A Pharmacogenomics Clinical Decision Support Service Based on FHIR and CDS Hooks.

OBJECTIVES: Pharmacogenomics (PGx) is often considered a low-hanging fruit for genomics-electronic health record (EHR) integrations, and many have expressed the notion that drug-gene interaction checking might one day become as much a commodity in EHRs as drug-drug and drug-allergy checking. In addition, the U.S. Office of the National Coordinator has recognized the trend toward storing complete sequencing data outside the EHR in a Genomic Archiving and Communication System (GACS) and has emphasized the need for "pilots that test Fast Healthcare Interoperability Resources (FHIR) Genomics for GACS integration with EHRs." We sought to develop a PGx clinical decision support (CDS) service, leveraging the emerging FHIR and CDS Hooks standards, and based on an assumption that pharmacogene sequencing data would be stored alongside the EHR in a GACS. METHODS: We developed a PGx CDS service as a functional prototype. The service is triggered by a medication order in the EHR. When evoked, the service looks for relevant genetic data in a GACS and returns corresponding recommendations back to the ordering clinician. Where the patient has no genetic data on file, the service can recommend pretreatment genetic testing where applicable. RESULTS: Overall, we were able to meet our objectives and deploy a functional prototype, interfaced with a commercial EHR. We identified several areas where FHIR or CDS Hooks lacked necessary semantics or have implementation ambiguity. Primary FHIR challenges included multiple ways to say the same thing, which exacerbated the complexity of variant to allele conversion and lack of representation of deoxyribonucleic acid region(s) studied. Primary CDS Hooks challenges included the complexity of executing an authenticated query against one system (GACS) upon being triggered by a different system (the EHR), and limitations in the types of actionable recommendations that can be returned to the EHR. CONCLUSIONS: In conclusion, we have found that PGx CDS based on FHIR and CDS Hooks appears to represent a promising means of genomics-EHR integration. More real-world testing along with a set of use-case driven GACS interface requirements will push us closer to the U.S. National Human Genome Research Institute vision of a plug-in PGx app.

Health level seven interoperability strategy: big data, incrementally structured.

OBJECTIVES: Describe how the HL7 Clinical Document Architecture (CDA), a foundational standard in US Meaningful Use, contributes to a "big data, incrementally structured" interoperability strategy, whereby data structured incrementally gets large amounts of data flowing faster. We present cases showing how this approach is leveraged for big data analysis. METHODS: To support the assertion that semi-structured narrative in CDA format can be a useful adjunct in an overall big data analytic approach, we present two case studies. The first assesses an organization's ability to generate clinical quality reports using coded data alone vs. coded data supplemented by CDA narrative. The second leverages CDA to construct a network model for referral management, from which additional observations can be gleaned. RESULTS: The first case shows that coded data supplemented by CDA narrative resulted in significant variances in calculated performance scores. In the second case, we found that the constructed network model enables the identification of differences in patient characteristics among different referral work flows. DISCUSSION: The CDA approach goes after data indirectly, by focusing first on the flow of narrative, which is then incrementally structured. A quantitative assessment of whether this approach will lead to a greater flow of data and ultimately a greater flow of structured data vs. other approaches is planned as a future exercise. CONCLUSION: Along with growing adoption of CDA, we are now seeing the big data community explore the standard, particularly given its potential to supply analytic en- gines with volumes of data previously not possible.

The HL7 Clinical Document Architecture.

Many people know of Health Level 7 (HL7) as an organization that creates health care messaging standards. Health Level 7 is also developing standards for the representation of clinical documents (such as discharge summaries and progress notes). These document standards make up the HL7 Clinical Document Architecture (CDA). The HL7 CDA Framework, release 1.0, became an ANSI-approved HL7 standard in November 2000. This article presents the approach and objectives of the CDA, along with a technical overview of the standard. The CDA is a document markup standard that specifies the structure and semantics of clinical documents. A CDA document is a defined and complete information object that can include text, images, sounds, and other multimedia content. The document can be sent inside an HL7 message and can exist independently, outside a transferring message. The first release of the standard has attempted to fill an important gap by addressing common and largely narrative clinical notes. It deliberately leaves out certain advanced and complex semantics, both to foster broad implementation and to give time for these complex semantics to be fleshed out within HL7. Being a part of the emerging HL7 version 3 family of standards, the CDA derives its semantic content from the shared HL7 Reference Information Model and is implemented in Extensible Markup Language. The HL7 mission is to develop standards that enable semantic interoperability across all platforms. The HL7 version 3 family of standards, including the CDA, are moving us closer to the realization of this vision.

The creation of an ontology of clinical document names.

The efficient use of documents from heterogeneous computer systems is hampered by differences in document-naming practices across organizations. Using an open-consensus method, the Document Ontology Task Force, with support from the Veterans Health Administration, addressed this pervasive problem by developing a clinical document ontology. Based on the analysis of over 2000 clinical document names, the ontology was used to formulate a terminology model which is currently being used to guide the creation of fully-specified document names in LOINC (Logical Observations, Identifiers, Names and Codes). Incorporation into LOINC will enable homogeneous management of documents in a widely distributed environment and will also give rise to a rich polyhierarchy of document names.

Kaiser Permanente's "metadata-driven" national clinical intranet.

This paper describes the approach taken to build Kaiser Permanente's national clinical intranet. A primary objective for the site is to facilitate resource discovery, which is enabled by the use of "metadata", or data (fields and field values) that describe the various resources available. Users can perform full text queries and/or fielded searching against the metadata. Metadata serves as the organizing principle of the site--it is used to index documents, sort search results, and structure the site's table of contents. The site's use of metadata--what it is, how it is created, how it is applied to documents, how it is indexed, how it is presented to the user in the search and the search results interface, and how it is used to construct the table of contents for the web site--will be discussed in detail. The result is that KP's national clinical intranet has coupled the power of Internet-like full text search engines with the power of MedLine-like fielded searching in order to maximize search precision and recall. Organizing content on the site in accordance with the metadata promotes overall consistency. Issues currently under investigation include how to better exploit the power of the controlled terminology within the metadata; whether the value gained is worth the cost of collecting metadata; and how automatic classification algorithms might obviate the need for manual document indexing.

The SNOMED RT Procedure Model.

BACKGROUND: SNOMED RT represents a fundamental change from prior versions of SNOMED. The logic-based structure of SNOMED RT enables concepts to be defined more explicitly, providing an opportunity to address inconsistencies and ambiguities present in prior SNOMED concept definitions. OBJECTIVES: Create a unifying organizational strategy for all SNOMED RT procedures, and remove ambiguities in procedure concept definitions. METHODS: A comprehensive model of procedures was developed, based on a set of guiding principles and a review of known existing models. RESULTS: All SNOMED RT procedures are categorized by a common set of "root procedures" (high level atomic actions), and are more explicitly defined by a shared set of defining relationships. CONCLUSIONS: While the objectives have largely been met, open issues continue to be addressed. The similarity between procedure models of SNOMED RT and the U.K. s Clinical Terms Version 3 is proving to greatly facilitate the full integration of the two terminologies into a merged vocabulary to be known as SNOMED Clinical Terms (SNOMED CT), slated to be released in the near future.

An update on HL7's XML-based document representation standards.

Many people know of HL7 as an organization that creates healthcare messaging standards. But HL7 is also developing standards for the representation of clinical documents (such as discharge summaries and consultation notes). These document standards comprise the HL7 Clinical Document Architecture (CDA). Last year we presented a high-level conceptual overview of the CDA. Since that time, CDA has entered HL7's formal ballot process (which when successful will make the CDA an ANSI-approved HL7 standard). This article delves into the technical details of the current CDA proposal. Note that due to space limitations, only a subset of CDA details can be described. Also, because the ballot process elicits considerable feedback, it is likely that the material presented here will undergo evolution prior to becoming a final standard. The most up-to-date information is available on HL7's web site (www.hl7.org).

Impact of attenuation correction on the accuracy of FDG-PET in patients with abdominal tumors: a free-response ROC analysis.

The aim of this study was to evaluate image quality and lesion detectability with and without attenuation correction in patients with abdominal tumors, using a free-response receiver operating characteristic (FROC) methodology. Thirty-four patients with various abdominal tumors were evaluated (11 men, 23 women, median age 48 years). Whole-body emission scans were performed 68 min (35-102 min) after intravenous injection of 4.3 MBq/kg fluorine-18 fluorodeoxyglucose (FDG). Images were reconstructed using the OS-EM algorithm and corrected for attenuation either using postinjection singles transmission (n=27) or by calculation and body outline (n=7). Total scan duration did not exceed 70 min. Studies were read independently by four observers unaware of any clinical data. The uncorrected (UC) images were systematically read before the attenuation-corrected (AC) images. All studies were given an image quality score ranging from 1 (unreadable) to 5 (excellent). Each focus of increased activity was then localized and given a probability of malignancy using a five-point scale. The average image quality score was similar for both UC and AC images. At the time of the positron emission tomography (PET) scans, 127 lesions (63 liver metastases, 9 retroperitoneal lesions, 50 peritoneal or bowel lesions, and 5 pancreatic carcinomas) were revealed by pathological or correlative studies. The areas under the FROC curves were consistently greater for AC images (range 0.8663-0.8867) than for UC images (range 0.7774 -0.8613). Overall, the difference between the AC images and the UC images was significant (P=0.019). In particular, correction for attenuation increased the sensitivity regardless of the location of the lesions. In conclusion, correction for attenuation significantly improves the diagnostic accuracy of FDG-PET for abdominal staging of neoplasms, without impairing the image quality.

A phase II study of two HIV type 1 envelope vaccines, comparing their immunogenicity in populations at risk for acquiring HIV type 1 infection. AIDS Vaccine Evaluation Group.

Several immunogens induce HIV-specific neutralization and in vitro lymphoproliferation in adults at low HIV-1 risk, but responses in persons at high HIV-1 risk are not known. We performed a multicenter, double-blinded, adjuvant-controlled trial with two gp120 vaccines in 296 HIV-1-uninfected volunteers, including 176 reporting higher HIV-1 risk activities. The immunogens were remarkably well tolerated. After three immunizations, 210 of 241 vaccinees (87%) developed neutralizing antibodies, which persisted in 59% after 2 years. The injection drug users receiving SF-2/gp120 had decreased antibody responses relative to the lower risk groups. Envelope-specific lymphoproliferation peaked after two immunizations, and 54% of vaccinees mounted a DTH reaction to gp120 after 4 years. In summary, these immunogens have low adverse reactogenicity and induce durable antibody and T cell responses to the prototype strains. Unexpected differences in antibody responses among diverse HIV-1 risk strata lend support to the conduct of expanded phase II trials in populations other than low-risk volunteers.

HIV vaccines for prevention of infection and disease in humans.

Phase I and II studies have been carried out with several candidate HIV-1 vaccines in seronegative volunteers. Vaccines consisting of rgp 120 stimulated moderate levels of neutralizing antibodies against homologous, TCLA adapted viruses, but did not induce CD8+ CTL responses. Canarypox vectors stimulate CD8+ CTL responses, but little neutralizing activity. The latter can be increased in titer by boosting recipients of canarypox vectors with rgp120 vaccines. Large-scale placebo-controlled efficacy trials are underway with two rgp120 vaccines: AIDSVAX B/B (VaxGen, Inc.) in the United States, and AIDSVAX B/E in Thailand. The canarypox-rgp120 combined regimen has been proposed for study in an intermediate-sized, "test-of-concept" efficacy trial by the NIAID-sponsored HVTN, with an experimental design intended to provide information on the potential in vitro correlates of immunity. The results from these studies, and the methodology used in their conduct, should facilitate the refinement of conventional and novel approaches to the development of safe and effective HIV vaccines.

HL7 document patient record architecture: an XML document architecture based on a shared information model.

The HL7 SGML/XML Special Interest Group is developing the HL7 Document Patient Record Architecture. This draft proposal strives to create a common data architecture for the interoperability of healthcare documents. Key components are that it is under the umbrella of HL7 standards, it is specified in Extensible Markup Language, the semantics are drawn from the HL7 Reference Information Model, and the document specifications form an architecture that, in aggregate, define the semantics and structural constraints necessary for the exchange of clinical documents. The proposal is a work in progress and has not yet been submitted to HL7's formal balloting process.

Validation of an XML-based process to automatically web-enable clinical practice guidelines: experience with the smoking cessation guideline.

OBJECTIVES: To validate the ease by which a Clinical Practice Guideline (CPG) can be web-enabled using an XML-based semi-automated process. DESIGN AND IMPLEMENTATION: An XML DTD for Clinical Practice Guidelines and an MS Word authoring template were created in an earlier project. We took an existing guideline, Bedside Smoking Cessation Intervention, placed it into the MS Word template, converted it into XML, and then to HTML for deployment over the Kaiser Permanent intranet. CONCLUSIONS: We were able to use the MS Word authoring template and automatically generate both an XML representation of our guideline, and an HTML representation, which we have deployed on our intranet. The Bedside Smoking Cessation Intervention guideline was automatically merged into the online guidelines collection. Placing it on our intranet allowed for rapid and easy access by physicians and other health care providers throughout the Kaiser Permanente Medical Care Program.

Advances in data exchange for the clinical laboratory.

The focus of the article is on the nuts and bolts of those standards relevant to the exchange of data between a clinical laboratory and an electronic health record. These include: Health Level 7 (HL7), Logical Observation Identifier Names and Codes (LOINC), Systematized Nomenclature of Human and Veterinary Medicine (SNOMED), and, most recently, the Extensible Markup Language (XML).

Immune responses to human immunodeficiency virus (HIV) type 1 induced by canarypox expressing HIV-1MN gp120, HIV-1SF2 recombinant gp120, or both vaccines in seronegative adults. NIAID AIDS Vaccine Evaluation Group.

A safety and immunogenicity trial was conducted in vaccinia-immune and vaccinia-naive human immunodeficiency virus (HIV)-uninfected adults who were randomized to receive 10(6) or 10(7) TCID50 of canarypox (ALVAC) vector expressing HIV-1MN gp160 or 10(5.5) TCID50 of ALVAC-rabies virus glycoprotein control at 0 and 1 or 2 months and ALVAC-gp160 or 50 microg of HIV-1SF2 recombinant (r) gp120 in microfluidized emulsion at 9 and 12 months; others received rgp120 at 0, 1, 6, and 12 months. All vaccines were well-tolerated. Neither vaccinia-immune status before vaccination nor ALVAC dose affected HIV immune responses. HIV-1MN and HIV-1SF2 neutralizing antibodies were detected more often (100%) in ALVAC-gp160/rgp120 recipients than in recipients of ALVAC-gp160 (<65%) or rgp120 (89%) alone. ALVAC-gp160/rgp120 also elicited more frequent HIV V3-specific and fusion-inhibition antibodies, antibody-dependent cellular cytotoxicity, lymphoproliferation, and cytotoxic CD8+ T cell activity than did either vaccine alone. Trials with ALVAC expressing additional HIV components and rgp120 are underway.

Evaluation of a "lexically assign, logically refine" strategy for semi-automated integration of overlapping terminologies.

OBJECTIVE: To evaluate a "lexically assign, logically refine" (LALR) strategy for merging overlapping healthcare terminologies. This strategy combines description logic classification with lexical techniques that propose initial term definitions. The lexically suggested initial definitions are manually refined by domain experts to yield description logic definitions for each term in the overlapping terminologies of interest. Logic-based techniques are then used to merge defined terms. METHODS: A LALR strategy was applied to 7,763 LOINC and 2,050 SNOMED procedure terms using a common set of defining relationships taken from the LOINC data model. Candidate value restrictions were derived by lexically comparing the procedure's name with other terms contained in the reference SNOMED topography, living organism, function, and chemical axes. These candidate restrictions were reviewed by a domain expert, transformed into terminologic definitions for each of the terms, and then algorithmically classified. RESULTS: The authors successfully defined 5,724 (73%) LOINC and 1,151 (56%) SNOMED procedure terms using a LALR strategy. Algorithmic classification of the defined concepts resulted in an organization mirroring that of the reference hierarchies. The classification techniques appropriately placed more detailed LOINC terms underneath the corresponding SNOMED terms, thus forming a complementary relationship between the LOINC and SNOMED terms. DISCUSSION: LALR is a successful strategy for merging overlapping terminologies in a test case where both terminologies can be defined using the same defining relationships, and where value restrictions can be drawn from a single reference hierarchy. Those concepts not having lexically suggested value restrictions frequently indicate gaps in the reference hierarchy.