Skip Navigation

American Health Information Community

Personalized Health Care Workgroup

Summary of the Tenth Web Conference of This Workgroup

Monday, November 26, 2007


PURPOSE OF MEETING

The meeting was convened to receive updates on the efforts of subgroups and staff, and to consider testimony on the implications of pharmacogenomic tests for electronic health records (EHRs). The meeting was called to order by the Co-chair John Glaser at 12:09 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 October 25, 2007 Meeting Summary (Doc 1)

The Workgroup approved the meeting summary as distributed.

  1. Prioritization of Personalized Health Care (PHC) Workgroup Needs for 2009 Use Cases (Doc 2)

Kristin Brinner and Lauren Kim, Office of the National Coordinator (ONC) for Health Information Technology, distributed a document delineating the use cases across all workgroups; this was an updated version of the document discussed at the October meeting. The revised version shows the decisions on priorities made at the previous meeting and notes the interconnecting needs and overlaps across the workgroups. For example, the newborn screening use case has implications for the registries being considered by the Population Health and Clinical Care Connections Workgroup. Clinical Decision Support (CDS) is a priority for several groups in addition to the PHC Workgroup, making it a high priority overall. Staff invited members to review the listings and submit comments by email. Staff will continue to consolidate the listings and prepare a comprehensive list to submit to ONC for approval. The final list will be available late January or early February 2008.

  1. Discussion of PHC Workgroup JAMIA Submission (Doc 3)

Kristin Brinner presented a draft article, “Advancing Personalized Health Care through Health Information Technology: An Update from the American Health Information Community’s Personalized Health Care Workgroup,” for submission to the Journal of American Medical Informatics Association. Dr. Brinner reported that many are interested in the products of the PHC Workgroup; the Journal editors are considering publishing an article. Staff and a contractor prepared the draft based on the reports and recommendations produced by the group. Feedback suggested that the article focus on the products rather than the history and process of the Workgroup. The Journal editors may be interested in additional articles on, for example, family health history and Confidentiality, Privacy, and Security (CPS). Perhaps the article should mention that several companion papers are being prepared for publication. Dr. Brinner requested feedback from members.

Members discussed the relative merits of providing more background on the Workgroup versus describing its products. It was noted that other journals may desire publishing articles on the Workgroup’s efforts.

Greg Downing announced that after consultation with the Workgroup co-chairs, it was agreed that all members of the Workgroup will be entitled to authorship inclusion. The final draft will be circulated for approval prior to submission.

Action item #1: Staff will receive comments and continue to prepare the paper for publication.

  1. Update on Newborn Screening Subgroup Activities

Michele Lloyd-Puryear, Health Resources and Services Administration (HRSA), reported that the Subgroup has completed two conference calls. Members divided into three sub-subgroups: one will focus on the entities and codes that are necessary for communicating lab results; the second will deal with newborn hearing screening; and the third will address the mutational analysis of genetic evaluation. The Subgroup agreed to produce a use case for presentation to the Workgroup in January. The Subgroup’s work will show how population information is integrated into personal health information to support primary care decision making.

The Subgroup delineated 13 areas for potential development and prioritized the first four for inputs in the use case. A primary consideration is that data elements are quantitative and interchangeable with the lab results. The group is focusing on the information technology (IT) implications of newborn screening. Staff noted that the Subgroup is on track.

  1. Pharmacogenomic Tests and Personalized Health Care: Implications for the Electronic Health Records and Health Information Exchange (Doc 4)

Greg Downing, ONC staff, offered context for the presentations. The science of pharmacogenomic medicine is progressing, but mainstream clinical application lags behind. To date, the Workgroup’s efforts have focused primarily on single gene type abnormalities associated with inherited forms of mutations acquired in cancer processes or biological variations that describe drug metabolism. The Workgroup has considered other types of genetic tests being developed for medical applications, such as multiple gene expression patterns. The Workgroup has not yet discussed index assays or other genomic analysis.

Dr. Downing pointed out that the purpose of the work on pharmacogenomic tests and PHC is not to facilitate the application of research findings in the clinical setting, but rather to support consistent standards in order that information can be shared as it is available. Goals are to:

He concluded by pointing to areas that the Workgroup may wish to consider in the future:

  1. Clinical Genomics Interoperability Using CDISC-HL7 Standards (Doc 5)

Becky Kush, Clinical Data Interchange Standards Consortium (CDISC), and Joyce Hernandez, Merck & Co., Inc., made the next PowerPoint presentation. Dr. Kush began with background information on clinical research standards and their application in health care. Information does not flow easily across the research, the Food and Drug Administration (FDA), and health care delivery environments. There is a great deal of repetitious data entry from the site of the clinical trial to the FDA. Reporting frequently interferes with work flow. Adverse events are not consistently reported because of the time and effort involved. 60% of clinical trials record data on paper. When data are collected electronically, 10 to 12 work stations or lap tops may be involved. Clinicians are often reluctant to be involved in trials due to the effort required and interference with work flow. To eliminate these barriers, we need to be able to collect, integrate and analyze information from multiple sites in a standard format, which is what the CDISC is working toward. The requirements of medical research should be taken into account now in conjunction with Health Information Technology (HIT) initiatives.

CDISC has established worldwide industry standards to support the electronic acquisition,

exchange, submission, and archiving of clinical research data and metadata to improve data quality and streamline medical and biopharmaceutical product development and research processes. Use of the standards facilitates compliance with the FDA guidance. The standards (http://www.cdisc.org/) are harmonized with Health Level Seven (HL7) through Biomedical Research Integrated Domain Group (BRIDG).

Dr. Hernandez showed the results of CDISC’s work with the HL7 Clinical Genomics Special Interest Group to integrate pharmacogenomics data with clinical information. Integration can help with, for instance, decisions about medication when genetic mutations indicate the risk of an adverse event. Once the clinical genomic data are captured in the EHR, the information can be accessed through a clinical practice view or a research view, which gives genotypic information through what are known as clinical research plans.

Next, Mollie Ullman, Harvard Partners, explained the methods used in developing the HL7 clinical genomic standard. She pointed out that within the clinical realm the concern is not only with the visual representation of the data in the health record, but also with the machine readable representation. The machine readable representation is encapsulated and then either exposed or bubbled up via CDS within the appropriate clinical context and guidelines. Ms. Ullman went on to describe several implementations, which are summarized as follows:

Lessons learned from these projects will be described in an implementation guide that will be available via HL7 for comment in December. Dr. Kush concluded the presentation by describing the CDISC Healthcare Link, a joint project with Duke Clinical Research Institute, Novartis, Merck, Johnson & Johnson, and Microsoft. The project demonstrated clinical information interoperability between EHR and pharmaceutical clinical trials systems based on open standards. CDISC is using standards to work toward an environment in which data are collected once from various sources and used for multiple downstream purposes, including registries and traditional review boards, contract research organizations, partner companies, regulatory authorities, and clinical care. Dr. Downing acknowledged that the use cases developed under this project had helped to inform the Community use cases.

  1. PharmGKB (Doc 6)

In introducing the presentation, Rochelle Long, National Institutes of Health (NIH), said that the purpose of the project was to develop standards for linking phenotype and genotype for research. Michelle Whirl Carrillo, Stanford University, described the pharmacogenomics and pharmacogenetics knowledge database. Dr. Carrillo began by saying that she was describing a basic research knowledge base, which is somewhat different from other areas that the Workgroup has explored. NIH created in 2000 the Pharmacogenetics Research Network to investigate how genetic variation impacts drug response. PharmGKB integrates, aggregates, and annotates pharmacogenomics data and knowledge (visit http://www.pharmgkb.org/). PharmGKB accepts research data submissions from scientific researchers in pharmacogenetics and pharmacogenomics. Data are de-identified. Several standard formats have been established for submitting the data. Users must be registered. PharmGKB does not interact directly with EHRs nor use LOINC or HL7. Dr. Carrillo acknowledged the potential usefulness of standards, but pointed out that the research community uses a variety of terminologies.

In response to a question, Dr. Carrillo described the efforts of the Warfarin Consortium, a group of researchers interested in the effects of pharmacogenomics on Warfarin. PharmGKB volunteered to collect phenotype and genotype data from the participants and to try to standardize the disparate data sets. They are currently working on standardization of the phenotype data. The genomic data are more straightforward because only a few different genes are involved and the nomenclature for those genes is well accepted. The researchers identified the elements that required standardization.

Q&A

In response to a question about the development of a template to use as standard practice, Dr. Carrillo said that the merging of data sets that she at first envisioned was not occurring. The capacity to collect large data sets in a single study with a single sample has increased; consequently, much more analysis can be done with that one data set. Researchers are not inclined to agree on protocols and standards; there is much more discussion of privacy, ethics, and access than standards.

Another participant asked if EHRs were available for research use, could they be used. The usefulness of the data contained in the EHR would depend upon the level of detail required by the research protocol. In deciding what level of detail to include in the EHR, one is faced with anticipating research interests and including everything to answer very specific questions versus including information on who performed the test and the contact for acquiring more information.

Marc Williams suggested that consideration be given to requiring NIH grant applicants to agree to use specific formats or standards in submitting findings. Joyce Mitchell pointed out that lab reports often do not include the sequencing data; hence they are not incorporated into the EHR. The provider must order another test if at a later time she needs sequencing information. Another participant pointed out that information on the quality of the data may be absent as well. Ronald Farcus noted that different tests for the same gene give somewhat different information; tests have certain inherent biases even being affected by the subject’s ethnicity. Tests that are used in clinical care may not be adequate for research purposes.

Work is being done to develop an EHR profile that would allow an EHR to be certified for clinical research purposes. Another approach is the study data tabulation model; however, health care organizations typically have limited IT resources and reformatting data becomes burdensome. Another barrier is the lack of common vocabularies.

Action item #2: Staff and the co-chairs will discuss the presentations to identify possible aspects of data interchange between the research and clinical care environments for the Workgroup to consider.

  1. Clinical Decision Support Vignettes (Doc 7)

John Glaser, Co-chair, reported on the status of the Ad Hoc CDS Planning Group, which is scheduled to meet in January to agree on recommendations. The recommendations will be presented to the Workgroup for action prior to submission to the Community. Each workgroup has been asked to submit vignettes for CDS use case development. The Ad Hoc CDS Planning Group will examine the vignettes for common threads and recommendations. Insofar as there were few comments, Dr. Glaser suggested that members review the vignettes (Doc 7) and submit comments and/or recommendations to staff.

In addition to the three vignettes submitted to date (breast/ovarian cancer, pharmacogenomics, and biosurveillance/screening), Marc Willams and Joyce Mitchell offered to develop one on colorectal screening and Michele Lloyd-Puryear agreed to work on newborn screening. It was suggested that one of the vignettes include detection of an allergy.

Action Item #3: Marc Willams and Joyce Mitchell will draft a CDS vignette on colorectal screening. Michele Lloyd-Puryear will draft a CDS vignette on newborn screening.

  1. Next Steps

The next meeting of the Workgroup is scheduled for Wednesday, January 30, 2008, from 1 to 4 p.m. Staff will provide an update on a paper on CPS that is being prepared for submission for publication. The PHC draft detailed use case, which is expected to be published for public comment in late December or early January, will be presented to the Workgroup for comments. Staff will bring two topics to the Workgroup for its future consideration. One topic is standards for population genomics; the National Human Genome Research Institute (NHGRI) has awarded a $7million grant to the Research Triangle Institute to develop standards for large genomewide association and epidemiological studies. The second is the role of genomics in transplantation; CDISC, Inc. is working on a family model based on genetics and genomics for tissue typing to better understand donor-recipient matches.

  1. Public Comments

None.

  1. Adjournment

Meeting adjourned at 2:10 p.m.

SUMMARY OF DECISIONS AND ACTION ITEMS

Decision: None

Action item #1: Staff will receive comments and continue to prepare the paper for JAMIA publication.

Action item #2: Staff and the co-chairs will discuss the presentations for identify possible aspects of data interchange between the research and clinical care environments for the Workgroup to consider.

Action Item #3: Marc Willams and Joyce Mitchell will draft a CDS vignette on colorectal screening. Michele Lloyd-Puryear will draft a CDS vignette on newborn screening.

Meeting Materials

Agenda

DRAFT Meeting Summary (October 25, 2007)

PHC Use Case Priorities

PHC Workgroup JAMIA Submission Draft

Pharmacogenomics Overview

CDISC-HL7 Presentation

PHC Scenarios for Clinical Decision Support

CDISC IHE Flyers

Clinical Genomics Data Standards for PGx

PharmGKB Article

Integrating Large-Scale Genotype and Phenotype Data

Updated PHC Newborn Screening Subgroup Roster

Personalized Healthcare Workgroup Members and Designees

Participating in the Web Conference

Co-chairs
John Glaser Partners HealthCare Systems
   
Office of the National Coordinator for Health Information Technology Staff
Kristin Brinner  
Gregory Downing  
Lauren Kim  
Scott Boyle  
Alan Zuckerman  
   
Members
Beryl Crossley American Clinical Laboratory Association, Quest
Paul Cusenza Consultant
Andrea Ferreira-Gonzalez SACGHS, AMP President & Virginia Commonwealth University
Rebecca Fisher Patient Advocate
Emory Fry Department of Defense
Alan Guttmacher NIH/NHGRI
Kathy Hudson Genetics and Public Policy Center
Betsy Humphreys HHS/National Library of Medicine
Amy McGuire Baylor College of Medicine
Steve Teutsch Merck & Co., Inc.
Janet Warrington Affymetrix
Dennis Williams HHS/HRSA
Marc Williams Intermountain Healthcare
   
Senior Advisors:
Greg Feero NIH/NHGRI
Katherine Kolor for Muin Khoury HHS/CDC
Michele Lloyd-Puryear HHS/HRSA
Dina Paltoo NIH
Ron Przygodzki Department of Veterans Affairs
Lisa Rovin FDA
Mollie Ullman-Cullere Harvard Partners Center for Genetics & Genomics
   
Others, Including Presenters:
Michelle Carrillo PharmGKB
Mark Carroll Indian Health Service
Ronald Farkas FDA
Edward Helton CDISC
Joyce Hernandez Merck & Co., Inc.
Terri Klein CDISC
Rebecca Kush CDISC
David Lobach Duke University
Rochelle Long NIH
Joyce Mitchell University of Utah
Philip Pochon CDISC
Grant Wood Intermountain Healthcare

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.