American Health Information Community

Personalized Healthcare Workgroup

Summary of the Fifth Web Conference of This Workgroup

Tuesday, May 15, 2007

PURPOSE OF MEETING

The meeting was convened to receive testimony relevant to and to proceed with the work of the subgroups formed at the April meeting. The meeting was called to order by the Co-chairs at 1:06 p.m. Meeting materials and documents referenced below are available at http://www.hhs.gov/healthit/ahic/healthcare/phc_archive.html.

KEY TOPICS

1. Approval of April 17, 2007 Meeting Summary

There were no objections to the acceptance of the summary as distributed.

2. Report of April 24, 2007 AHIC Meeting

Personalized Healthcare (PHC) Workgroup Co-chair Doug Henley reported that the changes to the Workgroup’s broad and specific charges (agreed on at the April 17 meeting) were presented and accepted by the AHIC on April 24. The AHIC also was informed of the formation of the four subgroups.

3. Progress Reports on the Four Subgroups

Kristin Brinner, Office of the National Coordinator staff, reported that each of the subgroups had had one conference call. The Genetic/Genomic Tests and the Family Medical History subgroups are expected to work on use case development recommendations to be presented to the AHIC in July. Dr. Brinner presented PowerPoint slides describing the projected activities of these two subgroups.

Genetic/Genomic Tests The subgroup members discussed the format for recommendations. A Chair has yet to be selected. Immediate issues for use case development are the electronic recording of laboratory data associated with genetic/genomic testing analyses and the inclusion of relevant genetic/genomic test results in the electronic health record (EHR). Complementary activities are expected to include State-mandated newborn metabolic and genetic/genomic screens and voluntary submission of clinical pharmacogenomics data. Longer-term activities are the inclusion of information that describes the analytical validity, clinical validity, and clinical utility of genetic/genomic tests; incentives for the development and evaluation of new genetic/genomic tests and their incorporation into routine clinical practice; and consumer education about the potential benefits and risks associated with genetic/genomic tests.

Family Medical History Marc Williams and Alan Guttmacher reported that immediate issues for use case development are the standardization of family relationship nomenclature and other critical items related to family medical history and the communication of family medical history information between the personal health record (PHR) and the EHR. Longer-term efforts are to include infrastructure and incentives to use PHRs to improve consumer-clinician communication, characterization of the validity and utility of use of family medical history in making clinical decisions, and effects on outcomes.

The PHC Confidentiality, Privacy and Security (CPS) Subgroup and the Clinical Decision Support Subgroup will focus on issues that cross other AHIC workgroups.

CPS Becky Fisher and Amy McGuire reported that they had discussed the work of the subgroup but had yet to convene a call with members. They expect to implement a general approach from two perspectives a top-down perspective on larger issues and a bottom-up perspective on practical issues. They hope to develop a conceptual framework that can include a number of different domains. Global issues that must be addressed include the informed consent process, responsibilities and process for patient education, use of algorithms, validity of information, analysis and interpretation of information, and security.

Clinical Decision Support John Glaser summarized the background of this subgroup. A roadmap was submitted in June 2006. It likely will be the basis for the development of use cases on decision support. Objectives delineated in the roadmap included but were not restricted to the following: improving the clinical adoption and use of clinical decision support interventions, assessing and refining the experience with these technologies, and advancing care-guiding knowledge. In keeping with the Workgroup’s priorities for use case development recommendations on genetic/genomic tests and family history, discussion on clinical decision support will not take place until the fall of 2007.

4. Testimonies on Standards Setting

Robert Dolin gave a brief overview of standards based on his experience at Kaiser Permanente and his participation in various national and international committees, including the Health Level Seven (HL7) board and SNOMED editorial board. Kaiser uses SNOMED and LOINC for its electronic medical record (EMR), LOINC for lab data and SNOMED for diagnoses, orders, microorganisms and most other data. Kaiser developed some additional codes to fill gaps in SNOMED and these codes have been submitted for consideration. Ownership of SNOMED has transferred from the College of American Pathologists to an international standards body, which may have benefits in terms of cost and efforts to harmonize SNOMED and LOINC. A number of European countries use IUPAC rather than LOINC so there are issues with integration from an international perspective. He is involved in the development of a new standard using SNOMED Clinical Terms (CT) and HL7 version 3, which is being written at the level of the implementer. He went on to say that a committee had been formed to work on SNOMED and genetics testing, but he is not a member of that committee and is therefore not able to describe its work.

Next, Kevin S. Hughes, MD, FACS, Co-director, Avon Comprehensive Breast Evaluation Center, Massachusetts General Hospital gave a PowerPoint presentation on HL7 and data standards for family history. HL7 for family history is well-developed and is being adopted for all HL7 version 3 specifications. There are at least eight family history software programs in use at various institutions. The HL7 model provides a standard intermediary and common language by which each of the eight programs can be translated into any of the other programs. Dr. Hughes suggested that all PHRs and EHRs contain the data elements needed to draw pedigrees, run clinical decision support programs, and import and export to HL7. HL7 is available for a small fee.

In response to a question about how a physician would interpret the information from running an algorithm, he said that the models shown in the slides give a percent risk of having a mutation in a gene that can cause either breast cancer in one case or colon cancer in the other case. This information helps the physician to determine whether to refer that patient for genetic testing. Different models show different risks. A 10 percent risk would result in a recommendation for genetic testing. The algorithm helps in assessing risk. The information is not necessarily shared with the patient.

Grant Wood, Senior IT Strategist, Intermountain Healthcare, discussed his PowerPoint presentation on HL7 Clinical Genomics Update. In addition to describing the status of current efforts, he discussed the interactions among three standards bodies the HL7 Clinical Genomics Special Interest Group (SIG), LOINC, and National Cancer Institute (NCI) Ontologies. The goals of the HL7 Clinical Genomic SIG include, for clinical researchers, to integrate clinical laboratory tests, genetic tests and image data; and, for patients, to build systems that coordinate care across physicians and ensure that their genetic and genomic profile is properly informing decision-making.

Maki Moussavi, MS, Cerner Corporation, gave a PowerPoint presentation on Clinical Bioinformatics Ontology (CBO), a semantically structured controlled vocabulary for clinical molecular diagnostics. CBO addresses the need for consistent descriptions and representations of molecular biological and cytogenetic entities currently performed in clinical diagnostics in a standardized and machine readable format. CBO focuses on gaps in the existing vocabulary of, for instance, SNOMED CT and LOINC to meet the needs of the molecular and cytogenetics world to transmit data electronically, to store it in the EHR, and potentially to implement clinical decision support. SNOMED is a very good phenotypic resource for describing the problem that has been detected and LOINC attempts to describe the procedure itself and in some cases the clinical result. The CBO is an open source tool that can be accessed from the Cerner Web site. It is included within the NCI meta-thesaurus as well.

The Cerner Controlled Medical Terminology (CMT) is a meta-vocabulary associating most of the common medical vocabularies, including SNOMED and LOINC. CMT enables integration of concept-based capabilities into the Cerner Millennium suite of clinical software.

5. Testimonies on Complementary Activities

Mary Beth Bigley, Office of the U.S. Surgeon General, reported on forthcoming efforts with the U.S. Department of Veterans Affairs (VA) and the U.S. Department of Defense (DoD) on family history. She plans to meet with DoD and VA separately within the next three weeks about the family history tools that are incorporated into their respective EMRs. The Surgeon General wants to know how they are using their family history tools and how they may be transferable to other systems. As military personnel leave the system, they should be able to transfer their family history records to the VA and other healthcare systems. Dr. Bigley will attempt to form a DoD, VA, and U.S. Department of Health and Human Services (HHS) workgroup to adopt an interoperable family history tool.

Dr. Brinner pointed out that conversations with the Surgeon General have focused on the possibility of My Family Health Portrait as something that could be used across agencies. It will not be helpful to have these different agencies using different family history tools.

Felix Frueh, Associate Director for Genomics at the U.S. Food and Drug Administration (FDA) and Chair of the Interdisciplinary Pharmacogenetic Review Group the FDA-wide group responsible for the review of voluntary genomic data submissions summarized the program. There are three important aspects of the program. One is the submission itself, and the standards and formats that are used to submit such data to the FDA. The second is the data analysis that is associated with the review of the submissions and what is learned from it. Third is the concept of the submissions, the meetings and interactions to discuss the information. Having an informal data exchange platform has been very helpful to the FDA in ensuring that a framework is being developed for making regulatory decisions.

He distributed a paper that he co-authored with FDA colleagues titled, The Experience with Voluntary Genomic Data Submissions at the FDA and a Vision for the Future of the Voluntary Data Submission Program. As a result of several workshops and public comment opportunities, the FDA released the final Guidance for Industry: Pharmacogenomic Data Submissions in March 2005 and created a Web portal that provides up-to-date regulatory and background information on genomics. FDA officials believe the program has been instrumental in creating a review infrastructure for genomic data and has contributed to policy development. The program has provided a way to interact with industry focusing on the scientific rather than the regulatory interpretation of the results presented.

Next, Mike Watson, PhD, American College of Medical Genetics (ACMG), described ongoing efforts in the area of newborn screening, which he characterized as an area in dire need of interoperability standards. State law regulates newborn screening, and the requirements vary across States. States collect short-term follow-up information but generally do not have outcome data. ACMG was funded by HRSA to develop action sheets for primary care pediatricians to guide them in the case of a positive screen (http://www.acmg.net/resources/policies/ACT/condition-analyte-links.htm). The act sheets delineate steps to take from the receipt of a positive screen to a confirmed diagnosis. The act sheets were designed to be used with EMRs. For instance, they could be incorporated as pop-ups and would be applicable for clinical decision support. Intermountain Healthcare has incorporated the act sheets into its EMRs. The project is expanding rapidly from the 31 conditions for which act sheets have been completed. The National Institute of Child Health and Human Development is involved and interested in the National Newborn Screening Translational Research Network. The project will eventually result in outcome data that are reported back to the States.

The regional collaboratives also will try to collect data on those conditions that are under consideration for newborn screening to establish a better evidence base on which to make decisions. All of this requires standardized language formats for the laboratory information, not just the screening laboratory information but also the clinical diagnostic laboratory information and the clinical evaluation information. Building natural histories via these data collection efforts may eventually reduce the cost of clinical trials.

6. Next Steps

The next meeting of the Workgroup is scheduled for June 7, 2007. The agenda will focus on recommendations for submission to the AHIC on July 31. Recommendations are expected from two subgroups use case development for genetic/genomic tests and family medical history.

7. Public Comments

Mollie Ullman-Cullere, Harvard Partners Center for Genetics and Genomics and Co-chair of the HL7 Genomics Group, asked to augment information provided during the meeting. Work is currently under way for the new version 3 formatting of HL7 as well as for version 2, which is more broadly adopted in the healthcare arena. The version 2 collaboration is among Intermountain Healthcare, Partners HealthCare, and the National Institutes of Health National Library of Medicine. The work includes LOINC coding for both the genetic ordering of the tests and the structured results of the test. The HL7 version 3 is going on the ballot this summer and fall.

Partners HealthCare has implemented an end-to-end solution for genetic testing results with clinical decision support. The work with HL7, LOINC, and NCI will enable outside testing laboratories using software unrelated to any type of vendor product or a particular vendor solution to add data into EMRs independent of software constraints.

8. Adjournment

Meeting adjourned at 2:52 p.m.

SUMMARY OF ACTION ITEMS

No action was taken.

MEETING MATERIALS

Agenda

Subgroup Activities:

Brinner - Genetic/Genomic Test Priority Area

Williams/Guttmacher - Family Medical History

Immediate Activities:

Hughes - Data Standards for Family History

Hughes - Specification of Family History HL7 XML File

Moussavi - Clinical Bioinformatics Ontology

Williams/Wood HL7 Clinical Genomics Update

Complementary Activities:

Frueh/Mansfield - Voluntary Genomic Data Submissions

Watson - Newborn Screening Collaborative

Personalized Healthcare Workgroup

Members and Designees Participating in the Web Conference

Co-chairs

John Glaser Partners HealthCare Systems

Douglas Henley American Academy of Family Physicians

Members
Paul Cusenza	23andMe
Andrea Ferreira-Gonzalez	Association for Molecular Pathology
Becky Fisher	Patient Advocate
Felix Frueh	HHS/Food and Drug Administration
Alan Guttmacher	HHS/National Institutes of Health/National Human Genome Research Institute
Betsy Humphreys	HHS/National Institutes of Health/National Library of Medicine
Stephen Matteson	Pfizer Global Research and Development
Deven McGraw	National Partnership for Women and Families
Amy McGuire	Baylor College of Medicine
Steve Teutsch	Merck & Co., Inc.
Janet Warrington	Affymetrix
Marc S. Williams	Intermountain Healthcare
Office of the National Coordinator for Health Information Technology Staff
Kristin Brinner
Gregory Downing
Senior Advisors
Mary Beth Bigley	Office of the Surgeon General
Armando Oliva	HHS/Food and Drug Administration
Lisa Rovin	HHS/Food and Drug Administration
Other
Maren Scheuner	RAND Corporation

Disclaimer: The views expressed in written conference materials or publications and by speakers and moderators at HHS-sponsored conferences do not necessarily reflect the official policies of HHS; nor does mention of trade names, commercial practices, or organizations imply endorsement by the U.S. Government.