AMIA 2020 Informatics Summit Panels

Monday, March 23

S01: Panel – Recruitment Innovation Center: Assessing Institutional Capabilities and Appropriate EHR Recruitment Strategies

P. Embi, J. Harper, Regenstrief Institute, Inc., Indiana University ; D. Hood, Regenstrief Institute, Inc.; U. Tachinardi, Regenstrief Institute, Inc., Indiana University

The ability to support a learning health care system at a national level requires reliable data to inform policy and clinical practice decisions. Controlled intervention trials, which depend on effective recruitment and enrollment, continue to be one of the most recognized sources of data for this effort. The Recruitment Innovation Center (RIC) is developing recruitment and retention strategies to improve the quality of clinical trials, raise awareness of the value of research and increase trial enrollment and health outcomes across America. During the past 12 months the RIC has enabled innovative work by developing methods to support sites who would like to leverage their local electronic health record (EHR), their EHR data, and other informatics tools for recruitment and retention in clinical trials. This panel will examine three interrelated activities and how they can positively impact the challenges of recruitment and retention: EHR driven recruitment strategies; 2) Governance and maturity models for implementation and EHR recruitment; 3) Implementing recruitment strategies in a Cerner/Epic environment.

Tuesday, March 24

S06: Panel – Building Longitudinal Records for Research while Preserving Patient Privacy

A. Gregorowicz, The MITRE Corporation; T. Ong, University of Colorado; T. Carton, Louisiana Public Health Institute

Conducting patient centered outcomes research typically involves a process of assembling information on the individuals being studied. When answering research questions, there is often the need to build a longitudinal record which includes a comprehensive clinical summary of the patient. Generating a longitudinal record requires gathering all of the relevant information on an individual may require the integration of data across multiple information systems and organizational boundaries. To successfully gather this information, researchers must be able to correctly identify patients across organizations and maintain their privacy in the process. This panel will discuss the approaches taken by the Childhood Obesity Data Initiative (CODI) to address this issue. First, the panel will discuss the use of Privacy Preserving Record Linkage (PPRL) techniques. This allows the linkage of patient records across organizational boundaries without disclosing personally identifiable information (PII) between the parties. This will include a brief primer on fuzzy matching with PPRL. Then the panel will review how these concepts have been applied to queries on PCORnet, the infrastructure leveraged by CODI. In closing will be a discussion and question and answer session on how CODI is building longitudinal records across organizations to answer research questions related to childhood obesity interventions.

S11: Panel – Addressing Competing Interests in Engaging with the Learning Healthcare System within an Academic Institution

B. Goldstein, Duke University; D. Dorr, Oregon Health & Science University; C. Lindsell, Vanderbilt University; D. Mann, New York University; E. Hinz, Duke University

With the advent of the modern electronic health record (EHR) system the dream of learning healthcare system (LHS) – one where lessons learned from clinical encounters are integrated to improve clinical care – has been in reach. However truly realizing a learning healthcare environment requires the cooperation and coordination of experts with diverse skill sets. Moreover, since most of these efforts occur at academic medical centers, most of the work is being conducted by academics. This raises important questions of how best to integrate this work into one’s academic research agenda.

In this panel, we will convene a diverse panel of experts who are actively engaged in developing a LHS at their local institutions. Each panelist has a different academic background and plays a different role in helping to realize a LHS at their local institution. We will cover the various needs of a LHS - overall structure, clinical integration, quantitative expertise, & research dissemination - and some of the implicit challenges within each.

S15: Panel – Improving Access to Interoperable Data for Research

K. Chaney, T. Zayas Caban, Office of the National Coordinator for Health IT; P. White, Salt Lake City Veterans Affairs Medical Center

Widespread adoption of electronic health record (EHR) systems and consumer electronics has resulted in large volumes of health-related data potentially available for research. However, realizing the value of these data for research has been slow due to challenges in both the data and the health information technology (IT) infrastructure that supports it. This panel will discuss efforts by the Office of the National Coordinator for Health Information Technology (ONC) to guide the development of a future health IT infrastructure that supports use of electronic health data for research. This work addresses three objectives: (1) articulate a vision for an ideal health information ecosystem that supports research; (2) identify stakeholders’ priorities needed to address challenges within the current ecosystem; and, (3) propose a Policy and Development Agenda that will contribute to realizing an ideal health information ecosystem in which both the health IT infrastructure and the data it supports are optimized.

Wednesday, March 25

S19: Panel – Early Experiences of the Cancer Moonshot’s IMPACT Consortium in Implementing Patient Reported Outcomes for Cancer Symptom Management

J. Richardson, RTI International; A. Wilder Smith, National Cancer Institute; A. Cheville, Mayo Clinic; M. Bass, Northwestern University; M. Hassett, Dana-Farber Cancer Institute

As part of its Cancer MoonshotSM, in 2018 the National Cancer Institute established an initiative to fund a consortium that aims to improve the monitoring and management of patients’ cancer-related symptoms using informatics solutions. The consortium, Improving the Management of symptoms during And following Cancer Treatment (IMPACT), is comprised of three research centers and a coordinating center that are tasked with collecting and sharing symptom data from across the cancer care continuum – at the point-of-care in oncology clinics and via remote settings – and evaluating the effects that cancer-related symptom management tools and data have on patients, through interventions conducted in care delivery organizations. The research centers leverage informatics strategies such as novel electronic health record user interfaces, patient-directed mobile health interventions, and symptom-based clinical decision support tools. This panel will discuss their efforts with developing and implementing symptom management tools, and highlight the solutions and challenges encountered. Learning objectives include understanding current issues with developing and implementing systems for tracking cancer-related symptoms and using decision support tools for effective supportive care and symptom management.

S20: Panel – Facilitating Research through the Childhood Obesity Data Initiative

A. Goodman, R. King, Centers for Disease Control and Prevention; P. Mork, A. Gregorowicz, The MITRE Corporation; M. Daley, Kaiser Permanente

Childhood obesity impacts almost 14 million U.S. children and is associated with serious and costly health risks. Research that assesses childhood obesity interventions is hindered by an inability to link patient records across information systems where pediatric health-related data are stored. The Childhood Obesity Data Initiative (CODI) aims to improve data capacity for childhood obesity research by facilitating access to patient-level, linked, longitudinal data that include health outcomes, information about weight management interventions, and risk factors. This panel will include a federal project lead, technical experts, and a clinical researcher who will present:

  1. ) The studies that CODI enables—childhood obesity comparative effectiveness research, program evaluation, and population health surveillance.
  2. ) The CODI data model; architecture for contributing data; process for assembling longitudinal data; and means by which research queries are processed.
  3. ) The perspective of a CODI participant, including research interests and the steps taken to participate in CODI.

S24: Panel – Open Source Research Analytics: Using Leaf for Cohort Discovery and Data Extraction

N. Dobbins, University of Washington; P. Nagy, Johns Hopkins University; G. Weber, Harvard Medical School; S. Mooney, University of Washington; K. Stephens, University of Washington; M. Beno, Case Western Reserve University

Academic medical centers and health systems are increasingly challenged with supporting appropriate secondary use of clinical data. Enterprise Data Warehouses (EDWs) have emerged as central resources for these data, but often require an informatician to extract meaningful information, limiting direct access by end users. Concurrently, the open source community is enabling open, modular, and innovative tools for data extraction, manipulation and visualization. In this panel, we will focus on the open source Leaf software, which we developed, and use cases within academic medical centers and their research and analytics strategy. Leaf is a lightweight self-service web application for querying clinical data from heterogeneous data models and sources. We believe that Leaf fits into a larger more modular, interoperable and shareable research analytics ecosystem. Here, we will introduce Leaf, provide an overview of Leaf functionality and then present several use cases from organizations participating in the Leaf community. These include the University of Chicago, Johns Hopkins University, Harvard University, Case Western University and the University of Washington.

S28: Panel – Real-world Implementation of Patient-reported Outcomes (PRO) into Electronic Health Record (EHR) Systems

S. Garcia, ONC; D. Meeker, University of Southern California; K. Bradford, Research Action for Health Network (REACHnet), Louisiana Public Health Institute

Advancement in clinical research and decision support is hindered by limited inclusion of patient reported outcomes (PRO) data in electronic health records (EHRs) and other health IT solutions, preventing the availability of patients’ perspective in their own care. Although there are some EHRs that capture PRO, such as the National Institutes of Health (NIH)-funded Patient Reported Outcomes Measurement Information System (PROMIS) instruments, collecting this data at the point of care is not common practice. Digital tools to streamline the collection of PRO are not widely adopted due to challenges in workflow integration and lack of standards. Patient perspective and autonomy remains a powerful tool to ensuring healthcare decisions are informed by patients and used to impact prevention, diagnosis, treatment and long-term care. This panels presents findings from a project that aims to address gaps in the availability of PRO data by standardizing the integration of PRO with EHRs and other health IT systems.

Thursday, March 26

S32: Panel – Leveraging Decision Support Analytics, FHIR, and Case Reporting for Information Exchange

N. Mishra, Centers for Disease Control and Prevention; J. Duke, Georgia Tech Research Institute; S. Karki, Centers for Disease Control and Prevention; N. Collins, Public Health Informatics Institute

Historically, reportable conditions were manually submitted to public health agencies. This practice is now being modernized by electronic reporting from either the laboratory or from the electronic health record (EHR). The electronic lab report (ELR) is a reliable method to report cases to public health as commercial and clinical labs send positive test results to the local health authorities. However, ELRs have less information available as it lacks demographic and additional clinical data. We added clinical and demographic data to the ELR via Fast Healthcare Interoperability Resources (FHIR) queries. To complete the information loop, we have used automated case reports as a data source for analytics and decision support. Patient data, including demographics, medications, and allergies, were analyzed by the clinical decision support algorithm to return gonorrhea treatment and screening recommendations to the provider. This continuous update of patient records (using FHIR) and bi-directional exchange of information enables electronic case reporting and actionable clinical decision support (CDS) to simulate a learning health system prototype.

S35: Panel – FHIR – Implementing the HL7 Interoperability Platform

C. Jaffe, Health Level 7 (HL7); J. Mandel, Microsoft; V. Nguyen, Stratametrics; J. Campbell, Epic; C. McDonald, National Library of Medicine

HL7 FHIR (Fast Healthcare Interoperability Resources) is now 10 years old. To date, it has been implemented in over 3000 sites worldwide. FHIR supports the broad continuum including patient care, population health, evolving payment models and clinical research. In March of last year, both the Center for Medicare and Medicaid Services (CMS) and the Office of the National Coordinated for Health IT (ONC) released two complementary Notices of Proposed Rulemaking (NPRM) to improve the Interoperability of health data. This panel will explore the emergence of FHIR implementation across the broad continuum of biomedical research, clinical care, patient empowerment, value-based payment systems, and population health.

S38: Panel – Research Data Network Ontologies for Precision Cancer Medicine supporting i2b2 and OMOP

W. Campbell, J. Campbell, University of Nebraska Medical Center; C. Reich, IQVIA; R. Belenkaya, Memorial Sloan Kettering Cancer Center

Research data networks created over the past decade including OHDSI and PCORnet offer the promise of access to large data sets to investigators for hypothesis testing. The data sets based on information collected and stored in the EHR are largely managed by common data models to support query interoperation. These data models, however, are insufficient to address the precision cancer medicine use case. The introduction and extension of network ontologies based on ONC standards greatly extends the utility and functionality of these research network to serve the cancer medicine use case.

The development, deployment and use of ONC standards into a shared ontology for cancer precision medicine research will be discussed by this panel. In particular, the integration of SNOMED CT, LOINC and RxNorm will be described along with research use cases in PCORnet and OHSDI networks and across the OMOP and i2b2 data structures. Information capture including discrete pathology and genomics data and its incorporation into these data warehouses will be reviewed.

Participants will gain insights into management and deployment of ONC ontologies and reference terminologies to support research objectives. Database requirements and operational considerations for successful use of ONC-based ontologies will be provided, as well as, example research queries.