoperability. The ability for one healthcare tool to communicate with another would be an example of one level of interoperability. One system being able to interpret assets from another system would be another level of interoperability. Becoming familiar with the requirements of each level is an essential step in making sure your healthcare tool can communicate with any other safely and securely.

Many developers working in the healthcare sector use the HIMSS Interoperability Maturity Model, a standard created by the Healthcare Information and Management Systems Society to evaluate the ability to share, integrate, and use healthcare data across different systems, identify gaps in interoperability, and plan strategic improvements that lead to better care coordination, efficiency, and patient outcomes as a way to visualize different levels of interactions.

The Interoperability Maturity Model comprises four levels:

• Level I, the Foundational Level, specifies the requirements necessary for one component to communicate with another.
• Level II, the Structural Level, defines the requirements for data exchange at the field level as well as the necessary syntax and data organization.
• Level III, the Semantic Level, allows underlying models to standardize and codify data using publicly available datasets and coding vocabularies.
• Level IV, the Organizational Level, sets the standards for data governance, legal, and organizational standards for exchanging information both within and outside the organization.

4. Understand FHIR Specifics

FHIR is one of the most popular healthcare data exchange standards currently in use. It uses modern web technologies like JSON, XML, and RESTful APIs to share and access patient information seamlessly. FHIR simplifies the process of exchanging healthcare data by breaking it down into standardized, reusable resources like patient records, medications, or clinical observations, allowing them to be easily shared across different platforms.

FHIR is the latest incarnation of HL7, which was created in the 1980s to manage electronic health records (EHRs). FHIR was explicitly designed to make use of web-friendly formats like JSON and XML, which are far more familiar to casual users than the highly specialized interfaces of HL7. This familiarity means that FHIR makes secure, interoperable healthcare data available for users with limited technical experience.

5. Make Use of Modern AI Data Platforms

Modern healthcare providers are working with more data than at any point in history. They’re also dealing with more complexity, with data coming from many different sources and stored in many different locations. AI data cloud platforms help minimize many of these issues, allowing large amounts of data to be assessed, sorted, and processed in many different ways without increasing your development team’s workload.

Using an AI data platform also helps guarantee that your healthcare data is secure and compliant, as it can be customized to work with different specifications. Best of all, AI data platforms help make sure your healthcare data complies with current regulations without sacrificing speed, as large language models (LLMs) can process many other queries, processes, and tasks without putting much of a strain on your system.

To illustrate how modern AI data platforms can improve interoperability, consider the case of UH Cleveland, where nurses were struggling to collect, compile, and interpret patient data in a timely and efficient manner. This bottleneck caused everything from miscommunication to unnecessary delays and reduced decision-making capabilities.

To solve these problems, UH Cleveland developed EdgeHuddle in partnership with Edgility, an AI-driven data platform that aggregates patient data from EHRs and medical devices, analyzes it, and presents actionable insights during each shift’s safety huddle. The system highlights emerging risks, trends, and leads to more focused discussions. Not only does this prevent the possibility of accidents from occurring, but it also leads to better care for patients as it allows nurses and healthcare providers to focus on meaningful work instead of wasting time on unnecessary tasks.

6. Get Everybody On The Same Page

Healthcare providers have many different components, each of which has its own goals to meet and priorities to fulfill. When you’re optimizing your organization’s healthcare data for interoperability, it’s a good idea to sit down with all of the different teams and learn about their specific goals and needs.

The C-suite might let you know they’re only comfortable budgeting $500 a month for hosting data, for example. The cybersecurity team might reveal an uptick in cross-site scripting (XSS) attacks through a public-facing endpoint. Taking the time to learn about each team’s goals and needs will allow you to come up with a thorough plan on how best to update your API to improve interoperability.

7. Emphasize a Data-Driven Culture

Data policies work best with widespread adoption. Otherwise, organizations can be left with blind spots where they’re forced to guess or make assumptions, eliminating many of the benefits that data can offer. It’s important to emphasize a data-driven culture throughout your entire organization. This can help each team decide on which metrics are important for their specific goals and which data needs to be shared with the organization. This is an important step in making sure that everybody stays on the same page, as clear metrics let each team know how they’re performing and ways that they can improve.

Consider the case of Lakeview Health Systems, a healthcare provider that acquired three community hospitals that each had their own EHR system. This created all manner of complications for healthcare data, preventing staff from accessing necessary data and sometimes even resulting in duplicate tests. This data disconnect convinced Lakeview Health Systems to lean into a data-driven culture, integrating a central data platform to unify clinical, operational, and patient data across all systems. Focusing on data reduced hospital readmission rates by 18% within a year, signifying both an improvement in healthcare as well as a more cohesive culture across all four hospitals.

8. Standardize Your Data

Standardizing your data is another essential step in improving interoperability. Enforcing data standards enables all of your enterprise’s data to be used interchangeably throughout your entire organization, making it as observable and secure as possible.

Important data classification standards in healthcare include:

• Systematized Nomenclature of Medicine — Clinical Terms (SNOMED CT), the globally recognized, comprehensive clinical terminology used in EHRs
• Logical Observation Identifiers Names and Codes (LOINC), the most widely used terminology standard for health measurements, observations, and documents
• The 10th revision of the International Classification of Diseases (ICD-10), a medical classification list by the World Health Organization

Using standardized coding systems helps ensure that your data is compliant with the current standards and best practices around medical data. Having data standards in place makes it easier to translate and convert data from outside organizations and vendors, too.

9. Establish a Data Center of Excellence

A data center of excellence (CoE) acts as a central hub for all of your organization’s data initiatives. A data CoE can serve as a central repository for all of your organization’s data. Such a group can decide on goals and important key performance indicators (KPIs), create a knowledge base around your organization’s approach to data, and share it throughout the company. Establishing a data CoE can make use of model-driven interoperability, which is an excellent way to make sure your healthcare API is efficient and secure.

10. Test Your APIs

With API security and data compliance being so important, there are numerous tools for testing to make sure your APIs are secure and compliant. HealthIT.gov offers a suite of tools to make sure your healthcare APIs comply with Fast Healthcare Information Retrieval (FHIR) standards. HealthIT.gov’s Inferno test kits check that your healthcare APIs comply with ONC Certification Standards, US Core Standards, UDS Test Kits, UDS+ Test Kits, and many, many more.

Final Thoughts on Improving Interoperability in Healthcare APIs

Healthcare is only going to become more digitized from here on out. This means that the demands for API security and data compliance will only become more pronounced as time goes on, as well. This means that the demand for interoperability is only going to become more pronounced.

To help guarantee that your healthcare API’s interoperability is up to par, you should look into some of the AI platforms that are out there to make sure your healthcare data is secure, compliant, and scalable. Cultivating a data-oriented culture in your organization is one of the best things you can do for data interoperability, as it will help enforce data conformity and governance both within and outside the organization.

Having a thorough understanding of the current guidelines and best practices around official standards will help you implement data standards throughout your organization, as well. Finally, tools like HealthIT.gov’s Inferno toolkit let you check that your healthcare APIs are compliant with the current standards around data interoperability. Following all of these best practices and making use of the tools out there will help you build, develop, use, and grow your healthcare API without fear or stress.