By Cindy Atoji

November 11, 2008 | Molecular diagnostics laboratories that are doing clinical genomics “try to fit a round peg into a square hole, especially when it comes to workflow,” says Mark Hoffman, director of translational medicine at Cerner, the Kansas City-based health-IT player which has been targeting this market since 2005. “They try to use systems that were designed for traditional pathology and shoehorn those into a very unique and contemporary workflow, going into contortions to do that within their information system, especially with infectious disease testing.”

Cerner’s genomics solution is Millennium Helix, an offering that incorporates molecular diagnostic data into the electronic medical record, says Hoffman, who helped develop the bioinformatics system, which he says also streamlines processes and positions labs on the cutting edge of diagnostic medicine.

DHP: You helped define the Cerner strategy for integrating genomics with the electronic medical record. What is that strategy?

HOFFMAN: Genomics is a complex topic but our strategy is very straightforward. As we incorporate genetic test results into the electronic medical record, we associate with them with a unique code. So we’ve developed a new coding vocabulary for genetic test results; we call it the CBO, or Clinical Bioinformatics Ontology. And so, within the EMR, any of our genetic test results are associated with the concepts from the CBO, whereas a traditional observation like a pulse or blood pressure might be associated with a SNOMED (Systematized Nomenclature of Medicine) concept. All results are going into the same patient record, and that enables clients to use that information in a variety of creative ways, including personalized medicine and decision support.

Say a patient has a genetic test for a gene that indicates whether they’re predisposed for an adverse drug reaction, within our architecture, the system can evaluate medication orders against both the genetic test results as well as other factors like drug allergy and drug interactions. So it enables us to take a very holistic approach to personalized medicine, where it’s not just genetic, not just allergy, but a combination of any factors that need to be taken into consideration.

DHP: How is Cerner building on its base of clinical labs and the IT infrastructure already in place at these labs?

Millennium Helix is part of our overall offering to lab diagnostic laboratories, so it’s a modular approach. Many laboratories are doing microbiology and clinical chemistry, so this a modular on par with that, focused on a section of laboratory that is doing DNA- based testing.

DHP: Can you explain more about the CBO, which is at the heart of the Cerner Millennium system?

HOFFMAN: We began the CBO in 2003, and it has approximately 12 thousand codes within it. We evaluated other vocabularies like SNOWMED and LOINC (Logical Observation Identifiers Names and Codes) to make sure we weren’t duplicating any of their efforts, and we very quickly noticed that when comes to genomics, they had not dedicated a large part of their efforts to that. So we thought it would be a good contribution to the community to develop a vocabulary, and so for the past six plus years, we’ve been continuously updating the CBO. We publish it out to the public so anyone can download and utilize it. It’s open content, available through a broad license that is modeled after open source licenses, and we have registered users around the world. It’s been exciting to see the adoption of it.

DHP: What are the differences between Millennium Helix, and clinical genomics solutions from other vendors?

HOFFMAN: We think that our underlying informatics are very robust and very focused on providing both providing practical benefits with respect to workflow but also scientific benefits in term of how we structure and codify the information. I think we’re a leader in terms of even approaching this topic, or the issue of ‘What are the appropriate means to manage genetic information in the electronic health record?’ Being a first mover has given us a chance to define many of the important topics that other groups are now just beginning to discuss.

DHP: How is information presented in a way which users can handle?

HOFFMAN: We try to simplify the complexity of molecular data capture and provide immediate access to genetic information. I think many providers are uncomfortable with genetic information because the process of getting to those results is very complicated. I’ve observed that providers are more comfortable with more of a matter-of-fact presentation, where the interpretation is really the main focus—rather then how you got to that interpretation. Our clients configure the systems in a variety of way. We don’t tell them how to present the results, but a clear consensus seems to be to focus more on information and less than methodology, in order to increase the comfort level of providers.

DHP: How will Cerner expand its genomics offerings in the future?

HOFFMAN: One of our next areas of emphasis is: now that you have the genetic test results in the EMR, what do you need to do with them in terms of decision support? What can you do with those genetic test results to ensure that providers are making good use of that information? We’ll be rolling out some decision support systems to better enable providers to make good use of these results as well to know when a particular patient is eligible for a genetic test. Helping clinicians to recognize when it might be appropriate to order one of these tests is also a scenario of upcoming focus for us.

In addition to the overall comfort level to genetic tests, the other concern is that some physicians don’t feel the evidence-based is strong enough. I participate in a group that sponsored by the CDC, it’s called EGAPP, for Evaluation of Genomic Applications and Practice and Prevention. EGAPP is really focused on how can we better strengthen the evidence-bases for the value of genetic testing in terms of positive patient outcome, and by integrating some of analytic efforts with our genomic efforts, we’re also hoping to make contributions to that. Evidence-based is the other dimension of personalized medicine.