Search

Integration Navigation

 

One of the most frustrating things in healthcare is that there isn't a central patient ID, so sending data between systems requires customization. In the early EHR days, promises of HL7 compatibility made us hopeful that we'd all be connected, but integration engine & custom mapping costs often exceeded the cost of the EHR itself.

 

Here are several options to consider for data exchange:

 

Application Programming Interface (API) is basically a set of tools for developers to integrate with applications.

  • API stands for Application Programming Interface, and is basically a set of tools so developers can integrate with applications.  

  • An application company decides what data is shared.

  • API access can be obtained through app stores (or marketplaces, as some EHRs call them), by accessing published APIs, or through custom, privately-shared APIs.

  • Companies may charge to share data through their stores, or for making API access available.

 

Health Level 7 (HL7) standards were implemented to ensure data could be exchanged between platforms.  Because each system has its own naming conventions and data mapping, it’s not as simple as it sounds.  Some examples:

  1. Last Name vs Surname vs Full Name

  2. Discrete data vs free text

  3. Picklists with different values:  Even a system that has added a third Gender can cause data confusion when other systems are looking for two genders.

Most EMRs were sold as “HL7 Compliant,” so there was mass frustration in the industry when integrations were an additional, significant cost to practices, and pushback still exists.

 

 

Fast Healthcare Interoperability Resources (FHIR) combines HL7's standardized fields and API's more modern web protocols to create a simpler interface. It's becoming more prevalent, already in use in some EHR marketplaces, and should be in widespread use over the next two years.

 

 

Web Data Extraction is more frequently known as screen scraping. A developer creates code to perform tasks that a human would perform, and the data from those screens is harvested and sent to the receiving party. An example would be to tell the system to look for a new patient, gather the demographic & clinical information, and save it to the receiving platform.

 

 

Additionally, creative solutions through projects like Carequality, database connections, and other methods are available.