On September 9, the Health IT Policy Committee submitted a letter to the National Coordinator for Health IT, Dr. Farzad Mostashari, with recommendations for vocabulary standards. The committee recommends adopting and incorporating a smaller set of vocabulary standards into the Stage 2 certification requirements for Electronic Health Records (EHRs) Meaningful Use. The ONC offered a variety of vocabulary choices for Stage 1 certification, and this proposal would significantly limit the choices for Stage 2.
So why is this a good thing?
In a post I wrote in June, titled CDA Levels of Interoperability, I discussed the importance of not just exchanging the data, but working towards Level 3 type interoperability for CDA documents. There are three levels of interoperability, and the richness of the data increases with each individual level. The richness of data can be defined as the degree to which the receiving application fully understands the clinical meaning of what was captured by the sending application. The varying degree to which this meaning is conveyed is what HL7 refers to as “incremental semantic interoperability.”
Semantic interoperability can be most easily achieved when all applications are using a consistent set of coding standards that have enough detail to accurately describe the clinical event, with little or no required translation. This is analogous to two humans speaking the same language, with the same dialect, resulting in a fluid conversation without any misunderstandings.
It is also true for machines: if we can limit the amount of translation, we lower the likelihood that meaning will be lost. According to Dr. John Halamka, committee co-chair of the Health IT Policy Committee, the presence of multiple coding standards “creates a dizzying amount of mapping because vendors must support every variation.”
Level 3 interoperability calls for the coding of entries within the XML body of the CDA instance. The coding itself is optional as defined by the CDA standard, but ideally the entries should be encoded with the full power of the RIM with vocabulary code sets such as LOINC, SNOMED, RxNorm, etc.
The panel made recommendations on code sets in 23 categories. These categories range from Problems to Medication to Family History. And while it is impossible to use the same code set for all 23 categories, the panel limited the recommendation to a small set that would adequately address each of the categories. Some of the primary standards that the panel recommended across multiple categories include:
- Systemized Nomenclature of Medicine-Clinical Terms, also called SNOMED-CT.
- Logical Observation Identifiers Names and Codes, also called LOINC.
- RxNorm.
When discussing interoperability, the richness of the data is equally as important as getting the data to its rightful destination. These select, and already widely accepted, coding schemes will help move the industry in the direction of consistent and precise communication of events leading to semantic interoperability.
How will this impact the healthcare industry moving forward?
The initial proposal by the CMS for Accountable Care Organizations (ACOs) stressed their intent to align the technical requirements for an ACO with the criteria of the ONC EHR certification process for Meaningful Use. However, with the Stage 1 requirements as they exist today, there is little to ensure that the semantic meaning of the clinical events will follow the patient throughout their continuation of care.
Currently, there are a variety of vocabulary standards that can be utilized, and there are limited requirements to what sections must be filled in. But if an ACO is going to ultimately improve the quality of care while reducing the costs across a patient episode, full information of the patient’s previous care along with detailed coding of what has already been discovered must be shared among the applications.
The proposed vocabulary standards give applications the guidance they need to utilize a common language (code set) that conveys the consistent and detailed information that is required for quality care. With all applications using the same code sets, the ACO goals of quality and efficient care can be realized more quickly.
Rob Brull
Latest posts by Rob Brull (see all)
- Opinions about HL7 FHIR in the JASON Report - January 22, 2015
- What the JASON Report says about CDA - January 8, 2015
- The CCD/CCR Fork in the Road. Did Health IT Take the Wrong Path? - July 30, 2014