Lab test dummies

By Brian
02:03 PM

For physicians, receiving simple, accurate, electronic versions of lab test results could be the single biggest incentive for adopting electronic health records.

There’s been a steady stream of technical standards in this area in the past few years. First were the Logical Observation Identifiers Names and Codes (LOINC) and the Systematized Nomenclature of Medicine standards. Then came the EHR-Lab Interoperability and Connectivity Standards (ELINCS) for sending electronic reports between labs and ambulatory care sites. We are now in the era of the Healthcare Information Technology Standards Panel’s Interoperability Specifications.

Unfortunately, even though substantial progress has been made toward standardizing the content of electronic messages from labs, there is no way yet of making sure the message gets to the intended recipient. That’s partly because labs have different ways of identifying recipients and patients.

“If every vendor used ELINCS, we would be close to being able to interoperate, but still these other things keep turning up,” said Dr. J. Marc Overhage, president and chief executive officer of the Indiana Health Information Exchange and director of medical informatics at the Regenstrief Institute. “It always feels that we are 80 or 90 percent of the way there. Most of the hard work has been done, but it’s not yet plug and play.”

A national patient identifier would help interoperability because it would ensure a certain level of confidence that the information that’s being exchanged refers to the right person, said Joshua Temkin, a senior director at General Dynamics IT’s Health and Human Services Group.

An agreement on a common coding scheme would also be beneficial, he added. Even though LOINC has been around for some time and there are now more than 40,000 codes available through it, most labs have their own way of representing test results. A tool is needed to convert results to LOINC before they are transmitted.

“There are a number of software services that are in the early stages of allowing that, but there’s still a lot of work that needs to be done, particularly given that labs continue to change their coding schemes,” Temkin said. “It’s an ongoing battle.”

And it’s a costly endeavor, both in dollars and human resources. Labs must map their coding schemes to LOINC and update them as the standards evolve, he said.

There are also gaps in the standards that define how information should move from the lab to the physician. For example, ELINCS defines the shape of a message — the patient identifier, the lab that performed the tests, etc. — but it doesn’t specify the appropriate network transport protocol for such messages.

Although most labs that rely on the Health Level 7 (HL7) standards use TCP/IP to send their messages, it’s not a given that all labs will send messages that way. They have many options for sending messages, including via the Web, File Transport Protocol or virtual private network.

In the end, no standard
“There is no standard, defined way of doing this,” said Jonah Frohlich, senior program officer at the California HealthCare Foundation, which developed ELINCS. “In practice, it’s up to the lab and the individual doctor’s office to make that decision.”

Most interoperability discussions focus on point-to-point solutions, where messages go directly between a lab and a doctor’s office. That’s the simplest scenario, Frohlich said, and the one that addresses the most immediate need.

However, the requirements change dramatically once more complex sharing arrangements, such as a regional health information organization, come into play. That model requires a more complete EHR that enables a doctor to see information about the care patients have received in any setting. It could also involve incorporating results from a number of labs.

Ultimately, it’s a question of cost. “On a purely technical basis, this is doable and is being done now,” Frohlich said. “But it’s relatively expensive, so there’s an opportunity cost to all of this. [For people switching to electronic messaging], it’s very hard to quantify the difference between fax and electronic.”

Non-standard standard
The health IT community’s goal is to incorporate all of the various standards into HL7. But even if everybody on the vendor and user sides agrees to use HL7 for electronic messaging, there’s still no guarantee of interoperability. Not everyone uses the same version of HL7, and some might prefer to stick with an older, proven version that doesn’t include the newest standards.

And even if everyone uses the same version, there are differences based on how organizations implement the specification, Temkin said. That makes it difficult to build any data exchange.

Developers are trying to overcome the problem through revisions to the HL7 Reference Information Model, he said, but success hinges on the implementation guide that is delivered with the final specification.

“All of this is absolutely required” for interoperability, he said. “This is the next biggest hurdle in the field.”

In the end, however, the bigger barrier to interoperability will not be technical. Rather, it relates to age-old concerns about privacy, security and return on investment.

States have different rules about what kind of medical information can be transmitted electronically and to whom. Such restrictions hamper interoperability, which requires that labs be able to freely send information to physicians and other labs, even if they are in different states.

Efforts are under way now to harmonize privacy rules across state lines, but experts say we are years away from developing a national consensus.

But the biggest barrier of all for interoperability — as it is for health IT in general — is that there isn’t yet a financial model to spur the adoption of the standards and tools that would enable it, said Melissa Chapman, vice president of health IT strategic programs at General Dynamics and former chief information officer at the Health and Human Services Department.

“Even if the technology issues are overcome, the core model still hasn’t been modified to allow for the adoption of IT processes” that enable interoperability, she said. “The return on investment is challenging enough for established technology, let alone for something as fluid as health IT.”

As things stand now, doctors don’t have an incentive to adopt health IT, she said, and the labs don’t see what they have to gain by adopting what in all likelihood will be expensive processes with no guarantee of a return.