AMIA 2020 Annual Symposium Workshops

*Please note: You can only register for one Workshop per timeslot.

Saturday, November 14, 2020

10:00 a.m. – 1:30 p.m. ET

A. Soares, University of Colorado; M. Afshar, Loyola University Chicago; J. Richardson, Veterans Administration Medical Center; C. Moesel, MITRE Corporation; A. Solomonides, Northshore University Health System; E. Barone, Veterans Administration Medical Center; E. Pan, Westat; M. Grasso, University of Maryland; L. Schilling, University of Colorado

Clinical Decision Support (CDS) is a key component of the Promoting Interoperability Program.1 In the National Academy of Medicine report, “Optimizing Strategies for Clinical Decision Support,”2 the authors emphasize that a learning health system (LHS) is driven by “the seamless and rapid generation, processing, and practical application of the best available evidence for the circumstance.” CDS systems require rapid and effective management of clinical knowledge and patient data to identify and deliver the best available evidence to the point of care. This is seldom realized in health systems, partly because the development and uptake of open-source and interoperable CDS tools remain slow.
Resources for CDS developers include human and machine-readable artifacts made publicly available by the Agency for Healthcare Research and Quality (AHRQ) and other federal agencies. The Health eDecisions initiative, for example, led to standards development for knowledge artifacts and CDS services, including the ONC- and CMS-sponsored Clinical Quality Framework to harmonize Health eDecision standards with measurements, and the development of the Fast Healthcare Interoperability Resources ( FHIR ® ) for the electronic exchange of healthcare information. The Clinical Quality Language (CQL) standard also has gained traction and CMS is looking to incorporate this approach into its operational standards.
This workshop will review several open-source CDS tools and resources that currently exist, including CDS Connect, CDS Hooks, and SMART on FHIR. Participants will also be provided with hands-on experience using the tools and resources. The focus will be on a broadly applicable and mature topic – statin use in at-risk adults. Following the statin guidelines, we will discuss and compare the approach and functionality of each CDS tool. Participants will work in a sandbox environment containing CDS tools based on FHIR ® and CQL. The interactive session led by Drs. Soares and Schilling will be facilitated by several “superusers”.

O. Uzuner, George Mason University; M. Yetisgen, University of Washington; D. Demner-Fushman, National Library of Medicine; A. Neveol, CNRS; H. Liu, Mayo Clinic

Advances in Natural Language Processing have made these methods a crucial component of many health applications. These advances have led to incorporation of NLP systems into many real-life health applications, going beyond English language and extending to multiple languages. Thus, the reach and impact of NLP technologies have been growing. This pre-symposium meeting aims to bring together academic, research, and industry participants from around the world with an interest in NLP and provide a venue for sharing recent ground-breaking developments in the field, while also providing an opportunity for mentoring of junior NLP researchers through a doctoral consortium. The sessions planned during this event include: a doctoral consortium, and a session on NLP for COVID-19 that presents methods, systems, and applications.

Y. Aphinyanaphongs, NYU Langone Health; J. Wilt, Oschner Health System; M. Sendak, Duke Institute for Health Innovation; C. Chivers, Penn Medicine

Artificial intelligence based applications are making inroads into multiple industries including healthcare. Due to factors that include a growing pool of skilled data scientists, open source machine learning and deep learning frameworks, AI commoditization by cloud vendors, the availability of digital streams and standards from the electronic health record, and a recognition by leadership of the potential transformational possibilities of artificial intelligence, more hospitals are thinking of or have invested in data science teams to address hospital needs. As of now, there is little guidance on how to build these teams, what the composition of the teams should be, how to interact with leadership on projects, how to vet new projects, how to evaluate impact on projects, resource allocations per project, and other operational considerations. This workshop aims to bring together a team that have been the pioneers in the space and collectively have 3 to 5 years of experience each building data science teams and innovation shops. This team is diverse and operates in different operational structures and yet we are all able to deliver significant projects that impact care and operations. The workshop will expose these inner workings and give guidance and suggestions to other institutions building their own teams or thinking about building a team. It will also highlight best practices in selecting projects and how to move them forward to deliver maximal value.

K. Zheng, University of California Irvine; T. Kannampallil, Washington University in St. Louis; A. Hettinger, MedStart Health Research Institute; V. Patel, New York Academy of Medicine

Most clinical environments resemble a paradigmatic complex system with its dynamic and interactive collaborative work, non-linear and interdependent activities, and uncertainty. Addition of new organizational and systemic interventions, such as health IT, can cause considerable cascading effects in the clinical processes, workflow, and consequently, on throughput and efficiency, which may have implications for clinican burnout and patient safety outcomes. A 2011 IOM report called for a sociotechnical approach for designing and incorporating health IT in clinical settings1. One of the critical aspects of a sociotechnical approach is to understand the progression and evolution of human interactions within a sociotechnical context. In this instructional workshop, we will discuss a set of convergent methodologies for analyzing human interactive behavior both with technology and with other humans or artifacts. These methodologies can help in capturing underlying patterns of human interactive behavior, and provide a mechanism to develop integrative, longitudinal metrics (e.g., metrics related to performance, or errors) for clinical activities for sustained interactive episodes that evolve over time. Such analysis of interactive behavior can also provide significant input for reducing clinician burnout and improving patient safety outcomes through the design of safer and more efficient health IT. In this workshop, we will (a) identify challenges to studying human interactive behavior in complex clinical contexts; (b) discuss new approaches for capturing and analyzing sequences of human interaction using sequential analysis and network-theoretic, time-series based methods; (c) utilize one or more of these techniques to demonstrate their effectiveness as a viable mechanism for developing insights on clinical work activities through hands-on sessions; (d) provide participants hands-on experience in using data collection and data analysis tools; (e) discuss how research on human interaction informs the cognitive design of clinical information systems to improve their usability and support for collaborative team work; and (f) discuss the implications of these techniques for new care delivery models and patient safety initiatives.

J. Herigon, C. Uptegraft, D. Brown, Boston Children's Hospital

This workshop will provide an introduction to the basic steps for developing a clinician-facing SMART on FHIR web-based application. Attendees of all skill levels are invited to participate, though this workshop will primarily focus on the basics of SMART on FHIR app development. After participating in this session, attendees will be able to create a basic functioning SMART on FHIR app and run it against a public FHIR server available via the SMART sandbox. Attendees will also be able to identify challenges and potential pitfalls when working with the FHIR standard, SMART, and vendor APIs.

N. Kutub, A. Das, R. Hekmat, C. Fox, IBM; G. Jackson, IBM, Vanderbilt University Medical Center

Conducting biomedical informatics research as a healthcare technology company poses specific challenges. Research activities to support product design, development, and evaluation must be conducted rapidly, to meet product delivery deadlines and to keep pace with rapid evolution of technology and the corresponding changes in business priorities. To enhance scientific integrity, research can be conducted in collaborations with clients or academic partners, and such collaborations require careful protection of confidential health data and intellectual property. Scientific publications by technology companies are often scrutinized in the peer review process, and thus the underlying science must have a foundation of exceptionally rigorous and reproducible methods.
In this instructional workshop, we will present an overview of lessons learned and best practices for conducting scientifically rigorous research at a rapid, pragmatic, industry pace. This workshop should be of interest to informaticians who are planning to collaborate with industry, wish to transition from an academic to an industry career path, or would like to accelerate their research productivity in any environment. The first 2.5 hours will be an interactive didactic session. A 30-minute working session will allow participants to get direct feedback from panelists on accelerating their research efforts.

O. Gevaert, Stanford University; W. Hsu, University of California Los Angeles; Y. Li, IBM TJ Watson Research Center; A. Rao, University of Michigan; F. Wang, Weill Cornell Medicine; L. Yao, Merck Research Laboratories

Vast amounts of biomedical data are routinely available for patients, ranging from clinical, radiologic and pathologic imaging, and genomic data, spanning multiple biological scales (e.g., patient, tissue, cell) and modalities (e.g., clinical, imaging, -omics). Machine learning is increasingly being used to discover new knowledge and to link and provide clinical decision support for diagnosis, prognosis, and individualized treatment. This collaborative workshop will assemble an interdisciplinary group of experts to discuss challenges, opportunities, methods, and experiences towards integrating multi-scale, multi-modal data towards developing more rigorous clinical decision support tools. The overarching theme focuses on the current state of datasets and methods to bridge diagnostic imaging with clinical and molecular characterizations for accurate diagnosis and treatment selection. The event welcomes participants from all areas of informatics. Topics include extracting and retrieving semantic content from large imaging archives, utilizing imaging features as part of electronic health record-based phenotyping, applying machine learning to discover uncover correlations between images and other biological scales characterizing the significance of evolutionary features derived from images, and translating multi-scale disease models into practice. Our goal is to raise participants' awareness of the opportunities and relevance of imaging and machine learning to other biomedical informatics activities.

E. Bareinboim, A. Ribeiro, Columbia University; M. Adibuzzaman, Regenstrief Center for Healthcare Engineering, Purdue University

Identification of causal effect is one of the key challenges for scientific studies and is a major challenge in biomedical and health sciences. Randomized controlled trials (RCTs) are considered the gold standard for the identification of causal effects. The key idea of RCTs is to control for confounding bias, meaning control for spurious association due to factors extraneous to the relationship between the variables of interest. Despite the ubiquity of RCTs, RCTs have many limitations as they are often not feasible, time consuming or expensive, and often have limited data. On the other end of the spectrum, the recent development of sensor and data capturing technologies has enabled the accumulation of very large amounts of observational data, colloquially termed “big data”. The current generation of biomedical researchers face, therefore, a challenge, to understand how to leverage the vast, but imperfect, amounts of data, and translate it into new insights about causal interventions.

In this workshop, we will start with a chronological description of the major theoretical development in causal inference, starting from randomized controlled trial (Fisher, 1992; Fisher & Yates, 1943), regulatory adaption of randomized trials in health sciences for safety and efficacy evaluation (Greene & Podolsky, 2012), potential outcome framework (Rosenbaum & Rubin, 1983; Rubin, 1974), and recent advances in structural causal models (SCM) (Bareinboim & Pearl, 2016; Pearl, 2009; Pearl, Glymour, & Jewell, 2016). While we will provide a brief description of the major advancements in causal inference, our focus will be on graphical properties of causal inference, more specifically, SCM. We will then discuss basic graphical structures namely, collider, forks and chains along with their statistical properties. Next, we will discuss the concepts of identifiability, in the form of backdoor and front-door adjustments to do mathematical transformation using do-calculus between observational and experimental realities. We will also discuss the practical approaches to compute the causal effect such as inverse probability weighting (IPW) and propensity score matching. Finally, we will discuss a few use cases and provide hands-on experience of the concepts discussed with a web-based software tool, CausalFusion, developed by the CausalAI lab.

10:00 a.m. – 5:30 p.m. ET

H. Hochheiser, University of Pittsburgh; B. Kwon, IBM Research ; D. Wu, University of Cincinnati

The combination of interactive visual displays and statistical and/or machine learning analyses can be powerful tools for managing the interpretation of large and heterogeneous health and biomedical datasets. Researchers and practitioners in the AMIA community have applied these visual analytics techniques to a range of challenges, building systems for researchers, clinicians and patients. Since 2013, the Visual Analytics in Healthcare workshop has been a recurring event, providing a venue for exchanging ideas and developing new collaborations. A full-day workshop at AMIA 2020 will invite interested individuals at all levels of experience in visual analytics to learn from technical paper presentations, rapid “ignite-style” demos, a design challenge, and a panel discussion regarding challenges in the field.

K. Cato, Columbia University; L. Hoeksema, The Ohio State University; R. Freeman, The University of Vermont Health Network; D. Womack, Oregon Health & Science University; A. Jeffery, Vanderbilt University

The widespread implementation of electronic health record (EHR) systems and the application of data science methods enabled the reuse of healthcare data to support clinical decision making. This workshop celebrates the World Health Organization Year of the Nurse and Midwife by analyzing data collected by nursing, which, although often overlooked, is a significant portion of electronic patient data. Research has indicated these nursing data contain strong signals that data scientists can use for predictive analytics and clinical decision support (CDS). Examples of CDS driven by predictive models are in the form of clinical alerts, smart applications, passive dashboards, and other types of informatics interventions. When done effectively, CDS can facilitate evidence-based, high quality, safe patient care. This workshop will guide participants in the development of effective prediction-based CDS tools to facilitate evidence-based, high quality, safe patient care.

A. Rao, University of Michigan

The objective of this collaborative workshop is to assemble an interdisciplinary group of experts to share challenges, methods, and experiences towards sustained development of software tools in the research enterprise. We will discuss the principles of development from a sustainability mindset including the need for investigation of license models and funding/business models adapted towards long-term sustainability of software solutions. We will discuss lessons learned from various groups such as NCI ITCR-SIP (Sustainability and Industry Partnership) Work Group, NCATS CTSA, PCORI Clinical Data Research Network (CDRN), OoD BD2K and NCATS CD2H to understand common processes around the sustainable development of software. We will also discuss the feasible reward mechanisms to motivate long-term sustainability, the needs of a central library that makes federally funded software tools visible and reusable, and the potential resources that can provide industry standard support on code quality control, ecosystem collaboration, security, and dependency hygiene. The goal of this panel is to help open source software (and algorithm) developers to understand the options they can employ to ensure the success of long-term sustainability of open source software (OSS). This will be accomplished by modeling their efforts in concert with a collection of business model archetypes that can serve as sustainability plans for federally funded programs.

2:00 p.m. – 5:30 p.m. ET

B. Alper, EBSCO Information Services, University of Missouri-Columbia; K. Shahin, EBSCO Information Services; M. Michaels, Centers for Disease Control and Prevention; B. Rhodes, Dynamic Content Group, LLC; L. Schilling, University of Colorado Anschutz Medical Campus

Evidence-based practice, also called evidence-based medicine (EBM), requires the use of research results to inform care. Fast Healthcare Interoperability Resources (FHIR) provides a standard for electronic exchange of healthcare information to achieve interoperability but FHIR version 4.0.1 does not support communication of research results. The FHIR Resources for Evidence-Based Medicine Knowledge Assets (EBMonFHIR) project is extending FHIR with an Evidence resource to communicate explicit descriptions of populations, exposures, measured variables, statistics, and certainty of these statistics. The foundational architecture and functional model for machine-interpretable expression of evidence is evolving and changes with engagement from implementers exploring use cases for reporting summaries of populations samples, reporting research findings, reporting appraisal and summarization of research findings, and reporting recommendations or other methods for translating research to practice. Open participation is encouraged by anyone translating data to knowledge to action and is described at https://confluence.hl7.org/display/CDS/EBMonFHIR . This workshop will introduce tools that can be used to create Evidence resources to report research findings in JSON formats and provide participants with hands-on experience creating Evidence resources. Participants are encouraged to bring examples of research findings they wish to report in computable format. Following direct experience in creating Evidence resources we will facilitate group interaction to inform improvements in the Evidence resource architecture, the instructions for creating Evidence resources and the tools for facilitating Evidence resource creation. This workshop will provide AMIA members a unique opportunity for live in-person engagement with a broad-based collaborative standards development effort that is mobilizing computable biomedical knowledge and making biomedical evidence interoperable.

R. Schreiber, Geisinger Holy Spirit, Geisinger Commonwealth School of Medicine; A. Wright, A. McCoy, Vanderbilt University Medical Center; D. Sittig, University of Texas Health Science Center; M. Grasso, University of Maryland School of Medicine

Analysis of current research in clinical decision support (CDS) and alerting reveals a wide disparity in definitions, metrics, and efficacy of CDS. Investigations report multiple different metrics which makes comparative research difficult. The result is that CDS efficacy is difficult to measure and difficult to compare across vendor products and institutions. There is no coherent consensus regarding terms such as acceptance rate, override rate, alert or CDS efficiency or effectiveness. Particularly troublesome are definitions of “override” and “ignore”. Timing metrics, such as think time2 and dwell time, which estimate how long clinicians contend with alerts, have different definitions and are difficult to assess and compare. There are numerous alert metrics, including total alerts, or alerts expressed as ratios, such as alerts per provider, patient, order session, specific order, or specialty. Efforts to assess efficiency and efficacy of CDS include metrics such as savings/alert. The first goal of this workshop is to provide researchers and operations personnel with better CDS definitions and metrics. The second goal is to develop strategies to justify alert appropriateness, problem identification, benefit determination, appropriateness, timing, and workflow consistency. This includes coming to a consensus regarding build principles, such as the CREATOR proposal which stresses the need for alerts to be Consistent with organizational strategy and principles; Relevant and timely; Evaluable; Actionable; Transparent; Overridable; and Referenced. The third goal focuses on alert burden, which is one of the chief causes of clinician burnout with the electronic health record, and which will discuss the need to correct design inconsistencies, tailor alerts to target audiences, engage directly with impacted clinicians, standards for inclusion, and periodic review for appropriateness. A fourth goal is to determine remaining fundamental questions applicable to CDS research.

H. Xu, The University of Texas Health Science Center at Houston; O. Patterson, S. Duvall, University of Utah

Natural language processing (NLP) technologies, which can unlock information embedded in clinical narratives, have received great attentions in the medical domain. While early clinical NLP systems are often rule-based, recent clinical NLP tools are often based on machine learning algorithms. More recently, deep learning (DL) techniques have begun to dominate in NLP because of their simplicity (no need for hand-crafted features), efficient processing (assuming dedicated, massively parallelized hardware), and state-of-the-art results (on a plethora of tasks). In our recent review article published in JAMIA, we clearly identified a trend of shifting to deep learning approaches in clinical NLP research and development (Wu et al. 2020). Nevertheless, it is still challenging for end users to select, apply, or extend deep learning-based NLP methods or tools for their specific clinical applications. In fact, there is a lack of best practices for building successful deep learning-based NLP solutions in the medical domain.
In this workshop, we would like to introduce methods, tools, and best practices on building deep learning-based NLP solutions for clinical applications. We will introduce basic concepts of clinical NLP, illustrate state-of-the-art deep learning methods and available tools, and demonstrate important applications of deep learning in diverse clinical NLP tasks. We plan to use lectures, demonstrations and hands-on exercises to cover the basic knowledge/tools and use case studies to illustrate important trade-offs in the design, development, and implementation of deep learning based clinical NLP applications. Each instructor has over 10 years of experience in clinical NLP research and application and they will share their recommendations in building successful NLP applications in clinical research and operation.

P. Payne, A. Lai, Washington University; J. Guinney, J. Eddy, Sage Bionetworks

The conduct of clinical and translational research programs that span traditional organizational boundaries has increasingly become the norm. Supporting data, information, and knowledge management needs in such collaborative settings requires new approaches to information architecture, technology delivery, and governance. Given the ubiquity of cloud computing infrastructure and capabilities, such environments offer a number of potential benefits in terms of supporting the preceding use cases and needs. These benefits include scalability, elasticity, intrinsic security controls, and overall flexibility in terms of enabling or end-user facing technologies and frameworks. However, the use of cloud computing in this context differs substantially from more traditional on-premises approaches to research computing, which remain common in the academic and healthcare delivery system settings. Such differentiation includes dimensions such as: 1) systems architecture; 2) data architecture; 3) technology delivery and management; 4) cybersecurity; and 5) governance. Building upon the experiences of the NCATS-funded Center for Data 2 Health (CD2H), this workshop will provide attendees with a primer on the preceding dimensions, as they apply to the use of cloud computing to support and enable both single- and multi-site clinical and translational research programs. In addition, the workshop will engage in a comparative analysis of such technologies in comparison to traditional, on-premises approach to research computing, thus equipping attendees with the knowledge needed to select and promote appropriate computing approach’s given their individual- or project-level needs. Finally, the workshop will include a collaborative exercise in which participants will apply the preceding content to a prototypical project design process, resulting in both a systems architecture model, as well as a business case for presentation to relevant decision-makers.

S. Abedian, E. Sholle, Y. Zhang, Weill Cornell Medicine

This instructional workshop was inspired by the current approaches taken at Weill Cornell Medicine's Research Informatics team, supporting the clinical studies utilizing the Social Determinants of Health (SDoH) data by creating a comprehensive SDoH dataset - Variables Affecting Care and Community in New York (VACCINe). This workshop provides an introduction to augmenting clinical data with publicly available SDoH datasets to support the research enterprise at large. In this workshop, we cover a brief introduction to SDoH and the current landscape of SDoH data collection methods. We review the current approaches to obtain SDoH data from federal and local government agencies, as well as individual SDoH data. During the interactive part, the audience learns how to merge and harmonize these publicly available datasets and is able to recreate the VACCINe dataset. We then introduce methods of large-scale geocoding for the patient population. Lastly, we will demonstrate how researchers can use the OMOP Common Data Model combined with the SDoH dataset, which is created at the workshop.

K. Wagholikar, J. Klann, Harvard Medical School; M. Mendis, Partners Healthcare; S. Murphy, Harvard Medical School

This half-day workshop is a hands-on introduction to the ‘Informatics for Integrating Biology and the Bedside’ (i2b2) data platform. It will provide an overview of the platform functionality and discuss approaches to install the platform and to import data.
The workshop suited for researchers, clinicians, IT programmers, educators, leaders in healthcare participating in projects in health information technology. Intermediate and advanced users of i2b2 can benefit from a recap of the fundamentals and learn about new tools and approaches using i2b2. i2b2 has been deployed at over 200 institutions across the world to enable researchers to identify and analyze patient cohorts for clinical studies. The aim of the workshop is to introduce i2b2 to novice users with hand-on training, to educate the research community about the i2b2 data tooling, and to evolve good practice guides for i2b2.

M. Ozkaynak, University of Colorado; K. Unertl, Vanderbilt University Medical Center

The workshop’s purpose is to appraise current workflow tools, frameworks and methodologies through the lenses of data science and artificial intelligence. New opportunities, challenges and future directions will be discussed. Participants will be engaged in hands-on experience through exploration of real-world case studies and participatory exercises.

J. Starren, N. Rothrock, Northwestern University; D. Meeker, University of Southern California; T. Nelson, Northwestern University

Patient-reported outcomes (PROs) represent health information directly reported by the patient receiving care. There are increasing efforts to incorporate PROs in clinical practice as a component of quality measurement, quality improvement, symptom management, and patient engagement. To enable this, organizations are attempting to integrate PROs into their electronic health record (EHR) environment. This integration presents numerous sociotechnical challenges including: new technology to support administration and scoring of computer adaptive tests, implementation of significant workflow changes for patients and clinicians, and management of the expectations of the many stakeholders impacted. This workshop will present challenges, strategies, and specific tools that can be employed to increase success. Among the tools is a clinical implementation planning process that leverages a multi- stakeholder survey instrument to identify critical workflow issues and organizational challenges. We will also review common pitfalls and solutions. The workshop will conclude with strategies for post-implementation evaluation and monitoring.

Sunday, November 15, 2020

10:00 a.m. – 1:30 p.m. ET

A. Solomonides, NorthShore University HealthSystem; K. Fultz Hollis, Oregon Health & Science University; S. Rosenbloom, Vanderbilt University; L. Sheets, University of Missouri; J. Smith, AMIA

What is data governance? As we shall explore in this workshop, it comprises the regulatory principles, policies and strategies adopted, the functions and roles that must be created to implement these policies and strategies, and the consequent architectural designs that provide both a home for the data and, less obviously, an operational expression of policies in the form of controls and audits. The reasons for the extraordinary measures taken by institutions to protect the data lie in the value of that data as a strategic asset and in the internal and external threats to the data. The workshop will include a presentation of background knowledge of principles (especially of recent developments), and an opportunity to role-play various data governance-related positions in an organization. Discussion of principles and of the simulated experience will complete the program.

P. Lynch, National Library of Medicine; Y. Wang, National Library of Medicine; A. Kanduru, National Library of Medicine; X. Luan, Y. Sedinkin, L. Amos, C. McDonald, National Library of Medicine; J. Buchanan, University of Wisconsin;

Fast Health Interoperability Resources (FHIR) is a web-based, easy to implement, standard for exchanging healthcare information electronically. Apple, Google, Microsoft, large health IT companies, federal agencies (ONC, CMS, Veterans Administration) and Big Pharma have embraced it, as has NIH (NIH, NOT-OD-19-122). NIH sees it a possible pathway to their long-term goal for reusable and interoperable data (NIH, Data Sharing Guidance 2003); FHIR providing a mechanism for accessing and leveraging routine clinical data. Last fall, NIH solicited public feedback on the use of HL7 FHIR for capturing and sharing clinical data for research purposes (NIH, NOT-OD-19-150). AMIA supported the use of FHIR for these purposes.
NLM has a longstanding interest in the support and development of clinical terminology and message standards. Within NLM’s Lister Hill Center, we have developed apps and tools for implementing many of FHIR’s capabilities. Specifically, tools for 1) detecting errors and facilitating data entry; 2) authoring FHIR compliant input forms, rendering those forms in real time 3) generating flowsheets on selected patents and 4) finding data across medical records for research purposes. This workshop will describe these tools, explain how many of FHIR's capabilities are implemented and demonstrate their use for research and clinical use. We will also describe and discuss problems we have encountered and potential fixes for them. The tools we developed can be used in EHRs, PHRs, and research systems and can be downloaded and/or accessed via https://lhcforms.nlm.nih.gov/. All tools are open source; most are available on GitHub and some also function as SMART on FHIR apps.

N. Shimpi, Marshfield Clinic Research Institute; K. Nguyen, University of Florida, College of Pharmacy; M. Durkin, Washington University School of Medicine in St. Louis; R. Mishra, University of Washington; S. Hebbring, Marshfield Clinic Research Institute

Precision medicine has been identified as an important element in tailoring healthcare management for improved patient outcomes 1 . Despite the definitions of clinically important pharmacogenetic variants with the potential to increase susceptibility for adverse drug events, (as demonstrated in large population-based studies) knowledge of what fraction of real-world patients that are given medications that may be contraindicated by their pharmacogenomics profile is currently unknown 2,3 . Recognizing strategies to improve integration of pharmacogenomics in clinical practice is of paramount importance. Adoption of innovative, evidence-based practices implemented in a learning health system environment with the support of informatics platforms, has created opportunities to develop clinical decision support models and to improve integrated, collaborative, patient-centric care.
The objective of this workshop is to equip attendees with practical competencies surrounding emerging concepts of pharmacogenomics and application of informatics to models of care across dental and medical practice. This introductory, interactive workshop brings together informatics researchers who leverage clinical and research data at point-of-care settings. The workshop covers the following broad areas:
a. Practice-based research in the area of informatics and pharmacogenetics service: System-level informatics approaches with applicability to pharmacogenetics
b. Using clinical decision support system to optimize dental antibiotic prescribing practices: Development of clinical decision support systems and mobile applications in electronic dental records
c. Precision medicine and applicability to global oral health disorders: Importance of oral precision medicine and its application to oral diseases and conditions
d. Estimating the efficacy of pharmacogenomics over a lifetime: Pharmacogenomic variants in clinical population and drug prescription and impact on adverse drug reactions

J. Faulkenberry, The Children's Hospital of Philadelphia; S. Craig, The Children's Hospital of Philadelphia; F. Holl, Neu-Ulm University of Applied Sciences; H. Fraser, Brown University; J. Ruff, Case Western Reserve University, MetroHealth

Over the last two decades, global health informatics (GHI) resources have been developed to support health care in low- and middle-income (LMIC) countries. Many products, including free and open source (FOSS) software, software-as-a-service (SaaS), digital health toolkits and maturity model assessment strategies have been developed and are available (1). However, some of these may not be well known or may be supported by varying business models. The maturity level of the local, regional, and national health systems also play a key role in finding the best fit between a GHI resource and the clinical context. This results in a still daunting task to discover, evaluate and implement tools and resources that work across the spectrum of GHI contexts. A practical hands-on approach is needed to guide informaticians in identifying the best digital health resource to yield a successful, sustainable, and scalable implementation for their use-case. This workshop continues the work from the 2011 Public Health Informatics Workshop: Building the Foundations of an Informatics Agenda for Global Health (2), and also the 2019 AMIA Symposium Workshop: Global Health Informatics Collaborations. Both workshops identified the need for increased pragmatic, effective collaborations among the GHI community to share resources and knowledge. Workshop participants of the 2019 AMIA Symposium expressed a need for guidance with GHI projects in areas with nascent and emerging health information systems. The content of this workshop will increase awareness of existing GHI resources by providing hands-on experience via demonstrations in practical, interactive sessions. Participants will discuss the trade-offs of systems that are open vs more closed or controlled, and how tools for disease surveillance or research differ from those for clinical care. Through these discussions, participants will gain practical insights to choosing, implementing, and maintaining a GHI resource.

P. Hsueh, Viome Inc.

In the past decade, we have witnessed tremendous progress in the use cases of incorporating electronic patient-reported outcomes (ePROs) and citizen science-inspired approaches in the fields of health system design and biomedical research. These use cases have enabled the curation of patient-generated health data (PGHD) to accelerate scientific discovery and advance patient-centered care. Many patient-powered research networks such as Patient-Centered Outcomes Research Institute (PCORI)-sponsored ones as well as informatics tools and platforms have been established to expand its scope and depth. In the past few years the AMIA Consumer and Pervasive Health Informatics Working Group has co-organized with academic and industrial leaders, as well as patient advocates, a series of events such as AMIA 2015 panel on PGHD, AMIA 2017 Policy Invitational, AMIA 2018 Citizen Science Workshop and ePRO Pre-Symposium and AMIA 2019 ePRO-Citizen Science joint pre-symposium. These events brought together like-minded researchers, practitioners, and patients to identify opportunities and challenges. One common observation is the lack of rigor in research methods and protocols for patient-involved research to further scale up the efforts of adopting real-world evidence for precision health application and clinical practice.
The goal of this workshop is to thus to take a step further to dive into case studies (e.g., community-based research/trial protocols, patient-centered participatory design, platforms for enabling scalable evidence generation and personalized recommendation) to learn from the informaticists and considered how to bridge the gap between research and practice. In particular, this workshop will focus on the best practice examples with an in-depth discussion on the research methods and protocols in support of transforming patient voice into valid research contributions from an informatics perspective. Drawing on the national and AMIA leadership initiatives and previously identified challenges, this workshop will enable a robust dialogue about how the informatics community can innovatively engage in ePRO and citizen science to meaningfully address informatics challenges in real-world evidence use. Workshop participants will join a panel of leading experts and stakeholders (including patient representatives) for scaling up patient-involved research and system design and creating a vision and strategy within informatics.

S. Rehman, Phoenix VA Health Care Systems, University of Arizona College of Medicine; H. Abbaszadegan, Phoenix VA Health Care Systems, University of Arizona College of Medicine; P. Dykes, Harvard Medical School, AMIA, Brigham and Women’s Hospital

Great leaders are great negotiators, they resolve seemingly intractable disputes and yet enhance working relationships. Their negotiation and communication skills determine their effectiveness. Physicians and non-physician members of AMIA are expected to negotiate with a vast array of third parties, including healthcare system governing boards, leaders in the C-suites, patients, end-user consumers, government, health plans, insurance companies, EMR vendors, and pharmaceutical companies. Additionally, negotiation skills are an essential competency and requirement for board certification for physicians (ABPM and ABP), yet one may not find any session on this topic in AMIA Symposium. It is time for all informatics professionals to be trained in effective negotiation skills. Law, business, and public policy schools offer classes in negotiation.
The ability to negotiate requires a collection of interpersonal and communication skills used together to bring about a desired result. It is based on exploring underlying interests and positions to bring parties together in a constructive way. Effective negotiators use innovative thinking to create lasting value and forge strong professional relationships. They take a deep dive in to what is behind the opponent and their own positions that may not seem logical at first but essential to understand the issues/ideas behind the problem.
The 3-hour highly interactive session provides evidence-based tools & interventions for identifying individual communication preferences, delivery methods, conflict resolution styles as well identifying best practices and “best alternative to a negotiated agreement” (BATNA).
The interactive session involves exercises and activities that will allow the participants to discover, learn and practice the negotiation skills and tools. We will also be including data from the literature review, needs assessment, and Informatic tools that are useful in negotiation. We also have incorporated the outcome measures used in data science for success factors upon completion. The interventions/tools discussed/practiced in the workshop are evidence based and supporting data will be provided.

R. Schreiber, Geisinger Holy Spirit, Geisinger Holy Spirit, Geisinger Commonwealth School of Medicine; J. Hollberg, Emory University; P. Fu, City of Hope; N. Safdar, Emory University

There are now 1,868 physicians board-certified in clinical informatics and 35 accredited Clinical Informatics Fellowship Programs.[1] There are also many non-board certified/eligible practitioners who need training in state-of-the-art applied clinical informatics. AMIA is uniquely suited to be the academic home for this community, because it provides a combination of personal experience with firm grounding in evidence-based biomedical informatics literature and theory, foundational knowledge, and proven best practices. A major part of that support is outreach to Chief Medical Information Officers (CMIOs) and those in similar roles who are charged with leading informatics change within their organizations. More than 350 individuals have attended the annual CMIO workshop since 2011, some more than once, ranging from seasoned CMIOs of large systems to those who are just beginning their applied clinical informatics careers. Yearly surveys confirm the need for an on-going workshop. In the 2017 workshop, of over 80 participants, 91% responded they desired to attend the workshop again; the most requested topics were practical leadership skills and guidance on change management.
The 2020 CMIO Workshop will focus on leadership development, including didactic and small group exercises. We will focus on successful negotiation strategies, change management processes, and interactions with other C-suite and key stakeholders within a single hospital as well as for large health systems. We will then apply the leadership and change management skills to case-based discussion of current contentious yet practical topics for CMIOs including management of problem lists, how to implement open notes, optimization projects, and best practices regarding EHR documentation and scribes. Didactic presentations will be integrated with structured group discussions. Participants in the workshop will engage each other during the group discussions to practice the concepts and teachings from the didactic sessions.
[1] Clinical Informatics Fellowship Programs. Found at: https://www.amia.org/membership/academic-forum/clinical-informatics-fellowships Accessed 5 March 2020.


Premier Sponsor


Sponsors



Student Paper Prize     AMIA Working Group Sponsors