AMIA: Why interoperability is 'taking so darn long'
Hospitals can have hundreds of IT systems. Vendors have built proprietary databases. Not everyone follows the same standards. Health systems fear sharing data with competitors. Policymakers have not focused on health information exchange or EHR usability.
[See also: AMIA: Regenstrief pumps up its clinical decision support ]
These are just a few of the reasons why true interoperability of health information remains so elusive, according to a panel of informatics luminaries.
"Technology is only one obstacle to interoperability," said Gilad Kuperman, MD, director of interoperability informatics at New York-Presbyterian Hospital, who moderated the panel at the just-concluded American Medical Informatics Association (AMIA) Annual Symposium about why interoperability is "taking so darn long.
"
[See also: Chicago proclaims Informatics Week as it hosts AMIA]
Charles Jaffe, MD, CEO of standards development organization Health Level Seven International (HL7) described a "circle of blame" involving government agencies and regulators, hospitals and healthcare systems, technology vendors, clinicians, academicians like those at AMIA and, yes, standards development organizations (SDOs), such as HL7. "The policy always preempts the technology," said Jaffe.
"And just like [in the 1983 Cold War movie] WarGames, in this finger-pointing, no one wins." He noted that not-for-profit HL7 in September made most of its standards and other intellectual property available free as a means of building trust for HL7 communications messaging. "Without trust, none of this is possible," Jaffe said.
Harry Solomon, interoperability architect at GE Healthcare, and a lecturer at Oregon Health and Science University, explained the road to interoperability with four numbers: 2, 4, 3 and 5.
There are two overarching concepts that need to be defined, namely interoperability and standards, and Solomon said "good enough" definitions exist from Institute of Electrical and Electronics Engineers (IEEE) and the International Organization for Standardization, known as ISO. Therefore, healthcare should not have to do any more in this arena. "We can't afford to have custom integrations for every data transfer that we have," Solomon advised.
The number 4 stands for the levels of interoperability specification: workflow, messaging, format, vocabulary.
The other two numbers represent three phases – standards development (generally handled by an SDO), product development (vendors), and system deployment (users) – and five process steps for each phase.
Showing 3 Comments
MedQuack say: The Short Order Code Kitchen Burned Down A Few Years Ago..
Actually I have used the title here quite a few times to address and educate folks on complexities...yeah it was easier to build and create when you had zero to build on and little integration for interoperability.
Congress doesn't get this either and makes it worse with some of their ambiguous questions at times...like the stimulus...duh:)
http://ducknetweb.blogspot.com/2012/10/senate-gop-members-call-for-meeti...
Dan Haley say: To achieve 21st century results, use 21st century tech
It is great to see so much focus on interoperability. But pondering our failure to interoperate in healthcare without considering the cloud is like bemoaning human inability to fly while ignoring the existence of airplanes. Simply put, the tools for effective interoperation exist. More, most of us use them in our daily lives every single day. For any number of reasons, healthcare has lagged decades behind the rest of the world in technological innovation. One reason for that lag is the fact that the healthcare information technology industry is dominated by large players heavily invested in legacy software systems jerry-rigged for the internet-enabled world. That makes no sense. Worse, government incentives intended to increase the "meaningful use" of health information technology have the unintended side effect of providing monetary life-support for those legacy software systems, which otherwise would have faded into the sunset by now. Boasting about use of direct messaging in 2012 is like boasting about sending an email in 2012. That might have been impressive in 1992; but these days, a direct electronic message sent between two points has the feel of rotary phone technology in an iPhone connected world. More thoughts along these lines here: http://www.athenahealth.com/blog/2012/10/24/there%E2%80%99s-no-such-thin...


Jeff Brand say: time to start over?
I agree that there are to many standards in healthcare. HL7 v.2, 2.2.5, CCD, CCDA, S&I Framework with many vocabularies. The complexity if extremely high. The number one selling EHR, Epic is the best example of Healthcare remaining in the past, Silos of data, non-relational database and one of the most difficult architectures to support interoperability. Why is healthcare still buying these types of systems in record numbers?
HL7 has taken a step in the right direction by promoting the new FHIR interface, a Health 2.0 architecture to ease the burden of connection of modern systems such as mobile. Though, a first step, legacy systems as well many of the new system continue to propagate siloed data.
Jeff Brandt
Collaborative author of mHIMSS Roadmap on Interoperability and Standards