The need for interoperability is readily apparent in the healthcare industry. But what does interoperability mean with regards to richness of data? The CDA document is defined with three different levels of interoperability:
- Level 1: Includes the CDA Header plus a body consisting of an unstructured blob, such as PDF, DOC, or even a scanned image.
- Level 2: Includes the CDA Header plus an XML body with narrative blocks. Each section identified with a code.
- Level 3: Includes the CDA Header plus an XML body with narrative blocks and entries. The section should be encoded with the full power of the RIM with vocabulary such as LOINC, SNOMED, CPT, etc.
The richness of the data increases with each individual level, and so does the usefulness with regards to interoperability.
In its simplest form, it is fully valid for a CDA document to include nothing but an unstructured blob in body of its structure. While it is a conformant Level 1 CDA, it does not provide much functionality to the applications exchanging the data. From an exchange perspective, the applications may pass the file flawlessly. But from a quality of data, and thus quality of use perspective, this format is rather limited.
In CDA Levels 2 and 3, coding becomes a requirement that helps add clarity and consistency to the data being exchanged. And even when using Level 3 entries, information can be encoded at varying levels of specificity. When encoded at varying levels of specificity, information can only be analyzed at the highest common level. In order to provide the quality and richness necessary for an EHR repository or many other clinical applications, full Level 3 coding must be utilized at the highest possible common level.
This varying degree of specificity is what HL7 refers to as incremental semantic interoperability. This term indicates that an implementer can begin with a Level 1 CDA, and then add more structure over time. This allows the industry to ease into the standards, which is a good thing. But when discussing interoperability, it must noted that the richness of the data is equally as important as getting the data to its rightful destination. For interoperability to truly be successful, EHR repositories and clinical applications need to be armed with the richness of quality data that is coded to its fullest extent. Quality data will help lead to quality care.