Data interchange standards in healthcare IT--computable semantic interoperability: now possible but still difficult,...

The following article on HL7 Version 3 will give readers a glimpse into the significant differences between "what came before"--that is, HL7 Version 2.x--and "what today and the future will bring," which is the HL7 Version 3 family of data interchange specifications. The difference between V2.x and V3 is significant, and it exists because the various stakeholders in the HL7 development process believe that the increased depth, breadth, and, to some degree, complexity that characterize V3 are necessary to solve many of today's and tomorrow's increasingly wide, deep and complex healthcare information data interchange requirements. Like many healthcare or technology discussions, this discussion has its own vocabulary of somewhat obscure, but not difficult, terms. This article will define the minimum set that is necessary for readers to appreciate the relevance and capabilities of HL7 Version 3, including how it is different than HL7 Version 2. After that, there will be a brief overview of the primary motivations for HL7 Version 3 in the presence of the unequivocal success of Version 2. In this context, the article will give readers an overview of one of the prime constructs of Version 3, the Reference Information Model (RIM). There are 'four pillars that are necessary but not sufficient to obtain computable semantic interoperability." These four pillars--a cross-domain information model; a robust data type specification; a methodology for separating domain-specific terms from, as well as binding them to, the common model; and a top-down interchange specification methodology and tools for using 1, 2, 3 and defining Version 3 specification--collectively comprise the "HL7 Version 3 Toolkit." Further, this article will present a list of questions and answers to help readers assess the scope and complexity of the problems facing healthcare IT today, and which will further enlighten readers on the "reality" of HL7 Version 3. The article will conclude with a "pseudo-code" argument in favor of the adoption of HL7 Version 3, framed by citing the recommendation of the Interoperability Consortium for the use of HL7 Version 3 as a critical component in the National Health Information Infrastructure.