Healthcare Integration and interoperability have started to play a key role in healthcare. Much of this can be attributed to healthcare transitioning from a volume-based incentive model to a value-based model. With how hospitals are set up today, providing value-based care is largely unsustainable. Healthcare organizations have had to alter priorities from provider to patient. To better provide for patients and to maintain compliance with the HITECH Act and other mandates, EHRs have become an integral part of the majority of hospitals. EHRs are a significant resource of clinical data and clinical workflows. However, because many EHRs are unable to talk to each other or talk to applications outside of their own silos, data within the EHRs remain largely inoperable.
Why Integration is Needed?
Being that there is a wealth of patient information siloed away within EHRs across thousands of organizations, it should serve as no surprise that many of the world’s cutting edge healthcare systems are looking to expand beyond the constraints of their silos to see the larger picture. However, in order to bring together all of the pieces of the puzzle integration becomes a necessary evil.
Integration is the exchange between providers, payers, vendors, and other important players that brings data or function from one application to another. Integration is important because when we look at the quantity and diversity of data involved in the healthcare system, it would be virtually impossible to process or analyze without breaking through the data silos. Traditional health IT systems like EHRs and PHRs utilize completely different technical and semantic standards to depict and house data. Making it extremely difficult to correctly and simply integrate data from various conflicting systems.
Levels of Data Integration
Data integrations vary primarily in the level on which they focus. There’s the application level, the API level, and the data exchange level. From an application level, data from various applications are reimplemented into one domain. From an API level, applications expose their API so other sources are able to access their data. At a data level, the integration establishes a common domain archetype or global schema. This gives independently developed applications the ability to exchange information. Of the 3, the data level has the most flexibility because it focuses on a common exchange.
In an effort to encourage the adoption of data exchanges and meet the standards of federal programs, HL7 has been introduced as the standard to transmit information. With HL7, there comes a handful of free templates. While this standard has been working (somewhat), and continuing to evolve many of its users view it as complex, flat, and delimited.
Challenges in Integration
Today integrations are challenging but important. They are the cornerstone for advanced healthcare applications. Once streamlined, integration opens up many benefits including a decrease in clinical errors, enhanced care coordination efforts, and increased patient care endeavors. Larger than that, applications give us the ability to further public health research opportunities and population surveillance to identify real-time issues.
An EHR stores a linear record of patient health info. It includes patient demographics, progress notes, problems, medications, vital signs, past medical history, immunizations, lab data, and radiology reports. By offering a clinician the EHR record, it helps automate and streamline their workflow.
Given a large part of the picture exists within the EHR, we would have expected integrating into and out EHR systems would have been easier for other sources. However, for the majority of hospitals, it requires a skilled team of experts to bring the idea to fruition. This team typically includes:
- Project Manager
- Operational Owner
- Systems Administrator/Network Engineer
- Interface Engine Analyst
- EHR/Application Interface Analyst
- EHR Web Service Analyst
- EHR Analyst
- Support staff
Healthcare Data Standards
In addition to orchestrating all of the involved parties, it is also important to make sure the hospital is meeting all of the healthcare data standards interpretations. Despite its complexity, there is a linear list to follow for the project initiative.
- Planning & Paperwork: Prepare contracts, Business Associate Agreements, and Kick-off meetings
- Gather Requirements: Sample HL7 Messages, sample JSON, data dictionaries, APIs, and associated documentation.
- Infrastructure/VPN: Infrastructure should be spun up with the VPN creation and verification
- Setup Interface: Make sure the front end of an organization’s endpoints are set up, including the interface
- Testing: Use the sample messages collected previously to review and acknowledge with inbound and outbound to resolve any potential issues
- Go Live: Migrate to production and deploy. Data feeds are opened and the integration will allow other sources to populate the system
- Ongoing Support– Unfortunately, this is not a set it or forget it situation. Make sure your support staff is alert, well trained, and ready to tackle all of the questions that will inevitably arise from this implementation.
As you can imagine, with all of the hands that have to go into creating and maintaining this integration, it can serve as a major bottleneck for health systems. Resulting in lower quality of care for patients and minimal ROI for providers. An EHR’s complexity can fall into 3 main buckets; custom data mappings, the security of data connectors, projects requiring different degrees of project management. Many organizations feel forced to limit the data they integrate because of a lack of understanding of what is possible. These issues become worse and more complicated with scale- the total opposite of what healthcare actually needs.
As a solution, think of the process as multiple linear steps. Stop thinking about HL7 message types and VPN handshakes and instead, work to enable the workflows your system needs to be successful. It goes beyond learning the technicalities of HL7 and FHIR. Your team will have to become domain experts, including on HIPAA because for the integration to be compliant, it must exist on a HIPAA compliant infrastructure.