Which organization is dedicated to creating standards for interoperability of electronic health information?

The Office of the National Coordinator for Health IT (ONC) is responsible for advancing connectivity and interoperability of health information technology (health IT).

What is interoperability?

According to section 4003 of the 21st Century Cures Act, the term 'interoperability,' with respect to health information technology, means such health information technology that— "(A) enables the secure exchange of electronic health information with, and use of electronic health information from, other health information technology without special effort on the part of the user; "(B) allows for complete access, exchange, and use of all electronically accessible health information for authorized use under applicable State or Federal law; and "(C) does not constitute information blocking as defined in section 3022(a)."

The Roadmap: Calls to Action and Commitments Organized by Participant Group

We all play a role and must work together to make health information electronically and securely available when and where it is needed to support the health and well-being of all Americans. The calls to action and commitments from the Roadmap were shaped by input from both experts and people on the ground, who understand the complex health IT ecosystem and sensitive nature of health and care. Explore the calls to actions and commitments organized by participant group as well as the statements of support from industry groups and organizations.

Vision Paper

2015 Edition Health IT Certification Criteria

The 2015 Edition Health IT Certification Criteria (2015 Edition) rule works in concert with the Roadmap and serves as a key component in ONC’s commitment to establishing an interoperable nationwide health information infrastructure. The 2015 Edition rule builds on the foundation established with the 2011 and 2014 Editions to advance interoperability, and incorporates changes designed to foster innovation, open new market opportunities, and provide more choices to the care community when it comes to electronic health information exchange.

The Federal Health IT Strategic Plan 2015-2020

The Federal Health IT Strategic Plan 2015-2020 (Strategic Plan), identifies implementation of the Roadmap as critical to enhancing the nation’s health IT infrastructure. The Roadmap directly aligns with the Plan’s mission of improving the health and well-being of individuals and communities through the use of technology and health information that is accessible when and where it matters most. While the Plan focuses on federal efforts, the Roadmap details the policy, technology and behavioral changes that public and private stakeholders must make to achieve nationwide interoperability.

Interoperability Standards Advisory

The Interoperability Standards Advisory (ISA) process represents the model by which the Office of the National Coordinator for Health Information Technology (ONC) will coordinate the identification, assessment, and determination of the “best available” interoperability standards and implementation specifications for industry use to fulfill specific clinical health IT interoperability needs. For detailed background on the Advisory, its purpose, and its processes please review the 2017 Interoperability Standards Advisory.

ONC also supports international efforts related to health IT standards.

Interoperability Infographic

AMIA Annu Symp Proc. 2018; 2018: 555–563.

Published online 2018 Dec 5.

Peter J. Haug, MD, 1 , 2 Scott P. Narus, PhD, 1 , 2 Joseph Bledsoe, MD, 1 , 2 and stanley Huff, MD 1 , 2

Abstract

During the last decade, software supporting healthcare delivery has proliferated. This software can be divided into electronic medical record (EHR) systems and applications that treat EHRs as platforms. These collect, manage, and interpret medical data, thereby adding value to associated EHRs. To reduce the burden of developing for multiple EHR platforms, a group of standards has evolved that allow software written for one vendor’s EHR to be introduced into settings supported by other vendors.

The Health Services Platform Consortium (HSPC) is a collaborative effort to advocate for standards that will make healthcare applications truly interoperable. In this document, we discuss the approach adopted by the consortium and the standards central to this approach. We discriminate between interoperability standards that support the plug-and-play transfer of applications from one vendor’s EHR to another and knowledge portability standards that allow knowledge artifacts used in one software environment to be introduced effectively in others.

Introduction

Over the last decade, the use of information systems in medicine has become ubiquitous. Recent efforts by the federal government to encourage the use of electronic health records (EHRs) has led to widespread availability of clinical information systems that manage patient records, oversee delivery of therapeutic interventions, provide billing and scheduling services, and contribute to core decision-making in the areas of diagnosis, prognosis, and treatment. While a number of large and capable health information systems exist, each system has limitations. Many of these limitations are amenable to software solutions, and new technologies, evolved both within and outside the healthcare field, are expected to enhance important aspects of healthcare delivery.

The ability to implement, test, and ultimately disseminate these new technologies across healthcare software vendors is eased when international standards can be used to integrate innovative applications into existing EHR platforms. When fully realized, this approach supports the development of applications that are fully interoperable, i.e. they can be moved from one EHR to another with no change in the underlying software.

This degree of interoperability extends the usual definition which is typically limited to “the ability to share information between systems and employ it meaningfully”. The extended definition of interoperability is rarely achieved with existing standards. However, it is the central goal of the Healthcare Services Platform Consortium (HSPC)1, an organization devoted to the identification, demonstration, and dissemination of a suite of standards rendered unambiguous by careful attention to the structure and semantics of the underlying information models.

Here, we discuss the key elements of HSPC’s goal, an environment that supports the development of fully interoperable, standards-based applications. We will review the standards currently identified as a part of this environment as well as additional standards under evaluation that would further extend the platforms capabilities. Our focus will be on that group of specifications identified by HSPC as high value targets to demonstrate the power of unambiguous, interoperability standards.

As a part of this discussion we will report experience with two applications. The first implements a pediatric growth chart that uses standards-based tooling to allow its integration into two different EHRs. The second focuses on a multistep protocol for the diagnostic workup of pulmonary embolism in the emergency department. The original application was developed using standards from the Object Management Group (OMG)2 and deployed to run against a legacy EHR. It has been further standardized by invoking Health Level 7 (HL7)3 data access standards as well as standards for integrating the application’s user interface into a hosting EHR.

The Healthcare Services Platform Consortium

The Healthcare Services Platform Consortium was founded on May 21, 2013, by a small group of healthcare organizations determined to refocus how healthcare applications and services are developed. In the short time since HSPC launched, more than 270 contributors have participated in the development of detailed organizational goals.

These include leading healthcare and government organizations, healthcare technical solutions vendors and a venture-led group of investors supporting an HSPC-centric portfolio of companies.

HSPC’s stated mission is “to improve health by creating a vibrant, open ecosystem of interoperable applications, knowledge, content, and services”. It is organized to operate as a provider-led consortium. Its goal is to blend the hundreds of active organizations involved into a healthcare services platform community able to support a market place for interoperable applications in a Service-Oriented Architecture (SOA) and knowledge-enabled environment. Through the resulting interoperable marketplace and services platform, the consortium seeks to foster a new level of provider-vendor collaboration and innovation to meet one of the industry’s greatest needs: accelerating the creation and delivery of promising software applications at the point of care.

In addition to its efforts to develop, support, and refine existing standards, HSPC supports a software development environment, the HSPC Sandbox4. This is provided to members to support the standards-based development of novel clinical applications. In addition, HSPC has recently released a document describing an initial, multi-year roadmap to provide this community an enhanced interoperability platform. Below we describe two different types of standards promoted by HSPC. Subsequently, we will discuss sample applications employing a subset of the standards described in the roadmap. These are selected to illustrate the character of the proposed platform. We will divide our discussion between the key interoperability standards and relevant knowledge portability standards envisioned in this document.

Standards

Modern healthcare computing standards address a number of different challenges. Principal among these is interoperability. This can be defined as the ability to transfer a useful application from one setting to another without significant alteration. In medicine, interoperability describes several different phenomena. The first of these is the ability to install an application in a new setting (featuring a new EHR) that was initially developed to extend the functionality of a different EHR. Success is defined in terms of 1) little or no reworking of the software to accommodate the new environment, and 2) behavior/benefits in the new setting that are identical to those seen in the original setting.

This description captures a key need in medical computing. Marked differences exist in the ways different EHRs support data storage and management. Standardized approaches to retrieving and storing clinical data and diagnostic/therapeutic orders are essential to the achievement of interoperable applications. Additionally, standards for integrating user interfaces into the workflows provided by EHR vendors are a necessary prerequisite for the successful installation of a promising application into a new EHR.

While interoperability standards are necessary for cross-platform implementation of entire applications, another class of standards exists that supports the export and import of medical knowledge independent of underlying interfaces to an EHR. These standards describe systems for encoding fundamental algorithms, workflows, and relationships in ways that allow the guidance provided by this knowledge to be reproduced in differing software environments. These knowledge artifacts, rather than being represented in standard programming languages, employ modeling languages specific to the character of the clinical knowledge targeted. Typical examples are representations of clinical rules and medical processes expressed using standards-based authoring environments. These authoring environments are designed to, at once, ease the task of representing medical knowledge and to produce an output that can be implemented using a standards-compliant execution engine. The advantage of achieving this type of interoperability can be broadly described as knowledge portability.

While the key goal of the interoperability standards is to minimize the technical challenges associated with transferring medical applications between clinical organizations, knowledge portability standards target the ability to exchange and manage the computable medical content utilized by these applications. The adoption of knowledge portability standards confers several technical and clinical benefits:

• Portability: While the interoperability standards lower the barriers associated with transferring entire applications from one organization to another, knowledge portability standards focus on the medical knowledge delivered by clinical software. Knowledge described in a standards-based format can be implemented using standards-based execution engines from different sources. These engines may be programmed in different languages, served by different operating systems, and provided by different vendors. The expression of clinical algorithms in this way invites a competitive marketplace in which multiple vendors provide execution engines that compete on speed of execution, scalability, ease of maintenance, etc.

• Comprehensibility: Programming languages are seldom understood by medical experts. Well-designed knowledge authoring environments allow effective review and critique of medical knowledge artifacts by domain experts, first within the organization that produces the artifact and subsequently at organizations seeking to adopt the artifact. The standards-based nature of the language and the execution systems means that the adopting organization can not only critique the fitness-for-use of a specific tool but can easily alter key parts to conform to local differences in workflows and resources.

• Complexity: A well-designed, standards-based, authoring system invites the application of decision-support technologies to medical problems of higher complexity. In particular, algorithms that span transitions of care and can project a care process across multiple locations and caregiver roles become more manageable.

• Maintainability: A permanent feature of modern healthcare is change. Science provides a continuous stream of new information bearing on the diagnosis, care, and prevention of disease. In this environment, an unmaintainable decision-support system may lose its effectiveness in a few years. Standards-based authoring systems are essential to an institution that wishes to maintain a comprehensive and relevant set of executable knowledge artifacts over time.

• Scalability: While “scalability” often refers to the ability to simultaneously execute many instances of a given software system, it may also apply to the ability to efficiently execute clinical systems distributed across multiple institutions while serving multiple patients being cared for by multiple caregivers over extended time periods. A standardized, well understood knowledge execution environment supports this broad view of scalability.

These two complementary classes of standards are the foci for the efforts of the Healthcare Services Platform Consortium. A number of existing and evolving standards are available to fill these roles. Below, we discuss an initial group of standards chosen as a core set by the HSPC.

HSPC-Adopted Standards

Based on input from the HSPC community, a group of healthcare standards was identified and included in version one of the HSPC roadmap. The key standards are those that lead to true interoperability. They provide a foundation upon which additional standards supporting knowledge portability can be exploited. Below we will briefly describe the roadmap’s standards.

Interoperability Standards

Developing standards to support interoperability has been a central focus of the international standards organization, Health Level 7 (HL7), for decades. The HL7 2.X standards5 for interfacing clinical systems are broadly implemented in the United States and abroad. In recent years, HL7 has sponsored a large and growing effort to define a standard set of interfaces (APIs) to computing services designed to provide connectivity to key EHR data and functions and intended for implementation by EHR vendors (Table 1). These form the essential infrastructure for standards-based applications in medicine.

Table 1:

Interoperability standards.

The effort to unambiguously define data-exchange and terminology standards also involves a collaboration between HSPC and the HL7 Clinical Information Modeling Initiative (CIMI) Workgroup6. This Workgroup brings a special focus to the standardization of the semantic models underlying data communication. It represents an international effort to develop the set of data objects necessary and sufficient to achieve true interoperability7.

Knowledge Portability Standards

The challenges associated with knowledge portability are not unique to healthcare. In a number of industries, the ability to capture and exchange executable knowledge has been recognized as a major enabler of innovative approaches to product delivery. One goal of the HSPC knowledge portability effort is to identify valuable, standards-based, knowledge modeling efforts designed for both medical and nonmedical activities and to incorporate these into the suite of standards endorsed by HSPC. This effort involves collaborations with organizations whose primary focus is outside of healthcare.

The standards listed in Tables 1 and ​2 encompass both relatively mature specifications, endorsed by their respective standards organizations as ready for routine use, and less mature specifications identified as Standards for Trial Use (STU). The STU standards are still rapidly evolving. Acceptance for routine use will follow testing and refinement. HSPC has incorporated these standards into its knowledge portability framework with the stipulation that, as experience matures the specifications, the HSPC endorsed version will track the current specifications.

Table 2:

Knowledge portability standards.

Table 2 contains a list of standards that have been identified in the HSPC roadmap as targets for integration into the HSPC sandbox and identified as preferred approaches to the development and sharing of clinical knowledge artifacts.

Of particular in the HSPC community are the Object Management Group’s (OMG) Bussiness Process Management Standards. These define specifications that are being rapidly adopted in the general bussiness domain and are receiving substantial support from competing suppliers of authoring and execution system. As an example, for the Bussiness Process Management and Notation (BPMN) specification, Wikipedia lists 29 BPMN tool vendors21.

Moreover, the inclusion of graphical authoring standards as an integral part of the specification helps overcome an important hurdle. If properly used, these standards can provide a common language in which decision authors and subject matter experts can design sophisticated clinical workflows and iteratively refine them with experience.

BPMN has filled a specific and well-defined business need. Modern businesses face a constant challenge due to continuing changes in the business environment and the increasingly complex efforts necessary to plan, document, and manage the workflows associated with these processes. Business analysts are continually developing and refining models for complex workflow processes. These workflows can involve participation from a variety of individuals, departments, and external partners. The challenge of managing these processes has led to a number of tools used to design and implement optimum workflows.

The delivery of medical care in many ways parallels the variety and complexity of processes in the business world. Many diagnostic and therapeutic activities are defined by complex, multistep processes involving a variety of caregivers across a multitude of care settings. A core challenge for any medical enterprise is to understand and manage the workflows necessary to deliver care.

Examples: Two Standards-Based Applications

As the HSPC has incorporated standards into its roadmap, it has encouraged initiatives to test these standards with real-world, clinical use cases. Below, we briefly describe two examples. The first illustrates the use of interoperability standards in an application designed to display and evaluate the pediatric growth chart. The second is an example of the use of the BPMN 2.0 standard to implement a clinical workflow designed to standardize the diagnosis of pulmonary embolism.

Standards-Based Growth Chart

Pediatric growth charts are a core constituent in the routine care of children. They have been use for decades and come in a variety of paper-based and electronic forms. They were an early target for developers of SMART applications.

The original growth chart tool was developed as a browser-based application at Boston Children’s Hospital to demonstrate the original Substitutable Medical Apps, Reusable Technology (SMART)9, but it was subsequently modified and extended by Intermountain Healthcare for use in a production setting. The University of Utah Health System then further enhanced Intermountain Healthcare’s version with additional features, which were then deployed by Intermountain. These two institutions use different EHRs, and the growth chart is an example of an interoperable application that is in active use on Intermountain’s Cerner system and is ready for deployment on the University’s Epic system. The data used by the app are accessed using HL7 FHIR-based services. The application’s visual displays and workflow are incorporated directly into the Cerner and Epic EHR using the SMART framework, which allows user and patient context to be directly shared between the EHR and the app. A version is available as open source code in Github22.

This application is available for general use within the Intermountain Cerner environment. We are currently recording usage data on the most recent version. As an indication of continued clinical interest, physicians have requested additional growth chart options (one for premature babies and one for Down’s Syndrome children). We have provided these, plus the ability to add new growth curves without having to do new coding.

Pulmonary Embolism Diagnostic Application

In contradistinction to the growth chart application, a second, standards-based application, developed at Intermountain Healthcare, began as an exploration of a knowledge portability standard and is currently being extended to be an example of the use of both knowledge portability and interoperability standards. It takes advantage of BPMN version 2.0, the OMG standard for the design and execution of complex, long-running, workflows potentially involving multiple interacting human users and software services.

A BPMN 2.0 implementation was used to support both authoring and delivery of an electronic Pulmonary Embolism protocol. The graphical authoring environment demonstrated its value as a tool to ease the communication between a group of (medical) content experts and a clinical process analyst. Using BPMN, we developed, implemented, and tested a multistep, executable workflow to assist in the diagnostic workup of Pulmonary Embolism. Figure 2 provides a high-level view of the protocol expressed graphically in the standards-defined authoring environment.

The Pulmonary Embolism Diagnostic Application. A BPMN diagram depicting the highest level of the PE diagnostic workflow. The thick-walled boxes indicated sub-processes each of which contains a detailed BPMN sub-workflow depicted using the standard, graohical authring system. In addition ato the graphical workflow representation, the authoring tools produce an executable representation that can be implemented using a standards-based execution engine.

Experience

In 2014, the automated workflow for diagnosing Pulmonary Embolism (PE) was made available on computer desktops in four emergency departments within Intermountain Healthcare’s Urban Central Region. A reminder to use this tool was triggered upon receipt of an order for a test associated with PE (D- Dimer, CT Pulmonary Angiogram, Ventilation-Perfusion scan). The workflow began by guiding the user through a series of evidence-based algorithms designed to estimate the likelihood of PE (based on the Pulmonary Embolism Rule-Out Criteria23 and the Revised Geneva Score24). It implemented the BPMN-based logic necessary to choose the most appropriate sequence of diagnostic tests. It concluded when sufficient information was collected to confirm or deny the presence of PE.

Between June 2014 and February 2016, the prototype system underwent clinical testing. The protocol was initiated for 1084 patients; it was exited before completion in 620 cases. In many of the latter, the clinician reviewed suggestions before exiting, and most of these “exits prior to completion” were in the setting of a negative D-dimer, which would have ended the protocol. Of 2471 total suggestions displayed to the user (suggest a D-dimer lab test, suggest imaging examination in general, and a suggestion for a specific imaging examination), 2353 (95%) were accepted and 118 (5%) were not accepted.

Further Standardization of the Pulmonary Embolism Application

The original pulmonary embolism diagnostic application focused on the ability to model multistep, stateful processes using BPMN. It was retired when Intermountain Healthcare adopted the commercial Cerner EHR and installed it across Intermountain’s 23 hospitals and 185 outpatient clinics. However, this application provides a key use case for HSPC’s development of a complete suite of healthcare standards. It is currently deployed against the HSPC sandbox and, in addition to its original BPMN 2.0 workflow engine, it enjoys access to a number of additional standards. In this environment, it can access and store data using FHIR DSTU 2.0 standards, it provides a user interface through the SMART standard, and it has access to new EHR events through a publish-and-subscribe interface consistent with the applicable HL7 standard.

The pulmonary embolism diagnostic system continues to provide a test bed where relevant standards can be explored in the context of a significant clinical problem. It provides a use case that drives extensions and modifications of the services provided in the HSPC Sandbox. We will be able to re-release it for clinical use once we have completed a version that fully conforms to the HSPC-endorsed standards and our EMR vendor provides services consistent with the enhanced standards available in the Sandbox.

HSPC Sandbox

As indicated above, HSPCs goals are best served by the adoption of unambiguous standards. Initially, development of applications designed to demonstrate standards compliance has suffered from the lack of well-constructed development environments featuring a comprehensive collection of relevant standards-based services. The Consortium has acted to promotes key standards in part by providing a standards-compliant, development environment, the HSPC Sandbox. As described above, our current work on the pulmonary embolism application takes advantage of the HSPC Sandbox. It is done on a separate web application server and treats the Sandbox as the hosting EHR. This application and others both rely on the Sandbox and provide opportunities to test and refine the specifications implemented there.

The HSPC Sandbox is an evolving environment that makes available to application developers the tools, data, and services necessary to construct compliant applications. Data relevant to a particular application can be inserted into its database. Healthcare software can be built, tested, and demonstrated in this environment with the expectation that transfer to a fully compliant EHR will require modest effort.

A variety of healthcare software developers have taken advantage of this site. A subset of their applications can be seen at http://gallery.hspconsortium.org.

Discussion

Although some of the standards described above have been in use in one form or another for more than a decade, many are relatively untested. Most are a few years old at best and represent transitional specifications as recognized by labels such as “Standard for Trial Use (STU)”. Moreover, a number of these standards are under-defined, leaving room for variability in implementation. This variability threatens to defeat the goal of truly interoperable applications and truly portable clinical knowledge. None the less, the rapid evolution of healthcare standards in recent years is a testimony to the need the healthcare industry has for unambiguous interoperability and knowledge portability standards.

The universal adoption of these standards holds substantial promise for vendors of innovative applications to participate more broadly in the healthcare marketplace. We believe that wide standards adoption also offers opportunities to improve key aspects of clinical research. If generally available, the kinds of interoperable applications described would have the ability to standardize the delivery of complex therapeutic protocols across healthcare institutions regardless of the incumbent EHR. Experimentation in the use of these interfaces with data warehouses such as integrating biology and the bedside (i2b2) repositories is underway25.

The benefits described above can only be realized through the general adoption of unambiguous standards. This is the long-term goal of HSPC and is reflected in the HSPC roadmap. The focus of this document is on phased development with regular deliverables. Among the near-term steps toward this goal are:

• The further development of tools to manage data models and medical terminologies.

• Publication of foundational medical knowledge content.

• Enhancement of the Sandbox with standard terminology services and other features.

• A shared interoperability maturity model to guide the evolution of relevant standards.

• A carefully selected set of use cases to drive and test the standards adopted by the organization.

As a part of these efforts, HSPC seeks to extend the range of collaborating organizations committed to this work. It plans shared activities with healthcare institutions, standards organizations, government institutions, healthcare software venders, and medical associations. Groups such as the Clinical Information Interoperability Council (CIIC) share many of HSPC’s goals. Moreover, these groups can bring additional levels of perspective and healthcare expertise to the standards definition process.

Conclusion

In summary, the Healthcare Services Platform Consortium represents a collaborative group of healthcare providers and vendors committed to the development of complete and unambiguous standards in healthcare. Through interaction with relevant standards organizations, participation in the development of standards-based applications, and the creation of environments that support the development of fully compliant, healthcare software, the HSPC expects to contribute to a marketplace in which high quality clinical applications are broadly available for a wide range of electronic healthcare platforms. We anticipate greater quality, more easily obtained clinical applications and improved environments for healthcare researchers who wish to pursue novel research designs across multiple healthcare systems.

​

The SMART-on-FHIR Growth Chart Application. This standard-based tool for visualizing the development of pediatric patients has benn integrated into both the Epic and Cerner EHRs.

Footnotes

*The SMART specification is copyright Boston Children’s Hospital and Health Level Seven International. The content is licensed under CC BY 4.0.

†In HL7 balloting process.

References

2. The Object Management Group [Internet] [cited 2018 Feb 18]. Available from: https://www.omg.org/.

3. HealthLevel 7 [Internet] [cited 2018 Feb 18]. Available from: http://www.hl7.org/.

7. The Clinical Information Modeling Initiative [Internet] [cited 2018 Feb 18], Available from: https://www.opencimi.org/.

8. Health Level 7: The Fast Healthcare Interoperability Resources [Internet] [cited 2018 Feb 18]. Available from: https://www.hl7.org/fhir/.

9. Mandel JC, Kreda DA, Mandl KD, Kohane IS, Ramoni RB. SMART on FHIR: a standards-based, interoperable apps platform for electronic health records. J Am Med Inform Assoc. 2016 Sep;23(5):899–908. [PMC free article] [PubMed] [Google Scholar]

12. The Systematized Nomenclature of Medicine - Clinical Terms (SNOMED CT) [Internet] [cited 2018 Feb 18]. Available from: http://www.snomed.org/snomed-ct.

14. Logical Observation Identifiers Names and Codes (LOINC) [Internet] [cited 2018 Feb 18], Available from: https://loinc.org/.

17. The Object Management Group: About the business process model and notation specification version 2.0 [Internet] [cited 2018 Mar 6], Available from: http://www.omg.org/spec/BPMN/2.0/.

18. The Object Management Group: About the decision model and notation specification version 1.1 [Internet] [cited 2018 Mar 6], Available from: https://www.omg.org/spec/DMN/About-DMN/.

19. The Object Management Group: About the case model management and notation specification version 1.1 [Internet] [cited 2018 Mar 6], Available from: https://www.omg.org/spec/CMMN/About-CMMN/.

23. Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. “Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism“. Journal of Thrombosis and Haemostasis. (2004);2(8):1247–55. [PubMed] [Google Scholar]

24. Le Gal G, Righini M, Roy PM. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. Annals of Internal Medicine. 2006;144(3):165–71. [PubMed] [Google Scholar]

25. Wagholikar KB, Mandel JC, Klann JG, Wattanasin N, Mendis M, Chute CG, Mandl KD, Murphy SN. SMART-on-FHIR implemented over i2b2. J Am Med Inform Assoc. 2017 Mar 1;24(2):398–402. [PMC free article] [PubMed] [Google Scholar]

Articles from AMIA Annual Symposium Proceedings are provided here courtesy of American Medical Informatics Association