Featured

Advancing to healthcare 2.0 with web 3.0

In the beginning, there were a few page authors on the world wide web.  People began to generate content – disconnected and isolated as it was.  In this genesis moment, the Yahoo!’s and AltaVista’s of the world wanted to organize all of the content of the web.  With yellow pages-like rigor, they formulated a taxonomy for the internet – news, sports, humor, etc.  And then at some point there was a big bang, the amount of context exploded, and taxonomies quickly fell by the wayside.  These aggregators of the early internet simply could not keep up.  The model changed from a world that knew exactly what existed and how to categorize it to one that knew of no boundaries to the information available and instead focused on making it connected and accessible.  Not only that, but the world demanded more information, more democracy in syndication of content.  All of this poured into the great well of information that the web had become and with the onset of this chaos, organization of content fell into the lap of consumers, not authors.  With the continued evolution of the “cloud,” the nebulous mass of information owned by individuals but living in the ether – organization was only usurped by the need for portability of that information.

Information is the lifeblood of healthcare, used by and influencing the decisions of everyone from patients to providers to payers.  But health information is far removed from whatever humble roots it may have had.  Government regulation and the demand for information as part of the never ending flow of funding for research has created information that lives in a largely disaggregated state, each element having a largely singular purpose and connected to the whole by and large by human beings looking for patterns and relationships that influence decisions.  Standards exist in various forms, and more and more are being imagined to find a panacea for the liberal exchange and use of this information.

Oddly, the world wide web has already tackled this problem, albeit more generically.  Some forward thinking people, including web founder Timothy Berners-Lee imagined a global network of information that would need to be modeled as a series of defined abstractions as to what data described.  The Resource Description Framework or RDF is a standard that suits exactly this purpose – aiming to make the description of a resource fully portable but not universal.  Additionally, these same forward thinking people envisioned a need to model relationships between resources – such as how a person might be related to a corporation. The same aim applied – to make these descriptions portable, but not universal.  For instance, in some case, a person may be a litigant against a corporation, which in another they may be described as being employed by the corporation.  For this we have the Web Ontology Language or OWL. Together, these two standards are the primary concerns of the semantic web.

In healthcare, we have several key motivators for information interchange and exchange of standards.  Business intelligence, the general idea of applying reason and analytics to understand the various measures of the business domain so that decisions can be influenced, is a big one.  For years, data warehousing and data federation have been the tools used to tackle problems of assembling domain information in ways that would allow meaningful analysis.  But these tools both exist because they are complimentary, not because they are necessarily competitive.  Secondly, data integration is an increasing concern amongst public and private entities alike.  While HIPAA mandates portability as part of it’s very name, the idea of a truly portable health record has been but a dream.  Competing standards are evolving, whether it be CCD or CCR, or even the amoeba-like HL7.  But as they are currently constituted – they are merely candidates for competitive evaluation rather than integration, and at the very most, subject to the same Extract-Transform-Load paradigm from data warehousing methodology or data federation of complimentary existing systems.

It is all of this effort and redundancy that make healthcare and portable health information a robust candidate for consolidation as part of the semantic web, the backbone of web 3.0.  In a world that is full of disorganization and competing standards, the use of RDF as interchangeable and mergeable pieces as from pieces of the business domain into an expanding, scalable ontology for that domain will spare the need for convergence on how to describe the domain and instead focus on enabling interactivity.  If the CCD or CCR were defined as RDF, anyone could easily combine these two interfaces which describe the domain of the medical record into an ontology that best describes the problem at hand.  Then, analytics can occur by understanding the domain and the relationships in that domain, rather than requiring a deep understanding of the data itself.

Tags: business intelligence, CCD, CCR, Data Federation, Data Warehouses, OWL, RDF, Semantics, XML

Ryan Norris is a technical architect at Medullan. He has a background working with physicians and clients in the eHealth space in developing and delivering solutions that better interconnect providers and create more informed patients through extensible service architectures and highly-usable interfaces.