Draft Preliminary Recommendations Precision Medicine Task Force September 18, 2015 Leslie Kelly Hall, Co-Chair Jon White, Co-Chair

Agenda Opening Remarks, Work Plan Review NIH ACD Update HL7 Genomics Workgroup Review and Finalize Recommendations Next Steps 1

Precision Medicine Task Force Membership MemberOrganization Co-Chairs Leslie Kelly HallHealthwise Jon WhiteONC / Agency for Healthcare Research and Quality (AHRQ) Members Mary BartonNational Committee for Quality Assurance (NCQA) Lisa GallagherHealthcare Information and Management Systems Society (HIMSS) David McCallie, Jr.Cerner Corporation Andrey OstrovskyCare at Hand Eric RoseIntelligent Medical Objects Andrew WiesenthalDeloitte Consulting, LLP Federal Ex Officio James BreelingVeterans Health Administration (VHA) Josh DennyNational Institutes of Health (NIH) Christina HeideHHS Office for Civil Rights (OCR) Betsy HumphreysNational Library of Medicine (NLM) Mitra RoccaFood and Drug Administration (FDA) Invited Guests Mina HsiangUnited States Digital Service (USDS) / Office of Management and Budget (OMB) Claudia WilliamsWhite House Office of Science & Technology Policy ONC Staff Maya UppaluruONC – Federal Staff Lead Debbie BucciONC - Technical Advisor 2

Precision Medicine Initiative Mission Statement To enable a new era of medicine through research, technology, and policies that empower patients, researchers, and providers to work together toward development of individualized treatments. 3

Task Force Charge Identify opportunities for innovative collaboration around pilots and testing of standards that support health IT interoperability for precision medicine Recommend existing standards that are currently ready to support PMI Identify emerging standards and reference implementations that may require further pilot testing in order to support PMI Identify gaps in available data standards related to PMI 4

MeetingsTask July 17, :00 am - 1:30 pm ET Kick-off Review charge, work plan Overview of the Precision Medicine Initiative - White House Office of Science & Technology Policy & National Institutes of Health Wednesday Jul 29, :30 - 3:00 pm ET Presentations from experts - 23andMe; NIH Precision Medicine Workshop; Institute of Medicine Genomics Roundtable Wednesday Aug 5, :00 - 1:30 pm ET Presentations from experts - Intel Corporation; Intermountain Healthcare; National Library of Medicine Wednesday Aug 19, :00 - 1:30 pm ET Presentations from experts - Duke Me Tree Project; eMerge Network; New York Genome Center; Sage Bionetworks Monday Aug 31, :00 - 1:30 pm ET Develop Preliminary Task Force Recommendations – Broad Institute Thursday Sept 10, :30 - 3:00 pm ET Revise Task Force Recommendations Friday Sept 18, :00-11:30 NIH ACD Update HL7 Genomics Workgroup Finalize Recommendations to HITSC September 22, 2015 – HITSC Meeting 10:00 am – 12:30 pm ET Present final recommendations Precision Medicine Task Force Workplan 5

NIH ACD Update 6

NIH ACD Recommendations Hybrid model of centralized and federated data – Coordinating Center will centrally store a curated, analysis-ready, core data set on all participants using common data models, and serve as a hub for accessing data and biospecimens. – Hybrid data and analytics architecture will leverage both centralized data storage of a growing amount of core data, as well as federated access to additional data at the nodes across the network Common, growing core data set – Collection of diverse set of data types: EHRs, claims, surveys, baselines health exams, mobile health, biologic investigations – Development of standardized and automated mechanisms to acquire clinical data from healthcare provider organizations (HPOs) – Rigorous data curation for analysis-ready data sets 7

NIH ACD Recommendations: Acquiring research data for the PMI cohort From participating healthcare provider organizations (HPOs) – A longitudinal record of care (e.g., academic medical centers, Federally Qualified Health Center (FQHCs), vertically integrated private health care organizations, and vertically integrated governmental organizations) – Ability to consent volunteers to the PMI cohort, to share EHR data, and to collect new biospecimens – [Those providers that] can contribute significantly to the size of the PMI cohort should be preferred. From individuals – Must be living in the U.S. or U.S. Territory and be able to visit a U.S. health provider – Provide certain amount of core data, including baseline phenotype assessment and data quality – Must be recontactable, provide a biospecimen, agree to share EHR data 8

NIH ACD Recommendations: Security & Privacy Ensure appropriate safeguards Use of de-identified data whenever feasible, provide appropriate data protections Discourage data from being copied outside the secure computing environment Clearly articulated breach response and notification plan NIH seeks to fill privacy gaps such as – FOIA exemption for genomic data – Unauthorized re-identification – Require data users to secure a certificate of confidentiality 9

NIH ACD Recommendations Core data set – Standard self-report measures via direct patient assessment – Brief, standardized baseline health exam at enrollment – Structured clinical data ICD & CPT codes with dates High value clinical lab results in structured format (e.g., eMERGE, PCORnet, Sentinel, MPOG). Measure and reference ranges. Medication data (e.g., start/stop date, inpatient or outpatient, dose, route, frequency, strength) Vital measurements Record of all encounters Health plan data (enrollment and disenrollment dates, type of coverage) – Biospecimen-derived data – Mobile health data Clinical data from direct volunteers – Ability to submit EHR data that to PMI cohort through existing data transfer protocols (Blue Button, patient portals & VDT) 10

HL7 Clinical Genomics Workgroup 11

HL7 Clinical Genomics Work Group Mission Create and promote standards by enabling the communication between…parties of the clinical and personalized genomic data Focus: Personalization of the genomic data and the linking to relevant clinical information. Facilitate the development of common standards for clinical research information management across a variety of organizations, including: – National and international government agencies – Regulatory bodies, private research efforts, and sponsored research Facilitate availability of safe and effective therapies by improving the processes and efficiencies associated with regulated clinical research 12

HL7 Clinical Genomics Work Group Leadership Co-Chairs – Gil Alterovitz, Boston Children's Hospital – Siew Lam MD MSc, Intermountain Healthcare – Bob Milius PhD, National Marrow Donor Program – Amnon Shabo PhD, Philips Healthcare – Mollie Ullman-Cullere, Partners HealthCare System, Inc. Facilitators – Amnon Shabo PhD, Philips Healthcare – Grant Wood, Intermountain Healthcare – Joyce Hernandez 13

Draft Recommendations 14

Recommendations Background - EHR likely to capture more phenotypic data from MDs and patient – Phenotypic data are collected already such as problems, medications, allergies, etc. – Core problem: Don't have a standard data model for EMR and categorical standard responses for many basic types of information Recommendations - Advance filling gaps related to key missing components with a short term focus on supporting the PMI cohorts. Gaps include: – Transport mechanisms (for transmission among EHR, lab and individual) – Data structure – Value sets Standards and Recommendations were placed into four categories: 1.Readily Applicable Standards for PMI (Green) – can be put to use to support the cohorts 2.Promising Standards for PMI (Yellow) – may require additional effort to bring to use 3.Standards Gaps for PMI (Red) – areas where considerable work is needed 4.Accelerators (Blue) – opportunities to advance / improve standards Recommended Actions to Advance need to be assigned to each standard, emerging standard, recommendation, etc. – Short-term versus long-term priorities 15

Readily Applicable Standards for PMI RecommendationActions to Advance Standards currently referenced in certification rule (CCDA, Direct for transport, etc.) HL7 Family Health History/PedigreeOpportunities to further develop? SNOMED for capturing family health history (need to name specific aspects of SNOMED that would be useful) 1 ** Open ID Connect & OAuth should be considered as means of access delegation and sign on to promote interoperability of precision medicine data 2 Standards for genetic data going from lab to EHR (need to name specific standards) Key: Actions to Advance 1 – Form Task Force to advance 2 - Apply accelerators (e.g., S&I Initiative, pilot project, policy guidance) to existing standards by ONC 3 - Follow existing standards process 16 1) Precision Medicine Task Force: 2) Transport & Security Standards Workgroup:

Promising Standards for PMI RecommendationActions to Advance FHIR could be included as an emerging standard 4 ; use of FHIR to send data from EHR to researchers Build off current work on SMART on FHIR Genomics 1 FHIR genomic concept transports biomarkers but needs to specify a value set Human Phenome Ontology (HPO) 2 Computable Patient Consent - standards exist but lack adequate implementation guidance 3 Include more complete authorization standards (e.g., IHE XUA, IUA, etc.); ensure authorization standards are compatible across disparate networks 5 IOM Genomics Roundtable, GA4GH work Key: Actions to Advance 1 – Form Task Force to advance 2 - Apply accelerators (e.g., S&I Initiative, pilot project, policy guidance) to existing standards by ONC 3 - Follow existing standards process 17 1) Precision Medicine Task Force: 2) Precision Medicine Task Force: 3) Interoperablity Standards Advisory Task Force - page 17http://healthit.gov/FACAS/sites/faca/files/HITSC_ISATF_Recommendation_Slides_ pdf 4) Ibid page 19 5) Ibid Page 19

Standards Gaps for PMI RecommendationActions to Advance ONC should convene a stakeholder group to address computable patient consent; there exist standards but without clear implementation guidance 1 Race and Ethnicity: OMB Standard may be suitable for some purposes but inadequate for precision medicine and directing therapy or clinical decisions 2 1 – Form Task Force ONC should work with stakeholders to define what is the minimum data set and/or means required to make precision medicine data useful in an EMR and in a clinical setting 3 1 – Form Task Force Microbiome data standards 4 Key: Actions to Advance 1 – Form Task Force to advance 2 - Apply accelerators (e.g., S&I Initiative, pilot project, policy guidance) to existing standards by ONC 3 - Follow existing standards process 18 1) Interoperablity Standards Advisory Task Force – Page ) Ibid. page 15 3) Precision Medicine Task Force: 4) Precision Medicine Task Force:

Accelerators RecommendationActions to Advance Additional ONC investment in pilots of FHIR for research/individual data donation use case? Incorporation of HPO in the UMLS Metathesaurus and connections between HPO and SNOMED CT 2 OMIM: Codes for phenotypes, genotypes and links between the two HUGO: De facto standard for naming of genes dpSNP and ClinVar: Opportunity to develop a service that would consumer data from these sources and synthesize so it’s digestible for a clinical information system Key: Actions to Advance 1 – Form Task Force to advance 2 - Apply accelerators (e.g., S&I Initiative, pilot project, policy guidance) to existing standards by ONC 3 - Follow existing standards process 19 1) Interoperability Standards Advisory Task Force – Page )

Next Steps 20

Appendix 21

Focus and Questions for Presenters 22 Focus The exchange of genomic and phenomic data among: – Patients/Participants – EHRs – Researchers – Testing labs (for both research and clinical care) Questions for Presenters What is the key problem or set of problems your organization is attempting to solve? What is the minimum interoperable data set of genome and phenome data for these data exchanges? Are there standards that can support this movement today? What gaps are there and what is needed in the future?