Sunday, November 5
3:30 p.m. – 5:00 p.m.
S06: Panel – The Personal Journey: Informatics from the Patient Perspective
M. Kamerick, Sequoia Consulting/TriNetX, Inc.; J. Tenenbaum, Duke University; W. Hammond, Duke University
As informaticists, we often create tools and manage data to assist patients in making difficult treatment decisions. What happens when one of us becomes a patient? How do we approach the dilemma of our own complex disease and possible response, especially when multiple treatment options are available? What is our decision-making modality? Do we engage in the same calm, rational process we espouse to others, or does our new-found sense of vulnerability overwhelm our analytic faculties? How has our own experience changed the way we approach the goals of our profession? A panel of informaticists discuss their personal encounters with disease, decision making, treatment, and outcomes.
S07: Panel – Clinical NLP in Languages Other Than English
A. Neveol, CNRS; N. Elhadad, Columbia University; S. Velupillai, KTH University of Technology; H. Xu, The University of Texas Health Science Center at Houston; G. Savova, Children's Hospital Boston, Harvard Medical School
Natural Language Processing (NLP) of clinical free-text has been an active area of research in the past decades. Clinical documents are routinely created across health care providing institutions and are generally written in the official language(s) of the country these institutions are located in. As a result, free-text clinical information is written in a large variety of languages. While many efforts for clinical NLP have focused on English, there is a steady interest to extend this work to other languages in order to gain medical information about patient cohorts in geographical areas where English is not an official language. Furthermore, adapting current NLP methods developed for English to other languages may provide useful insight on the generalizability of algorithms and lead to increased robustness. This panel aims to provide an overview of clinical NLP for a wide variety of languages including Germanic, Romance, Sino-Tibetan and Afro-Asiatic languages. We will also discuss initiatives to support clinical NLP research and sharing of clinical resources for NLP in multiple languages.
S08: Panel – Optimizing Patient Care through Clinical Decision Support: Call to Action by the National Academy of Medicine
J. Tcheng, Duke University Health System; K. Kawamoto, University of Utah; B. Middleton, Apervita, Inc.; J. Teich, Harvard University; S. Weingarten, Cedars-Sinai Medical Center
With support from the U.S. Office of the National Coordinator for Health IT (ONC), the National Academy of Medicine (NAM) convened a multi-stakeholder collaboration including clinicians, researchers, industry, patient advocates, and government representatives to identify opportunities and strategies for advancing clinical decision support (CDS) practice and adoption. Work groups focused on four key CDS areas: content authoring, integration with information platforms, clinical operations, and dissemination. In this panel, leaders of the NAM initiative will summarize the key findings and the resulting Priorities for Action.
S09: Panel – Big Data to Knowledge (BD2K) and the Application of Metadata
G. Jiang, Mayo Clinic; W. Campbell, University of Nebraska Medical Center; T. Clark, Massachusetts General Hospital; C. Tao, The University of Texas Health Science Center at Houston; M. Musen, Stanford University
Clinical and translational research studies increasingly involve the manipulation of large datasets. There is a critical need to develop metadata-driven informatics technology to both lower the effort required for, as well as incentivize the process of applying metadata in clinical and translational research. The NIH Big Data to Knowledge (BD2K) initiative seeks to develop an interactive data ecosystem through adopting the FAIR principles (i.e., data must be findable, accessible, interoperable and reusable), and is supporting various technology platforms in these contexts to make it easier for researchers to share their data with the community so that it can broadly benefit the community. This thematic panel will highlight five projects funded through the BD2K initiative that are developing technology infrastructure and tools for metadata management, standardization and applications. Attendees will learn about novel open source metadata informatics platform, and how data/metadata standards are used in data sharing platforms. Attendees can engage with panelists on metadata challenges and potential solutions in support of clinical and translational research applications.
Monday, November 6
8:30 a.m. – 10:00 a.m.
S17: Panel – A Spoonful of Structure Helps the Workload Go Down: Modeling Clinical Cognition in Inpatient Documentation Tools
M. Mai, Children's Hospital of Philadelphia; K. Zheng, University of California, Irvine; E. Shelov, E. Orenstein, Children's Hospital of Philadelphia; S. Airan-Javia, University of Pennsylvania Health System, Perelman School of Medicine at the University of Pennsylvania
As inpatient care has grown in complexity and front-line clinicians (FLCs) must create and update a larger volume of information, inefficient and fragmented documentation systems put hospitalized patients at risk of harm from communication errors. In addition to writing and updating admission notes, progress notes, and discharge summaries, FLCs must also update plans of care and problem lists in their handoffs – often independently. Further, the information therein is generally locked in free-text, which cannot be easily reused for clinical decision support. With this duplicative workflow, FLCs struggle to maintain high quality notes and handoffs, and discrepancies between information documented between the two are known to occur. Problem-oriented documentation tools from electronic health record vendors have begun to address these issues by mandating coded problem names and allowing users to specify a block of information as historical. However, these tools lack the granularity required to optimize task management and handoffs. In this didactic panel, we will review the successes and failures of structured and flexible documentation systems, describe strengths and weaknesses of a vendor approach to problem-oriented documentation, and discuss novel models that organize clinical reasoning documentation into discrete elements that can be reused to improve clinical workflow.
S18: Panel – Engaging Patients with Health Technologies to Improve Quality of Care and to Reduce Preventable Harm
W. Pratt, University of Washington; P. Dykes, Harvard Medical School; R. Greyson, University of Pennsylvania; C. Ruland, Oslo University Hospital/University of Oslo; D. Bates, Harvard Medical School
With recent studies highlighting the pervasiveness and severity of preventable medical errors, renewed efforts have turned to health technologies to improve quality of care and patient safety. However, few efforts target patients and caregivers as primary users of quality and safety-oriented technologies. In this panel, we will describe four new efforts in developing health technologies to engage patients and their caregivers to improve quality of care and to reduce preventable harm. These efforts include co-designing with both a pediatric and adult population, using sensors to promote safe and healthy mobility, engaging hospitalized patients and caregivers in the safety plan, and using gamification techniques to engage patients in chronic illness management. Attendees will learn new approaches for engaging patients and their caregivers in ensuring they receive high-quality, safe care while they are in the hospital. The panel will provide insights that information technology designers, developers, and researchers can build upon to address caregivers’ and patients’ safety needs, while realistically fitting within the workflow of an organization.
S19: Panel – "My Work Will Surely Speak for Itself”: Visibility, Networking, and Self Promotion in Informatics
W. Chapman, University of Utah; M. Beene, Department of Veterans Affairs; O. Ogunyemi, Charles R. Drew University of Medicine and Science; G. Melton, University of Minnesota; L. Wiley, University of Colorado
Women and men interested in better understanding how to obtain leadership positions and to enhance leadership opportunities for women within our field should attend this interactive panel sponsored by the Women in AMIA Taskforce. Women continue to be under-represented in leadership positions in medicine and within informatics. Contributing factors include a person’s confidence level and degree of networking. Our objective is to discuss the confidence and networking differentials between men and women and to share strategies for reconciling the fear many women (and some men) have of appearing over-confident with the need for self- promotion and visibility needed to achieve positions of leadership and increase impact. The panel will focus on four areas: (1) Selling yourself: the confidence spectrum; (2) Climbing the ladder: Academic career progression; (3) Women in the C-Suite; (4) Strategies for Students.
S20: Panel – The Data and Research Center of the All of Us Research Program: Framework for a National Cohort Program and Research Opportunities
R. Carroll, Vanderbilt University Medical Center; J. Mandel, Verily; K. Natarajan, Columbia University; S. Sutherland, Broad Institute; J. Denny, Vanderbilt University Medical Center
The Precision Medicine Initiative All of Us Research Program is a national effort to recruit and engage at least one million participants who consent to provide health information (including data from health surveys, electronic health records, and baseline physical measures), biospecimens, and to be recontactable in the future. The Data and Research Center (DRC) sits at the confluence of all of these varied data and, in collaboration with the many All of Us partners, has been creating a framework to shape and support biomedical research in the future. This panel will discuss a number of primary topics including operations support, the back end data structure, EHR data ingest, processing, and curation, and the researcher tools and public facing sites with a number of interwoven secondary topics including principles driving the creation of the All of Us Research Program, the ethical, legal, and social implications of the effort, and the privacy and security framework for protecting participants and their data. We are interested in engaging the AMIA community in discussions around the plans to engage and protect participants and empower this community of researchers.
Monday, November 6
10:30 a.m. – 12:00 p.m.
S27: Panel – Big Data in the Intensive Care Unit
M. Adibuzzaman, Purdue University; A. James, The Hospital for Sick Children; J. Zaleski, Bernoulli; P. Haug, University of Utah
Big data has brought much promise for discovery of treatment and therapies, drug safety, and the care delivery processes by identifying which treatment would work best for which patients. NIH has recently taken the initiative, ‘All of Us’, to collect one million or more patients’ data (electronic health records (EHR), imaging, genomics, environmental data, etc.) over the next few years. The intensive care unit represents a unique data source in this context, with carefully captured detailed high volume data from different systems such as EHR, electrocardiogram, blood pressure, infusion pumps, and photoplethysmogram, among other data. However, the mere availability of data does not translate into knowledge or improved outcome. Questions remain on what data is needed, how to integrate these high-volume data with high throughput infrastructure for near real-time decision making by the clinicians. In this panel, we present our work in integrating this heterogeneous high volume data with state of the art technologies for retrospective analysis and near real-time decision making with different systems such as Medical Information Mart from Intensive Care Unit (MIMIC III) and Artemis: both from technological and clinical perspectives.
S28: Panel – Getting Hooked on CDS: Toward an Open Standard Architecture for Clinical Decision Support in Leading Electronic Medical Records
B. Middleton, Apervita, Inc.; A. Boxwala, Meliorix Inc.; J. Overhage, Cerner Corp.; J. Doyle, Epic Systems, Inc.; T. Rothenhaus, Athenahealth
This interactive panel will discuss the current state-of-the art with respect to implementing clinical decision support in commercial EMRs, the state of standards used in clinical decision support, and newer approaches to externalizing knowledge management and clinical decision support as web services consumed by an EMR. The discussion is timely given the explosive growth of new knowledge creation and the need for enhanced decision support for clinicians, and their patients, in the areas of precision medicine, care coordination, chronic disease management, and in the setting of rapidly changing clinical practice or acute disease outbreaks (e.g. Zika). The goals of this panel are to facilitate a discussion with leading EMR vendors, application builders, and CDS platform vendors to describe about their approach to CDS and the challenges above.
S30: Panel – Innovation in Workflow Methods for Consumer Health Informatics
M. Ozkaynak, University of Colorado; R. Valdez, University of Virginia; G. Demiris, Y. Choi, University of Washington; L. Novak, Vanderbilt University Medical Center; C. Weir, University of Utah
Consumer health information technologies are becoming an essential part of wellness and management of chronic conditions. The literature on health informatics emphasizes that workflow studies are key to the design and implementation of health information technologies in clinical settings. Consumer health informatics interventions should also be informed by workflow in daily living settings, however, additional challenges exist such as, privacy, variability in living styles, preferences and daily routines. Workflow methods that have been utilized in clinical settings should be thoughtfully adapted to daily living settings. Compared to traditional approaches in institutional health care settings, novel methods may be needed in the home that account for the unstructured and personal nature of daily living settings. The success of workflow studies depends on the robustness of the methods used to capture, analyze and re-design workflows. This panel will begin with a discussion by each panelist of their experience in workflow research for consumer health informatics interventions and their use of innovative methods. This will be followed with an interactive session, where the audience and panelists will create a list of challenges. Finally, the audience and panelists will discuss effective strategies to address these challenges.
S31: Panel – Clinical Decision Support in the Era of Precision Medicine
M. Grasso, University of Maryland School of Medicine; B. Blumenfeld, RTI International; J. Rutter, National Institutes of Health; E. Siegel, University of Maryland School of Medicine; L. Wedemeyer, Greater Los Angeles VA Healthcare System
The All of Us Research Program from the Precision Medicine Initiative will create new opportunities and challenges for clinical decision support (CDS). An import goal of the program is to create a research cohort of at least one million U.S. volunteers, which will be used to develop customized approaches to treat diseases and improve health. This panel will discuss new opportunities and challenges for CDS using the large research cohort. This panel, which was organized by the AMIA Clinical Decision Support Working Group, brings together leaders from government, academia, and industry with diverse backgrounds in Biomedical Informatics.
Monday, November 6
1:45 p.m. – 3:15 p.m.
S39: Panel – The Good, the Bad, and the Ugly of Deploying and Adopting Machine Learning Based Models in Clinical Practice
Y. Aphinyanaphongs, New York University; D. Holmes, Mayo Clinic; P. Mirhaji, Albert Einstein College of Medicine; M. Draugelis, University of Pennsylvania
Over the past three decades, the growth of publications involving machine learning has grown in the healthcare literature. More recently, to leverage the potential benefits of these machine learning based models, institutions have begun making investments in infrastructure to translate this academic research into clinical practice. Deployment and adoption of models in clinical practice is not trivial and many challenges and issues emerge during implementation. This panel will provide a broad view of the machine learning model pipeline through model inception, evaluation, deployment, and adoption. Through a series of case studies, the panel will demonstrate the challenges and issues. The goal of this session is to begin to form a body of work for best practices, establish a like-minded community, and demonstrate potential scientific research opportunities toward successful model deployment and adoption.
S40: Panel – Successes and Challenges in Developing and Implementing Electronic Informed Consent Tools for Research
C. Harle, Indiana University/Regenstrief Institute; H. Kim, University of California; D. Nelson, University of Florida; K. Goodman, University of Miami
The U.S. health care system’s increased adoption and use of electronic health records (EHRs) over the past decade has increased the utility of health records for research. With these technological changes, new approaches and systems are needed to efficiently and appropriately inform patients about and consent them to participate in research, especially research that uses their electronic health information. In this panel, three researchers from different disciplines will describe three novel efforts to develop, implement, and evaluate point-of-care electronic systems that administer broad research consent. These systems span multiple institutions, are integrated with institutional EHRs, and involve consent for research re-contact and patient sharing of health records and bio-samples for research. At the conclusion of this panel, audience members will be able to: (1) discuss technological, process, and ethical challenges and keys to success in designing electronic consent tools for point-of-care use, (2) describe operational and technical challenges and keys to success in integrating research consent tools with EHRs and other enterprise information systems, and (3) critique qualitative and quantitative approaches to evaluating the effectiveness of electronic informed consent tools.
S41: Panel – The Jordan Experience: The VistA Experience
T. Cullen, Regenstrief Institute/Indiana University; F. Kamal, Electronic Health Solutions Co; N. Anthracite, Spectrum
The Hashemite Kingdom of Jordan is located in the Middle East at the crossroads of Asia, Africa, and Europe with a population of over 9 million. In 2009 under the patronage of His Majesty King Abdullah II Bin Al Hussein, Electronic Health Solutions (EHS) was founded to advance the healthcare sector in the country through its flagship program Hakeem. The Hakeem program facilitates efficient, high-quality healthcare in the Kingdom through the nationwide implementation of an Electronic Health Record (EHR). Jordan elected to use an open source version of VistA. The Jordanian version of VistA, a collaboratively supported free and open platform adopted from World VistA, is currently in use in over 104 facilities (primary healthcare clinics, comprehensive healthcare clinic, and hospitals) across various public healthcare authorities in the Jordan Ministry of Health, Royal Medical Services, and King Hussein Cancer Center. The benefits of open, collaboratively developed health information technologies (HIT) have been well described, and strongly advocated for over the past decade. Many have found it difficult to successfully initiate, grow, and maintain these initiatives. The Jordan experience with WorldVistA EHR provides a potent reminder of the possibilities and promise of open source HIT as well as the possibilities and impact of full spectrum HIT in low and middle income countries (LMIC).
S42: Panel – The Best of Imaging Informatics Research 2017
W. Hsu, University of California, Los Angeles; C. Kahn, University of Pennsylvania
The field of imaging informatics is rapidly advancing in its ability to address challenges related to clinical big data and harnessing this information for precision medicine. In the past year, the field has experienced growth in a variety of areas including machine learning (learning from imaging data for diagnostic and prognostic predictions), radiomics (the generation of high-dimensional features from images), patient-oriented sharing and communication of images and image-derived findings, and the growth of the use of imaging beyond radiology. Novel approaches go beyond pixel data to integrate imaging with other biomedical data, standardize imaging workflows, and improve the quality and utility of image-derived information in clinical practice. In this session, we will review key advances in imaging informatics research published this past year.
Monday, November 6
3:30 p.m. – 5:00 p.m.
S50: Panel – Current State of Visualization of EHR data: What's Needed? What's Next?
J. Caban, Walter Reed National Military Medical Center; V. West, Duke University; H. Kharrazi, Johns Hopkins University; D. Dowding, Columbia University; D. Wu, University of Cincinnati
As the amount of health-related information being generated and collected by medical organizations continue to increase, the amount of data available to healthcare providers within the electronic health records (EHRs) continues to growing at an unprecedented rate. This vast amount of health-related data poses cognitive and comprehension challenges for medical professionals to understand patients' medical histories and conditions at the point of care and for researchers studying individual- and population-level exposure and outcome data. Information visualization and visual analytics have recently received a significant amount of attention as potential techniques to support the cognitive processes, understanding, and decision-making for anyone gathering and using vast amounts of digital health data.
In this didactic panel, we will review the current state-of-the-art visualization techniques that have been incorporated within EHRs, discuss the areas of healthcare in which better tools and visualization techniques are needed, and examine the need for user evaluations to better understand the effectiveness of different visualization techniques and dashboards in clinical settings. The panelists will also summarize some of the work of the AMIA VIS working group in increasing the awareness of the importance of effective visualization techniques when analyzing large collection of clinical data.
S52: Panel – Computational Phenotyping on Diverse Data Sources
J. Sun, Georgia Tech; B. Malin, Vanderbilt; A. Kho, Northwestern University; M. Craven, University of Washington; J. Ghosh, University of Texas Austin
Phenotypes are the measurable biological, behavioral and clinical markers of a condition or disease. The process of deriving research-grade phenotypes from clinical data using computer-executable algorithms is called computational phenotyping (phenotyping for short) (Pathak, Kho, and Denny 2013). Phenotyping includes a range of approaches from finding a phenotype using expert-derived rules and those phenotypes emerging from novel computational methods that potentially represent new clinical entities. Traditionally, it focuses on algorithms for extracting clinical phenotypes from electronic health records (EHR). However, there is an increasing need to broaden the scope of computational phenotyping 1) to capture and process more diverse health related data such as omic and mHealth data; 2) to extract more sophisticated phenotypes such as temporal and spatial patterns; and 3) to speed up algorithm development, validation and deployment for supporting various clinical and research applications.
S53: Panel – Latin American e-Health Programs and WHO/PAHO Planning Toolkit
Y. Quintana, Harvard University; D. Novillo-Ortiz, Pan American Health Organization; M. Lopetegui, Clínica Alemana ; M. Cubillo, Hospital San Vicente de Paul de Heredia; H. Marin, Federal University of São Paulo; P. Otero, Hospital Italiano de Buenos Aires
The goal of this panel is to provide an overview major initiatives in Latin America, provide an overview of the Pan American Health Organization/World Health Organization (PAHO/WHO) national e-health strategic planning toolkit, challenges in implementation of e-health systems, overview of the PAHO e-health strategy for the Americas, and provide approaches for implementing programs using cooperative models.
Tuesday, November 7
8:30 a.m. – 10:00 a.m.
S61: Panel – Can the Clinical Information Modeling Initiative (CIMI) Enable the Semantic Interoperability Promise of Fast Healthcare Interoperability Resources (FHIR)?
S. Huff, Intermountain Healthcare/University of Utah; C. Nanjo, Cognitive Medical Systems; J. Skapik, Office of the National Coordinator for Health IT (ONC)
As implied directly in its name, the Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) standard is intended to enable semantic interoperability. However, the base FHIR-defined classes (known as FHIR “resources”) do not have sufficient specificity to enable such interoperability. Constrained versions of those classes (known as FHIR “profiles”) are intended to provide sufficient specificity to support the semantically interoperable exchange of clinical data. However, multiple, semantically inconsistent FHIR profiles on the same topic can and have been developed; there is no governance around profile development; many profiles are still insufficiently specified to enable semantic interoperability; there is no mechanism to ensure consistency between FHIR profiles; and vendor support for FHIR has focused primarily on resources and not profiles, and are therefore potentially (and likely) inconsistent with one another. HL7 CIMI is working with the FHIR team to address these problems and to enable FHIR to be truly semantically interoperable. This panel will be an interactive discussion to address two questions: first, is FHIR capable of supporting true semantic interoperability without an effort such as CIMI? And second, is CIMI the answer to FHIR’s challenges in this area?
S62: Panel – The PCORnet Learning Cycle
K. Marsolo, Cincinnati Children's Hospital Medical Center; L. Qualls, B. Hammill, L. Curtis, Duke Clinical Research Institute; J. Brown, Harvard Pilgrim Health Care Institute
The National Patient-Centered Clinical Research Network (PCORnet) is a distributed network-of-networks, where partners transform data from electronic health record (EHR) and claims data sources into a Common Data Model. In order to ensure that the underlying data are suitable for use in network queries, PCORnet employs a robust data characterization process. The tools and processes that support PCORnet are not static, however. They are constantly being refined in response to stakeholder needs as well as the findings from analytic queries and characterization results. This panel will describe the Learning Cycle that informs the activities of the PCORnet Distributed Research Network Operations Center (DRN OC) and will highlight the relationship between the PCORnet CDM, the Data Characterization process, the findings of the PCORnet Demonstration Projects, and downstream analytical tool development. In addition, the panel will discuss how all of these findings are communicated out to network participants and other stakeholders. Attendees will gain an understanding of why these components are all necessary to the successful operation of a distributed research network, and recognize how similar Learning Cycles can be used to improve the quality and reproducibility of any network’s research.
S63: Panel – PCMEDICS: Peace Corps' Home-grown EMR
T. Wilkinson, B. Shannon, K. Skrtic, M. Charles, L. Knauf, United States Peace Corps
In 2016, Peace Corps deployed a cloud-based electronic medical record system to support the healthcare of Volunteers deployed throughout the world. The solution was designed, built, deployed, and supported by the Peace Corps’ Health Informatics Unit (HIU) in Washington, D.C. Developers were embedded directly into medical operations at Peace Corps, and the resulting EMR has enjoyed significant support (even admiration) from its many end-users. This panel is a presentation of the HIU’s experience. It highlights some of the unique strengths of the EMR itself, explores issues around security in modern Federal IT systems, and describes the particular emphasis which the HIU has placed on a quality training and support program during go-live and in the year since.
S64: Panel – Enabling Knowledge-driven Care at Scale through CDS Hooks and the FHIR Clinical Reasoning Module
K. Kawamoto, University of Utah; K. Shekleton, Cerner Corporation; J. Doyle, Epic Systems Corporation; B. Rhodes, HarmonIQ Health Systems Corporation; H. Strasberg, Wolters Kluwer Health
Interoperable clinical decision support (CDS) holds great promise for ensuring that patients receive the best care possible based on the latest medical knowledge. Building on prior efforts at standards-based CDS interoperability, as well as rapid adoption of the HL7 FHIR standard by the EHR vendor community, various stakeholders including EHR vendors, knowledge vendors, and healthcare systems are working together to enable interoperable knowledge-driven care through the CDS Hooks and FHIR Clinical Reasoning specifications. These specifications are being harmonized into a single unified HL7 specification, and the goal of this unified specification is to enable EHR systems to both (i) invoke external CDS Web services and (ii) consume external knowledge artifacts such as order sets, rule specifications, and documentation templates in a standard manner. In this panel, the latest developments will be discussed by leaders of this effort from the EHR vendor community, the CDS knowledge vendor community, the standards development community, and the healthcare provider community. The panel will demonstrate interoperable CDS implemented on multiple EHR platforms using this approach. The panelists will also describe key challenges and lessons learned, solicit feedback from the audience, and describe planned future work to continue advancing health and health care through interoperable CDS.
S65: Panel – 20 Years of Digital Pathology – An Overview of the Road Travelled and What is on the Horizon
J. Saltz, Stony Brook University; A. Sharma, Emory University; A. Carter, Children’s Healthcare of Atlanta; L. Pantanowitz, University of Pittsburgh; T. Kurc, Stony Brook University
Over the past 20 years, Digital Pathology has developed into a thriving field with both clinical and research applications. FDA approval is anticipated this year to allow use of digital whole side imaging for primary Pathology diagnosis. Once this occurs, we expect digital Pathology to become as ubiquitous in clinical practice as digital Radiology. While the concept of the Virtual Microscope, a software tool that implements the operation of a whole slide microscope and enables viewing of large digitized tissue samples, is now two decades old, the adoption of Digital Pathology informatics tools in clinical practice is still a work in progress. This panel will describe the history, state of the art, promise and challenges in digital Pathology Informatics. Informatics related challenges include the need for interoperability and standards, integration of analytic pipelines and results into clinical decision support.
Tuesday, November 7
10:30 a.m. – 12:00 p.m.
S72: Panel – An Overview of Emerging Real-world OpenHIE Use Cases: Successes, Challenges, and Future Opportunities
S. Grannis, Indiana University/Regenstrief Institute/Indiana University School of Medicine; E. Manders, Centers for Disease Control and Prevention; F. Leitner, PATH; A. Ngaruro, ICF International; J. Bowie, Apelon
A growing international collaborative of diverse organizations and people have formed a community to develop open processes and technologies to support essential clinical data sharing needs of low and middle income countries throughout the world. OpenHIE (Health Information Exchange) is a global initiative convening communities of practice, implementers, and reference software components that together support national-level health information sharing activities. In this panel we will describe multiple real-world use cases supported by OpenHIE’s operational infrastructure, and will discuss the evolving community process, opportunities for broader collaboration, and future development plans. Learning objectives for the panel include: understanding the basic elements of a component-based eHealth architecture for low and middle-income countries, as exemplified by OpenHIE; describing specific use cases supported by OpenHIE; and identifying potential barriers and solutions for implementing specific health information exchange use cases for low and middle-income countries.
S73: Panel – A Structured Approach to Measuring Individual Nurses Contribution in Patient Outcomes
E. Harper, University of Minnesota; C. Ivory, Vanderbelt University; L. Moon, University of Minnesota; A. Garcia, Cerner Corporation
Nurses are key drivers of patient outcomes as well as a significant cost center in any healthcare setting. How do we measure nursing care value – or more importantly, the unique value each nurse contributes to patient outcomes? At the most basic level of definition, value is the relationship between outcomes of care and the costs to produce that care. These two basic elements of cost and outcome are measured using the individual patient as unit of analysis, Traditional hospital outcome metrics are measured in terms of length of stay, the presence or absence of adverse events such as infection, fall, or pressure ulcer, or inpatient mortality. However, the volume of data generated by clinical and operational electronic health records (EHRs), permits the examination of short term changes in patient outcomes attributable to nursing care. EHR data allows measurement of both the patient response to treatments or nursing interventions, at the individual nurse level and correlate both with nurse characteristics such as education and experience levels. This didactic panel will describe the methodological approach taken by the Nursing Value Expert Workgroup to structure, define, abstract and evaluate discrete data from disparate electronic systems to measure nurse’s contribution to patient outcomes.
S74: Panel – Are EHRs "Overloading" Health Professionals? Issues, Advances, and New Directions from Cognitive Science and Usability Engineering
A. Kushniruk, E. Borycki, University of Victoria; Y. Senathirajah, SUNY; D. Hudson, University of Alberta
Issues about the usability and safety of electronic health records have come to the fore. Reports of systems that user find difficult to use, interfere with healthcare workflows and that make them feel “cognitively overloaded” continue to be reported. The concept of “cognitive load” emerged from the cognitive literature several decades ago and refers to the amount of mental effort required to carry out tasks. In areas such educational design the concept has proven useful, as well as in other areas such as aviation and cockpit design. However, far less work in cognitive load has been applied in the area of human factors in healthcare. In this panel presentation, experts from the United States and Canada will explore the concept of cognitive load, including its definition, its measurement and its implications, for improving the usability and safety. Implications for the design and testing of a wide range of health information systems and technologies will be explored. The learning objectives include increasing awareness of cognitive load, understanding the concept, understanding how it can be measured, and understanding what can be done to decrease “cognitive overload”.
S75: Panel – Participatory Health Informatics for Data-driven Precision Medicine
L. Fernandez-Luque, Qatar Computing Research Institute; J. Srivastava, University of Minnesota; K. Mandl, Boston Children's Hospital; F. Martin-Sanchez, Weill Cornell Medicine
Lifestyle plays a major role in the prevention and management of chronic conditions. Sleep, physical activity and nutrition behaviors are being found to play a major inter-related role in physical and mental wellness of all of us. Worldwide empowered citizens are using new technologies such as social media, mobile apps and wearables to track their health and also get support from digital health services and peers. In that process, big and heterogeneous data sources are being generated. This panel will explore how data generated by empowered patients can facilitate the development of precision medicine. The panelist will address, among others topics, a) the use of social media to monitor lifestyles, b) data analytics of health wearables, c) the linkage between Genotype/Phenotype/Expotype data, and d) integration with other biomedical data sources. Using an interactive approach with the audience the panel will aim to identify key elements for the development of a roadmap for research and practice.
Tuesday, November 7
1:45 p.m. – 3:15 p.m.
S83: Panel – And Now for Something Completely Different: Successful Career Transformations in Biomedical and Health Informatics
A. Wang, Northwestern University; D. Fridsma, AMIA; J. Murphy, IBM; C. Patel, Applied Informatics, Inc.
The health and biomedical career landscape continues to grow and evolve as new technological opportunities and challenges arise. As professionals pursue new opportunities, careers take winding paths across unknown terrain. In an interactive panel discussion, informatics leaders who have experienced or guided colleagues through dramatic career transformations will address this topic. Panelists have transitioned between academic, research, educational, government, nonprofit, and industry realms. The panel will discuss the development of “soft” skills, such as vision, strategy, perseverance, resilience, communication, mentorship, and emotional intelligence, which are valuable for all professionals. Panelists will also address skills that are relevant to specific domains and roles. The presentation will begin with defined questions and then discuss questions from the audience. The purpose of the panel is to help informatics professionals identify and develop skills to understand the full breadth of career opportunities and paths and prepare and plan for career transformations. The intended audience includes informatics students or professionals who intend to transform their careers or mentor colleagues. This panel is jointly sponsored by the Education and Student Working Groups and supports AMIA’s core mission of professional growth and leadership development for our members.
S84: Panel – Design and Implementation of a Structured Sequencing Report Format: A Multi-Stakeholder Perspective from eMERGE
L. Rasmussen, Northwestern University; D. Ames, DNAnexus; S. Aronson, Partners HealthCare; L. Babb, GeneInsight; C. Overby, Johns Hopkins University/Geisinger Health System
With continued interest in adopting genomic medicine practices, an ongoing barrier has been around enabling the transmission of genomic laboratory results in a structured manner that allows a receiving institution to perform local computations on those results, including those for clinical decision support. The electronic Medical Records and Genomics (eMERGE) network, now in its third phase, has implemented a structured representation of results for a custom sequencing panel of over 100 genes and 1500 SNPs. This panel will describe the selection of the structured report format, the experiences of two clinical sequencing centers in aligning their laboratory systems to the format, how the format is received by eMERGE institutions, and the lessons to date that have potential relevance to other institutions exploring the integration of structured genomic results.
S85: Panel – Social and Behavioral Determinants of Health: Fundamental Informatics Challenges for Enriching Health IT Systems
J. Puro, Ochin, Inc.; D. Vreeman, Regenstrief Institute; A. Taylor, Department of Health and Human Services; M. Verzola, Epic Systems
In 2015, the Institute of Medicine released Capturing Social and Behavioral Domains and Measures in Electronic Health Records: Phase 2. This document identified core domains that were recommended for inclusion in patients’ health histories, and endorsed specific measures for each of these core domains. This work reflected the substantial impact of these factors on patients, families and communities. This panel is designed to share the experience of a diverse set of stakeholders who have been engaged in the recognition of social determinants of health (SDH) factors on health status and the current and future state of SDH domains and measures within electronic health records. The panel will present the current ONC perspective on the 2015 HIT certification criteria that included many of the IOM endorsed domains and measures, highlight the critical work that a standards development organization (LOINC) plays both currently and in the future, implementation challenges identified and addressed within a health care system (Verzola), and the integration and use of SDH data within primary health care (Puro). The panel presentations will emphasize lessons learned as well as current and future barriers that can be addressed to foster more inclusion of SDH factors in clinical care.
This panel is sponsored by the AMIA Social Determinants of Health (SDH) discussion group.
S86: Panel – The State of Interoperability and Health Information Exchange: Advancing Steadily or Treading Water?
W. Sujansky, Sujansky & Associates, LLC; J. Adler-Milstein, University of Michigan; R. Martin, Chesapeake Regional Information System for Our Patients
Prompt and efficient access to clinical data, regardless of its source, is one of the key promises of electronic health record technology as EHRs approach universal adoption. But in our nation’s complex and fragmented health care system, that goal requires interoperability and health information exchange across disparate information systems and business entities. In this panel, experts from academia and industry explore whether the U.S. healthcare system is now finally nearing the objective of broad-based technical and organizational interoperability for health data or still struggling against formidable barriers that have frustrated past interoperability initiatives. The panel presentations and discussion will assess the current state of interoperability and health information exchange (IHIE), as well as the potential contributions of contemporary IHIE initiatives, such as the FHIR standards and the Carequality and CommonWell industry consortia, to advance IHIE. The speakers will demystify these initiatives, highlight their novel attributes, and identify their remaining limitations in delivering on the promise of interoperability and health information exchange. The discussion will engage the audience and explore specific actions that informatics and health I.T. stakeholders can take to advance interoperability and HIE goals.
Tuesday, November 7
3:30 p.m. – 5:00 p.m.
S94: Panel – From Large-scale Network Analytics to Clinical Solutions in OHDSI
J. Duke, Georgia Institute of Technology; G. Hripcsak, Columbia University; P. Ryan, Janssen R&D; N. Shah, Stanford University
The Observational Health Data Sciences and Informatics collaborative (OHDSI, pronounced ‘Odyssey’) aims to create reliable scientific evidence through large-scale analysis of observational health data from around the world. OHDSI promotes the translation of this work to improve clinical research and medical decision-making. In the proposed panel, members of the community will discuss how OHDSI is working to support clinical research efficiency as well as support evidence sharing for medical decision-making by adopting the FHIR integration. To ensure transparency and reproducibility in our research, we have created shared knowledge artifacts such as phenotype libraries. In the proposed panel, we will discuss our approach to building shareable knowledge assets, conducting research using these assets, and supporting dissemination beyond the OHDSI community.
S95: Panel – Redesigning the "Choice Architecture" of the EHR to Improve Medical Decision Making
J. Ancker, S. Malhotra, Y. Zhang, A. Cheriff, Weill Cornell Medical College
Medical informatics was established to improve medical decision-making. In this panel, experts will illustrate how the design of health information technology inherently influences decision-makers and explore how design can be optimized for better decisions while in some cases minimizing the need for alert-based clinical decision support. These design interventions exploit the technology’s “choice architecture,” a term from cognitive psychology which describes the way a system presents choices to decision-makers. (For example, the US “choice architecture” presents organ donation as an opt-in decision; other countries present donation as an opt-out decision in which anyone who does not sign an opt-out registry is eligible to be a donor.) Through three case studies, panelists will demonstrate how decisions can be improved by choice architecture in the context of electronic prescribing, order sets, and patient portals. After the panel, participants will be able to: define “choice architecture”; describe how “choice architecture” interventions can provide alternatives to clinical decision support alerts; analyze ethical implications of intervening in the “choice architecture”; describe several successful EHR “choice architecture” interventions.
S96: Panel – The Impact of Human Microbiome on Precision Medicine
A. Alekseyenko, Medical University of South Carolina; L. Proctor, National Institutes of Health; P. Carlson, Food and Drug Administration; S. Jackson, National Institute of Standards and Technology; C. Chen, The Health and Environmental Sciences Institute
The human microbiome consists of the organisms that accompany us at all body sites our entire life and contribute to the health and disease processes. Actionable associations of the host microbiota with health conditions span many health areas, such as oral and dental health, inflammatory bowel disease, metabolic syndrome, and other systemic conditions, and cancers. In addition, the value of microbiome for diagnostic and prognostic tasks is becoming more prominent. As such, microbiome is becoming a vehicle for precision medicine research. The underlying data enabling microbiome research are high-throughput, complex, and heterogeneous. Extensive informatics expertise is required at all translational steps (basic, clinical, consumer and public health) to understand and utilize these data for precision medicine. The purpose of this panel is to discuss the representative precision medicine efforts in the microbiome domain at academic, public, and federal institutions. Our goal is to understand the breadth of deployed expertise in these efforts and achieve roadmap for alignment of informatics efforts to maximize the broad positive impact on precision medicine.
S97: Panel – Evaluating Impacts of Patient-facing eHealth Technologies: Connected Care in the Veterans Health Administration
S. Shimada, Department of Veterans Affairs/Boston University School of Public Health/University of Massachusetts Medical School; T. Hogan, T. Houston, Department of Veterans Affairs/University of Massachusetts Medical School; K. McInnes, Department of Veterans Affairs/Boston University School of Public Health; N. Evans, Department of Veterans Affairs
Patient-facing eHealth technologies such as patient portals, mHealth applications, home telehealth devices, and automated texting systems are increasingly common. Rigorously evaluating the impacts of these technologies is thus increasingly important. These ‘impacts’ can include both the intended positive effects on access, patient engagement, and health outcomes as well as other unintended adverse consequences. In this panel, we will present new data on the impacts of a range of patient-facing technologies being implemented across the Veterans Health Administration (VHA), including the My HealtheVet patient portal, Secure Messaging, and VHA’s automated, tailored text-messaging system. Attendees will learn about current and upcoming patient-facing technology implementation in the VHA, approaches for gauging different stakeholder experiences, and to select appropriate measures for positive and negative impacts of patient-facing technology. Implications for future evaluations will also be discussed.
The panel will present findings from a randomized trial of secure messaging and its impact on perceived healthcare access, an evaluation of the implementation of an automated texting system to support patient self-management, and a multi-modal study to identify and categorize unintended consequences of multiple patient portal features.
S98: Panel – Social Determinants of Health: Applied Informatics Approaches Incorporating Context into Care
M. Cantor, NYU School of Medicine; H. Estiri, MGH; R. Gold, Kaiser Permanente; T. Cullen, Regenstrief Institute
There is substantial empirical evidence that many of the factors that contribute to health outcomes originate from outside of the conventional confines of the U.S. healthcare system. Social Determinants of Health (SDH) constitute the majority of these factors. The uptake of Electronic Health Record (EHR) systems offers great potential for addressing such socio-contextual determinants of health. Integrating SDH data into EHRs would provide invaluable information for healthcare organizations, helping to improve precision in diagnosis, treatment, and monitoring of disease and care management programs. Yet, the current healthcare system generally lacks the infrastructure to assimilate SDH data within clinical practice and research. This panel will discuss theoretical aspects and practical steps for developing informatics infrastructures to incorporate data on socio-contextual determinants into healthcare practice and research.
Wednesday, November 8
8:30 a.m. – 10:00 a.m.
S107: Panel – Clinical Informatics in Medical Education: Innovations from the AMA Accelerating Change in Medical Education Initiative
W. Hersh, Oregon Health & Science University; S. Skochelak, American Medical Association; A. Spickard, Vanderbilt University; B. Takesue, Indiana University; P. Gorman, Oregon Health & Science University
Competence in clinical informatics is required of all 21st century health professionals. This panel will describe efforts to bring clinical informatics competencies, curricula, and integration into medical school curricula in the context of the American Medical Association initiative, Accelerating Change in Medical Education. After an overview of the initiative by one panelist, the other three will describe specific projects in curriculum development in clinical informatics for medical students, data-driven monitoring of medical student achievement and progress, and the development and use of a teaching electronic health record.
S108: Panel – Collaborative Pharming
M. Weiner, Temple University School of Medicine; J. Boehne, CHPSO, a Division of the Hospital Quality Institute; T. Kornet, R. Koppel, University of Pennsylvania
One EHR activity that significantly affects patient safety is medication ordering and administration, where the interaction and communication among physicians, pharmacists and nurses is crucial. The different information requirements and priorities of these clinical team members and EHR users creates design challenges in the development of a pharmacy interface in both the inpatient and ambulatory environments. Failure to achieve these goals creates a system that, while functional, is neither fully optimized nor conducive to medication safety.
This interactive panel will bring together stakeholders from the physician, usability/workflow analysis, nursing, and pharmacy communities to foster dialog across the groups. Participants will discuss their respective priorities and pain points in terms of addressing information needs, communication, workflow, and data recording requirements when ordering, processing and administering medications. Improved understanding of the other stakeholders’ needs should provide valuable insight into what information is needed at various points in the process, and the engagement of human factors expertise on the panel will help bridge the gap between existing medication ordering and administration interface design. The ideas and findings of this panel will help inform the design of a better pharmacy interface.
S109: Panel – Advancing Electronic Case Reporting (eCR) to Enable Public Health Disease Control and Emergency Response: Getting into the Technical Weeds!
C. Staes, University of Utah/Council of State and Territorial Epidemiologists; J. Loonsk, CGI Federal; K. Turner, Idaho Department of Health and Welfare; N. Arzt, HLN Consulting, LLC; P. Zarcone, Association of Public Health Laboratories
Many healthcare institutions grapple with the legally required mandate to report information to public health authorities when patients present with conditions of public health importance, such as pertussis, salmonella, Zika virus infection. Existing reporting of communicable diseases by health care providers is incomplete, untimely and predominantly a manual, paper-based process. Electronic case reporting (eCR) is relevant for over 200 conditions reportable in over 3,000 public health jurisdictions in the US. As such, eCR requires an ecosystem (knowledge, decision support, infrastructure, and standards) that can meet unique jurisdictional and clinical needs. The panelists represent leaders in the public health and informatics community involved in building the platform that supports standards-based services and defining the eCR-specific HL7 standards for reporting and response. The panel objectives are to a) describe the need for a scalable, flexible strategy for eCR that meets both routine and emergency reporting needs; b) describe the decision support services, infrastructure, and HL7 document standards that support bi-directional communication for eCR; and c) explore the strengths and limitations of the current eCR strategy. The Zika infection case reporting scenario will be used to illustrate key findings, and experience from pilot testing during 2017 will be shared.
Wednesday, November 8
10:30 a.m. – 12:00 p.m.
S119: Panel – Opportunities and Challenges for Development, Implementation, and Investigation of Acute Care Patient Portals (ACPP): Recommendations from the ACPP Task Force
A. Dalal, S. Collins, Brigham and Women's Hospital; V. Tiase, New York-Presbyterian Hospital; K. O'Reilly, Beth Israel Deaconess Medical Center; R. Greyson, University of Pennsylvania Health System
The adoption of patient portals has been predominantly in ambulatory settings–use for acute and post-acute care is a rapidly emerging field. An Acute Care Patient Portal (ACPP) task force consisting of stakeholders from early adopter institutions has identified opportunities and challenges for use of patient portals to improve the acute and post-acute care patient experience and is creating a road map to advance this evolving field. This interactive panel will discuss key findings and recommendations from the ACPP 2020 conference that engaged a national group of stakeholders. Session participants will discuss and vet recommendations, refine a road map for development, implementation, and investigation, and identify future directions for innovation.
S120: Panel – Problem List 2.0
J. Buchanan, University of Wisconsin School of Medicine and Public Health; W. Galanter, University of Illinois Hospital & Health Sciences System; D. Willett, University of Texas Southwestern Health System; A. Wright, Brigham & Women's Hospital
The value of a comprehensive, concise list of all a patient's active medical conditions to provide accurate "situational awareness" during clinical decision-making has long been recognized. EHRs offer the potential of centralizing the patient problem list, streamlining its maintenance, enhancing its accuracy, and providing flexible viewing to fit clinical information needs. Yet many practices and healthcare organizations today struggle with realizing these benefits. Why is that? What can be done to unlock this potential?
In this panel, we first briefly review the evolution of the problem list in its transition from paper to the electronic health record and summarize current challenges. Then we explore the use of clinical decision support and automated approaches for keeping up an accurate problem list. The recognized benefits of providing context-appropriate, problem-oriented views of relevant medical record information require both EHR technical capabilities and clinically-valid concept maps: we review both, and describe a project to develop the latter in the public domain. We then explore the wide range of potential applications for valid problem list representation of a patient’s health conditions, including clinical decision-making, clinical communication, quality measurement, registries, and clinical and translational research.
S121: Panel – Accelerating Evidence into Practice: AHRQ's Clinical Decision Support Initiative
E. Lomotan, T. Bright, Agency for Healthcare Research and Quality; B. Blumenfeld, RTI International; R. McCready, The MITRE Corporation
The Agency for Healthcare Research and Quality (AHRQ) has launched an ambitious multi-component initiative, focused on patient-centered outcomes research (PCOR), with two primary aims: to accelerate the movement of evidence into practice through clinical decision support (CDS) and to bring CDS closer to becoming more shareable, standards-based, and publicly-available. This panel features the leads of the initiative and its components: 1) The PCOR CDS Learning Network, a cooperative agreement with RTI International, which is building a community for engaging patients, clinicians, CDS developers and implementers, professional societies, and other stakeholders; 2) CDS Connect, a contract with the MITRE Corporation, which is building prototype infrastructure for sharing CDS, including a CDS authoring tool and national CDS repository; 3) research grants to develop new CDS and to scale existing CDS to new sites and systems; and 4) evaluation. Audience members will learn how the PCOR CDS Learning Network is engaging stakeholders to explore and advance the concept of “patient-centered CDS,” will learn how to contribute and use shareable CDS artifacts through CDS Connect, will learn about current grants and funding opportunities to move evidence into practice through CDS on a wider scale, and will learn how to participate in the overall initiative.