Times are displayed in (UTC-04:00) Eastern Time (US & Canada) Change
Custom CSS
double-click to edit, do not edit in source
11/18/2025 |
9:45 AM – 11:00 AM |
M102
S73: FHIR App Competition
Presentation Type: FHIR App Showcase
SmartHF: A patient-centered FHIR-based web application for medication optimization
2025 Annual Symposium On Demand
Presentation Time: 09:45 AM - 09:53 AM
Heart failure (HF) is the leading hospital discharge diagnosis among older US adults, with 40% readmitted within a year after initial admission. This causes substantial, often avoidable costs, as hospitalizations account for 70% of HF management expenses. A key reason for readmission is the failure to optimize chronic medications.
Despite widespread knowledge that guideline-directed medication therapy (GDMT) reduces hospital admission and mortality in HF with reduced ejection fraction (HFrEF, also referred to as systolic heart failure), medications are often not optimized in clinical practice. The American College of Cardiology recommends the use of electronic health records (EHR) to reduce errors, improve decision support, and facilitate GDMT for HFrEF. Yet, currently, there are no effective patient-centered EHR tools that can assess clinical characteristics and provide adaptive recommendations to optimize GDMT.
The FHIR-based solution, named SmartHF, is a patient-centered clinical design support tool. SmartHF is an adaptive web application to facilitate GDMT optimization and is being studied in a randomized controlled trial funded by AHRQ. Participants are randomized to the SmartHF intervention or no intervention before a clinic visit with their HF provider. The SmartHF intervention walks participants through the Epic MyChart OAuth, gathers EHR data using FHIR API endpoints, and analyzes the data using a computable algorithm designed to provide medication optimization recommendations in HFrEF. SmartHF delivers the participant with the medication recommendations in the web app, sends them an email with the recommendations, and asks the participant to discuss these recommendations with their doctor at their upcoming appointment.
Speaker:
Michael Dorsch, PharmD, MS
University of Michigan
2025 Annual Symposium On Demand
Presentation Time: 09:45 AM - 09:53 AM
Heart failure (HF) is the leading hospital discharge diagnosis among older US adults, with 40% readmitted within a year after initial admission. This causes substantial, often avoidable costs, as hospitalizations account for 70% of HF management expenses. A key reason for readmission is the failure to optimize chronic medications.
Despite widespread knowledge that guideline-directed medication therapy (GDMT) reduces hospital admission and mortality in HF with reduced ejection fraction (HFrEF, also referred to as systolic heart failure), medications are often not optimized in clinical practice. The American College of Cardiology recommends the use of electronic health records (EHR) to reduce errors, improve decision support, and facilitate GDMT for HFrEF. Yet, currently, there are no effective patient-centered EHR tools that can assess clinical characteristics and provide adaptive recommendations to optimize GDMT.
The FHIR-based solution, named SmartHF, is a patient-centered clinical design support tool. SmartHF is an adaptive web application to facilitate GDMT optimization and is being studied in a randomized controlled trial funded by AHRQ. Participants are randomized to the SmartHF intervention or no intervention before a clinic visit with their HF provider. The SmartHF intervention walks participants through the Epic MyChart OAuth, gathers EHR data using FHIR API endpoints, and analyzes the data using a computable algorithm designed to provide medication optimization recommendations in HFrEF. SmartHF delivers the participant with the medication recommendations in the web app, sends them an email with the recommendations, and asks the participant to discuss these recommendations with their doctor at their upcoming appointment.
Speaker:
Michael Dorsch, PharmD, MS
University of Michigan
Michael
Dorsch,
PharmD, MS - University of Michigan
Fenotyper: FHIR-enabled Phenotype Definitions
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 09:53 AM - 10:01 AM
Background
Precise, computable phenotype definitions from real-world clinical data are essential for large-scale clinical, translational research and quality improvement. Traditional methods rely on manual expert-driven rule authoring and validation. Advances in Large Language Models (LLMs), including Retrieval-Augmented Generation (RAG), offer promise for accelerating phenotype creation through human-AI alignment.
Objective
To design and evaluate a FHIR-integrated system that generates auditable, high-quality phenotype definitions from plain-language requirements for cohort-discovery in research and quality improvement.
Methods
Our system accepts natural language input (e.g., “I am studying children with congenital heart disease and ADHD”) and retrieves candidate SNOMED-CT codes via RAG. It generates phenotype definitions previously implemented in literature from the OHDSI library. Each concept is classified as include, exclude, or irrelevant, with a transparent, human-readable logic_reasoning field. The resulting definition--comprising terms, codes, and inclusion criteria--serves as shared memory for iterative refinement. Users can propose concepts, override decisions, or supply new reasoning examples, incorporated via in-context learning. When exporting, the system calls a FHIR API and ConceptMap resources to translate SNOMED-CT-based definitions to any FHIR ontology.
Results
Preliminary use cases show the system produces structured phenotype definitions with traceable logic, enabling efficient expert iteration. For instance, exclusion of “variant angina” can be revised in real time, with regeneration of code lists and reasoning. This reduces redundant coding, enhances transparency, and supports context-specific definitions.
Conclusion
We present a novel FHIR-enabled application for phenotype definition that unifies LLM-reasoning with user-oversight, lowers entry barriers, and supports reproducible research. Future work will evaluate performance across diverse phenotypes and pipelines.
Speaker:
Yona Kleinerman, Pursuing a Bachelors of Science in Biomedical Engineering
Health Systems Engineering & Informatics Laboratory, Dr. Vignesh Subbian, University of Arizona, Tucson
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 09:53 AM - 10:01 AM
Background
Precise, computable phenotype definitions from real-world clinical data are essential for large-scale clinical, translational research and quality improvement. Traditional methods rely on manual expert-driven rule authoring and validation. Advances in Large Language Models (LLMs), including Retrieval-Augmented Generation (RAG), offer promise for accelerating phenotype creation through human-AI alignment.
Objective
To design and evaluate a FHIR-integrated system that generates auditable, high-quality phenotype definitions from plain-language requirements for cohort-discovery in research and quality improvement.
Methods
Our system accepts natural language input (e.g., “I am studying children with congenital heart disease and ADHD”) and retrieves candidate SNOMED-CT codes via RAG. It generates phenotype definitions previously implemented in literature from the OHDSI library. Each concept is classified as include, exclude, or irrelevant, with a transparent, human-readable logic_reasoning field. The resulting definition--comprising terms, codes, and inclusion criteria--serves as shared memory for iterative refinement. Users can propose concepts, override decisions, or supply new reasoning examples, incorporated via in-context learning. When exporting, the system calls a FHIR API and ConceptMap resources to translate SNOMED-CT-based definitions to any FHIR ontology.
Results
Preliminary use cases show the system produces structured phenotype definitions with traceable logic, enabling efficient expert iteration. For instance, exclusion of “variant angina” can be revised in real time, with regeneration of code lists and reasoning. This reduces redundant coding, enhances transparency, and supports context-specific definitions.
Conclusion
We present a novel FHIR-enabled application for phenotype definition that unifies LLM-reasoning with user-oversight, lowers entry barriers, and supports reproducible research. Future work will evaluate performance across diverse phenotypes and pipelines.
Speaker:
Yona Kleinerman, Pursuing a Bachelors of Science in Biomedical Engineering
Health Systems Engineering & Informatics Laboratory, Dr. Vignesh Subbian, University of Arizona, Tucson
Yona
Kleinerman,
Pursuing a Bachelors of Science in Biomedical Engineering - Health Systems Engineering & Informatics Laboratory, Dr. Vignesh Subbian, University of Arizona, Tucson
Patient Artificial Intelligence Guided E-messages (PAIGE)
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:01 AM - 10:09 AM
Effective patient-provider communication is strained by the overwhelming volume of electronic messages, a major driver of clinician burnout. A significant portion of these messages are incomplete, leading to frustrating back-and-forth exchanges that delay care and increase administrative workload. For instance, a patient's simple request for a UTI prescription might omit critical symptoms like fever, forcing clinicians to spend valuable time seeking clarification.
To address this, we developed Patient Artificial Intelligence Guided E-messages (PAIGE), a Large Language Model (LLM)-powered chatbot integrated directly into the patient portal as a SMART on FHIR application. Unlike tools that help providers draft replies, PAIGE focuses on the source of the problem by helping patients compose a more comprehensive initial message. When a patient starts a message, PAIGE engages them in a brief, guided conversation, asking targeted follow-up questions to gather essential details. It then synthesizes this information into a clear, complete message for the patient to review and send.
PAIGE’s backend uses a Retrieval-Augmented Generation (RAG) framework with GPT-4o on a protected Azure platform at VUMC, querying a trusted knowledge base of nurse triage protocols and VUMC-specific protocols for follow-up questions on common conditions. It has the capability to securely access patient data (e.g., allergies, medications) via FHIR. Importantly, PAIGE does not provide medical advice; its sole purpose is to improve the quality of patient-provider communication, thereby accelerating care and reducing clinician burden.
Speaker:
Siru Liu, PhD
Vanderbilt University Medical Center
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:01 AM - 10:09 AM
Effective patient-provider communication is strained by the overwhelming volume of electronic messages, a major driver of clinician burnout. A significant portion of these messages are incomplete, leading to frustrating back-and-forth exchanges that delay care and increase administrative workload. For instance, a patient's simple request for a UTI prescription might omit critical symptoms like fever, forcing clinicians to spend valuable time seeking clarification.
To address this, we developed Patient Artificial Intelligence Guided E-messages (PAIGE), a Large Language Model (LLM)-powered chatbot integrated directly into the patient portal as a SMART on FHIR application. Unlike tools that help providers draft replies, PAIGE focuses on the source of the problem by helping patients compose a more comprehensive initial message. When a patient starts a message, PAIGE engages them in a brief, guided conversation, asking targeted follow-up questions to gather essential details. It then synthesizes this information into a clear, complete message for the patient to review and send.
PAIGE’s backend uses a Retrieval-Augmented Generation (RAG) framework with GPT-4o on a protected Azure platform at VUMC, querying a trusted knowledge base of nurse triage protocols and VUMC-specific protocols for follow-up questions on common conditions. It has the capability to securely access patient data (e.g., allergies, medications) via FHIR. Importantly, PAIGE does not provide medical advice; its sole purpose is to improve the quality of patient-provider communication, thereby accelerating care and reducing clinician burden.
Speaker:
Siru Liu, PhD
Vanderbilt University Medical Center
Siru
Liu,
PhD - Vanderbilt University Medical Center
Mere Medical: An open-source personal health record
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:09 AM - 10:17 AM
Objectives To enable patients to aggregate their fragmented health records across multiple health systems while retaining full ownership and control.
Materials and Methods We developed Mere Medical as a progressive web application using HL7 FHIR patient access APIs with SMART-on-FHIR authentication. The application employs a local-first architecture, storing all records directly on the user’s device. Core features include unified health timelines, consolidated summary views (e.g., immunizations), cross-system document search, laboratory value trending, and provider note access.
Results The application successfully aggregates records from major EHRs (e.g., Epic, Cerner) enabling users to digest their comprehensive histories spanning multiple health systems, better manage their chronic conditions over time, and tell a cohesive medical story to each new physician they encounter in their health care journey. Additionally, Mere illustrates the feasibility of building a patient-owned health record using patient access APIs.
Discussion We developed Mere Medical with three core principles: patient autonomy, data portability, and transparency. In contrast to solutions that focus on institutional interoperability, it shows how patient-directed access can complement national initiatives and support equitable information sharing. The open-source model enables collaborative development and shows that third-party developers can use existing standards to overcome barriers to patient access.
Conclusion Mere Medical demonstrates an open-source approach to health record aggregation building on FHIR standards, offering a foundation for future research and adoption in patient-centered care. Source code (MIT-licensed): https://github.com/cfu288/mere-medical
Speaker:
Christopher Fu, D.O.
Westchester Medical Center
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:09 AM - 10:17 AM
Objectives To enable patients to aggregate their fragmented health records across multiple health systems while retaining full ownership and control.
Materials and Methods We developed Mere Medical as a progressive web application using HL7 FHIR patient access APIs with SMART-on-FHIR authentication. The application employs a local-first architecture, storing all records directly on the user’s device. Core features include unified health timelines, consolidated summary views (e.g., immunizations), cross-system document search, laboratory value trending, and provider note access.
Results The application successfully aggregates records from major EHRs (e.g., Epic, Cerner) enabling users to digest their comprehensive histories spanning multiple health systems, better manage their chronic conditions over time, and tell a cohesive medical story to each new physician they encounter in their health care journey. Additionally, Mere illustrates the feasibility of building a patient-owned health record using patient access APIs.
Discussion We developed Mere Medical with three core principles: patient autonomy, data portability, and transparency. In contrast to solutions that focus on institutional interoperability, it shows how patient-directed access can complement national initiatives and support equitable information sharing. The open-source model enables collaborative development and shows that third-party developers can use existing standards to overcome barriers to patient access.
Conclusion Mere Medical demonstrates an open-source approach to health record aggregation building on FHIR standards, offering a foundation for future research and adoption in patient-centered care. Source code (MIT-licensed): https://github.com/cfu288/mere-medical
Speaker:
Christopher Fu, D.O.
Westchester Medical Center
Christopher
Fu,
D.O. - Westchester Medical Center
Lifestyle Intervention, Learning & Links for You
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:17 AM - 10:25 AM
Pregnancy is a vulnerable period where one in five birthing parents develop postpartum depression (PPD) within a year of childbirth. Despite evidence for lifestyle interventions to mitigate PPD risk, challenges for scalable delivery, access and implementation persist. We present Lilly (Lifestyle Intervention, Learning & Links for You), a FHIR-enabled digital health platform offering individually tailored lifestyle interventions and resources, designed to support pregnancy health and prevent PPD. With user consent, Lilly connects to electronic health records (EHRs) through FHIR, using resources including Patient, EpisodeOfCare, Condition, MedicationRequest, Observation, Procedure, and Encounter. This automated data pipeline enables LillyAI, Lilly’s generative AI agent designed to support pregnancy health, to assess users’ health and social characteristics relevant to PPD at regular cadence. Lilly will (1) recommend personalized, evidence-based lifestyle interventions (i.e., exercise, diet, sleep, and mindfulness), aligned with the latest recommendations from American College of Obstetricians and Gynecologists; (2) provide user-friendly reading materials summarized from peer-reviewed scientific articles, (3) identify neighborhood resources that meet users’ social needs; and (4) foster social networks among users with similar health backgrounds to strengthen social support. Lilly aims to fill critical gaps in the long-overlooked care continuum around the delivery of health education and social support for pregnancy and new motherhood, while engineering novel, generalizable frameworks for integrating patient-generated health data (PGHD) with EHR to promote positive health impacts. Currently, Lilly’s prototype is live and undergoing user-centered design and evaluation, with the vision of integrating into clinical pathways for PPD prevention in patients with low to moderate risk.
Speaker:
Yiye Zhang, PhD
Weill Cornell Medicine
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:17 AM - 10:25 AM
Pregnancy is a vulnerable period where one in five birthing parents develop postpartum depression (PPD) within a year of childbirth. Despite evidence for lifestyle interventions to mitigate PPD risk, challenges for scalable delivery, access and implementation persist. We present Lilly (Lifestyle Intervention, Learning & Links for You), a FHIR-enabled digital health platform offering individually tailored lifestyle interventions and resources, designed to support pregnancy health and prevent PPD. With user consent, Lilly connects to electronic health records (EHRs) through FHIR, using resources including Patient, EpisodeOfCare, Condition, MedicationRequest, Observation, Procedure, and Encounter. This automated data pipeline enables LillyAI, Lilly’s generative AI agent designed to support pregnancy health, to assess users’ health and social characteristics relevant to PPD at regular cadence. Lilly will (1) recommend personalized, evidence-based lifestyle interventions (i.e., exercise, diet, sleep, and mindfulness), aligned with the latest recommendations from American College of Obstetricians and Gynecologists; (2) provide user-friendly reading materials summarized from peer-reviewed scientific articles, (3) identify neighborhood resources that meet users’ social needs; and (4) foster social networks among users with similar health backgrounds to strengthen social support. Lilly aims to fill critical gaps in the long-overlooked care continuum around the delivery of health education and social support for pregnancy and new motherhood, while engineering novel, generalizable frameworks for integrating patient-generated health data (PGHD) with EHR to promote positive health impacts. Currently, Lilly’s prototype is live and undergoing user-centered design and evaluation, with the vision of integrating into clinical pathways for PPD prevention in patients with low to moderate risk.
Speaker:
Yiye Zhang, PhD
Weill Cornell Medicine
Yiye
Zhang,
PhD - Weill Cornell Medicine
MyLungHealth: An Innovative Patient-Facing SMART on FHIR Decision Aid for Lung Cancer Screening
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:25 AM - 10:33 AM
Lung cancer is the leading cause of cancer deaths in the US. Lung cancer screening (LCS) has the potential to save >10,000 lives every year in the US. However, screening rates are <10% among eligible patients, in part due to payor requirements to conduct shared decision making using a decision aid prior to screening. Our group, ReImagine EHR, previously developed a provider-facing LCS decision aid, Decision Precision+, which increased the odds of screening 5-fold. Yet, over 80% of patients still remained unscreened. Several key barriers remain, including insufficient provider time, lack of patient education, and incomplete or inaccurate EHR smoking history. To address this need, we developed MyLungHealth, a SMART-on-FHIR patient-facing decision aid seamlessly integrated into the PHR.
Critically, MyLungHealth has now been evaluated in a pragmatic, multi-site, patient-randomized controlled trial across University of Utah Health (UUH) and NYU Langone Health (NYULH). The intervention improved core outcomes: it increased identification of newly eligible patients among those with uncertain eligibility and increased LDCT orders among those with documented eligibility. At UUH and NYULH, LDCT orders rose to 32% vs 24% and 32% vs 29% in intervention vs control, respectively; LDCT completion also increased at NYULH for patients with uncertain eligibility (2.8% vs 2.3%).
MyLungHealth provides personalized risk assessment and education using >20 parameters gathered via FHIR. By activating patients before the visit and correcting missing or inaccurate smoking histories, MyLungHealth measurably increases screening actions and supports equitable, scalable screening.
Speaker:
Christian Balbin, PhD
University of Utah
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:25 AM - 10:33 AM
Lung cancer is the leading cause of cancer deaths in the US. Lung cancer screening (LCS) has the potential to save >10,000 lives every year in the US. However, screening rates are <10% among eligible patients, in part due to payor requirements to conduct shared decision making using a decision aid prior to screening. Our group, ReImagine EHR, previously developed a provider-facing LCS decision aid, Decision Precision+, which increased the odds of screening 5-fold. Yet, over 80% of patients still remained unscreened. Several key barriers remain, including insufficient provider time, lack of patient education, and incomplete or inaccurate EHR smoking history. To address this need, we developed MyLungHealth, a SMART-on-FHIR patient-facing decision aid seamlessly integrated into the PHR.
Critically, MyLungHealth has now been evaluated in a pragmatic, multi-site, patient-randomized controlled trial across University of Utah Health (UUH) and NYU Langone Health (NYULH). The intervention improved core outcomes: it increased identification of newly eligible patients among those with uncertain eligibility and increased LDCT orders among those with documented eligibility. At UUH and NYULH, LDCT orders rose to 32% vs 24% and 32% vs 29% in intervention vs control, respectively; LDCT completion also increased at NYULH for patients with uncertain eligibility (2.8% vs 2.3%).
MyLungHealth provides personalized risk assessment and education using >20 parameters gathered via FHIR. By activating patients before the visit and correcting missing or inaccurate smoking histories, MyLungHealth measurably increases screening actions and supports equitable, scalable screening.
Speaker:
Christian Balbin, PhD
University of Utah
Christian
Balbin,
PhD - University of Utah
Pediatric Intensive Care Unit Tracheal Intubation Clinical Decision Support
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:33 AM - 10:41 AM
Acute respiratory failure drives more than 20,000 pediatric intensive care unit (PICU) intubations annually, with upwards of 22% complicated by adverse airway outcomes (AAO). To address this, we transformed a validated paper airway checklist into an EHR-embedded SMART-on-FHIR application. The checklist enables concurrent access across desktops, tablets, and mobile devices, overcoming the limitations of single-location, paper-based tools. Real-time EHR data populate the checklist using FHIR resources and support rule-based clinical decision support for evidence-based preparation and timeout workflows. A predictive risk assessment leverages a machine learning model trained on over 1,400 historical cases to stratify AAO risk and guide escalation. The app has been used by over 450 clinicians in more than 6,000 distinct encounters at CHOP, improving situational awareness, communication, and safety during high-risk tracheal intubations (Tis), and has been scaled to 5 other health systems.
Speaker:
Jeritt Thayer, PhD
The Children's Hospital of Philadelphia
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:33 AM - 10:41 AM
Acute respiratory failure drives more than 20,000 pediatric intensive care unit (PICU) intubations annually, with upwards of 22% complicated by adverse airway outcomes (AAO). To address this, we transformed a validated paper airway checklist into an EHR-embedded SMART-on-FHIR application. The checklist enables concurrent access across desktops, tablets, and mobile devices, overcoming the limitations of single-location, paper-based tools. Real-time EHR data populate the checklist using FHIR resources and support rule-based clinical decision support for evidence-based preparation and timeout workflows. A predictive risk assessment leverages a machine learning model trained on over 1,400 historical cases to stratify AAO risk and guide escalation. The app has been used by over 450 clinicians in more than 6,000 distinct encounters at CHOP, improving situational awareness, communication, and safety during high-risk tracheal intubations (Tis), and has been scaled to 5 other health systems.
Speaker:
Jeritt Thayer, PhD
The Children's Hospital of Philadelphia
Jeritt
Thayer,
PhD - The Children's Hospital of Philadelphia
Nourishing Innovations: Saving Time, Reducing Burden, and Enhancing Nutrition Care with a Standardized Nutrition Provision Tool
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:41 AM - 10:49 AM
In the complex landscape of healthcare, Registered Dietitians (RDs) play a critical role in nutrition management for patients. The manual calculation of nutrition support provision from tube feeding, parenteral nutrition, and calorie-containing drips can lead to inefficiencies and increase clinician burnout. The Nutrition Provision (NP) tool, developed in partnership with the Stanford Emerging Applications Lab (SEAL), leverages SMART on FHIR technology to streamline these calculations by directly interfacing with data in the electronic health record (EHR). This application not only automates data retrieval but also generates comprehensive reports that enhance clinical decision-making. By reducing calculation time and minimizing human error, the NP tool significantly improves workflow efficiency for RDs, paving the way for more equitable workloads and impactful nutrition support for patients.
Speaker:
Alexandria Weaver, MS, RD
Stanford Health Care
Click to View Presentation
2025 Annual Symposium On Demand
Presentation Time: 10:41 AM - 10:49 AM
In the complex landscape of healthcare, Registered Dietitians (RDs) play a critical role in nutrition management for patients. The manual calculation of nutrition support provision from tube feeding, parenteral nutrition, and calorie-containing drips can lead to inefficiencies and increase clinician burnout. The Nutrition Provision (NP) tool, developed in partnership with the Stanford Emerging Applications Lab (SEAL), leverages SMART on FHIR technology to streamline these calculations by directly interfacing with data in the electronic health record (EHR). This application not only automates data retrieval but also generates comprehensive reports that enhance clinical decision-making. By reducing calculation time and minimizing human error, the NP tool significantly improves workflow efficiency for RDs, paving the way for more equitable workloads and impactful nutrition support for patients.
Speaker:
Alexandria Weaver, MS, RD
Stanford Health Care
Alexandria
Weaver,
MS, RD - Stanford Health Care
Primary Record: Delivering the Last Mile of Interoperability into Communities and Family Homes
2025 Annual Symposium On Demand
Presentation Time: 10:49 AM - 10:57 AM
Primary Record is a patient- and family-owned health record that unifies siloed patient portals, clinical data, and family-shared updates into a secure, AI-supported hub. Families, caregivers, and community-based professionals use it to coordinate care across complex journeys such as aging and pediatric bone marrow transplants. Leveraging FHIR APIs, Primary Record retrieves and organizes data from EHR portals and integrates it with uploaded files, notes, and real-time communication tools. The app reduces fragmentation, strengthens family engagement, and supports clinicians by making information accessible and actionable. Already in use with care management professionals and life care planning attorneys, and nearing our first contract with a children’s hospital, Primary Record focuses on the last mile of interoperability, bringing health data into homes and communities where it can be used during the in-between encounters of care that directly impact health outcomes.
Speaker:
Jean Ross, MHA, BSN, RN
Primary Record
2025 Annual Symposium On Demand
Presentation Time: 10:49 AM - 10:57 AM
Primary Record is a patient- and family-owned health record that unifies siloed patient portals, clinical data, and family-shared updates into a secure, AI-supported hub. Families, caregivers, and community-based professionals use it to coordinate care across complex journeys such as aging and pediatric bone marrow transplants. Leveraging FHIR APIs, Primary Record retrieves and organizes data from EHR portals and integrates it with uploaded files, notes, and real-time communication tools. The app reduces fragmentation, strengthens family engagement, and supports clinicians by making information accessible and actionable. Already in use with care management professionals and life care planning attorneys, and nearing our first contract with a children’s hospital, Primary Record focuses on the last mile of interoperability, bringing health data into homes and communities where it can be used during the in-between encounters of care that directly impact health outcomes.
Speaker:
Jean Ross, MHA, BSN, RN
Primary Record
Jean
Ross,
MHA, BSN, RN - Primary Record
S73: FHIR App Competition
Description
Custom CSS
double-click to edit, do not edit in source