John Glaser


At a Glance

In health care, interoperability- the ability of healthcare information systems to work together and share information within and across organizational boundaries-involves:
 

  • Data exchange  
  • Infrastructure interoperability  
  • User interface interoperability  
  • Process interoperability  

The nation's healthcare IT agenda focuses on furthering the adoption of electronic health records (EHRs) that are interoperable-defined by the Healthcare Information and Management Systems Society as the ability of healthcare information systems to work together and share information within and across organizational boundaries.

More specifically, interoperability means:

  • The ability of computer systems to exchange and use information, usually in a large heterogeneous application network
  • The ability of diverse systems and organizations to work together
  • The ability to connect people, data, and diverse systems, taking into account social, political, and organizational factors

In health care, such sharing of data should promote more efficient, safer, and higher-quality care. Interoperability also will be necessary if providers are to perform well under payment mechanisms based on capitation, bundled payment, and episode payment arrangements.

From a healthcare technology perspective, interoperability specifically involves:

  • Data exchange
  • Infrastructure interoperability
  • User interface interoperability
  • Process interoperability

Data Exchange

For more than 25 years, interfaces between separate healthcare IT systems have facilitated the exchange of a broad variety of data, including admission- discharge-transfer (ADT) data and data related to orders, results, and charges. Originally, many of these interface solutions were designed to solve specific connectivity issues and were written as direct connections between two specific systems. Today, these interfaces handle the vast majority of data interoperability among healthcare IT systems.

Over time, interface standards were developed to structure the data for the most common types of interfaces. Vendors implemented these standard interfaces in ways that best suited their system configurations, but that often produced proprietary "standard" interfaces that required modification before data could be transferred to a foreign system.

To streamline these modifications, "interface engines" have been developed to format and sometimes reinterpret data as the data were moved from one system to another. These interface engines receive the data from one system, reformat the data as needed for the receiving system, and then forward the data on. Modern interface engines perform additional functions, such as verifying the receipt of data and auditing interface reliability and performance.

Nonetheless, the world of healthcare IT remains fragmented; providers usually rely on solutions from multiple vendors and custom applications using a wide variety of technologies. The use of a standard implementation process for custom and model interfaces that includes thorough testing and appropriate documentation helps to ensure that the entire integration or interface environment is stable and supportable and achieves the healthcare organization's objectives for data exchange.

Although long-established interfaces can be powerful, it may not be easy to expand the reach of such interfaces to allow for nationwide or global information sharing. This traditional approach has many limitations, including technological, economic, and privacy and security concerns.

Infrastructure Interoperability

The exchange of data is the focus for driving interoperability efficiency in a modern healthcare system. But efficiencies also can be realized by tying systems and infrastructures together more effectively. The following technologies enhance the ability of an organization's IT infrastructure to support interoperability.

User authentication and identification. Increasingly, providers are looking to tie user-credentialing administration for multiple applications into one directory, allowing them to establish a single sign-on that works across multiple applications. Initially, such efforts were focused on maintaining identities of users who manage application access. However, standards are being developed for use of the same infrastructure to manage other information related to users, staff, and medical practitioners at healthcare facilities, such as the physical location of the user.

Provider directories. According to the HIT Standards Committee (HITSC), a provider directory is "an electronic searchable resource that lists all information exchange participants, their names, addresses and other characteristics and that is used to support secure and reliable exchanges of health information." These directories could facilitate patient information exchange across the country. They would become the "master provider indexes" for regional healthcare information exchanges.

A special provider directory task force within the HITSC has made several recommendations to the U.S. government relative to the formation of these directories and is developing additional recommendations. The first priority of the task force is to standardize queries for provider digital certificates that can be used to securely encrypt electronic messages. This initiative will help facilitate the rapid increase in secure electronic health information exchange throughout the industry.

Service-oriented architecture. A service-oriented architecture (SOA) is a modern method of building an application or enterprisewide infrastructure comprising assorted applications.

SOA infrastructures share several common characteristics:

  • They involve loosely coupled applications and components that interact directly with each other without the need for interface engines.
  • They reuse application-provided functionality, allowing a service to be built once and used in multiple contexts and processes.
  • They enable workflow/process choreography based on the composition of services that provide new business functionality and patterns not originally conceived by the application designers.

Well-established technical standards, such as web services, hypertext transfer protocol, simple object access protocol, and representational state transfer, allow for syntactic interoperability and network interoperability. Healthcare-related semantic service standards-such as HL7 v3, a set of standards created by Health Level Seven (HL7), a not-for-profit organization devoted to the development of international healthcare informatics interoperability standards-lead the way toward increased capability, interoperability, and semantic interoperability.

Workflow engine (business process management). A workflow engine is a software component or system that actively manages and executes business process models by starting, continuing, and ending processes in response to enterprise events. The workflow engine "listens" for events of interest, invokes business and clinical rules, and interacts with data and business services. Rounding out the capabilities of a workflow engine are managing work lists or queues and communicating information to and from users.

A workflow engine can provide "process interoperability" that crosses system boundaries. Processes are usually measured for effectiveness, quality, time, or cost improvements. To that end, workflow engines can be coupled to analytics tools that enable monitoring of key performance indicators.

The level of maturity for standards in workflow models and engines is still evolving. Some standards, such as business process execution language and business process modeling notation, have some traction, but are incomplete in expressing the workflows completely.

User Interface Interoperability

Most healthcare organizations rely on several specialized applications to enable clinicians and other professionals to provide care to patients. Often, the same data or functionality must be available in several different applications if a user is to perform a given workflow. User interface interoperability enables the user to move easily among multiple applications quickly and accurately.

The simplest form of user interface interoperability is the ability of one application to launch another, such as through a button or link in the primary application that launches a second application on the same workstation. From that point, the two applications simply run on the same workstation with no other interaction.

A more sophisticated and helpful solution for user interface interoperability involves context sharing. In this interoperability framework, a context may be defined as "the array of common elements that define and limit the way users interact with applications." Context sharing occurs when one application not only launches another, but also passes on unique information, such as a medical record number, to improve end-user efficiency and accuracy in completing a workflow objective. The information that has been passed on will vary depending on the applications involved or the needs being addressed by the launch. In healthcare scenarios, the most common context sharing involves sharing the identity of the person using the application and the identity of the patient whose data are being managed.

Process Interoperability

A central goal of clinical data exchange is to improve the interoperability of patient care processes within and between organizations.

In some ways, the current clinical data exchange approach assumes passive relationships between organizations or departments. Two organizations may need to establish a connection and ensure that they follow standards and have necessary legal agreements in place. However, once "implemented," a clinical data exchange does not require that two organizations routinely work together. The care processes within one organization can largely be ignorant of the care processes in the other. As an analogy, the presence of a global phone system that enables interoperable voice communication does not require that any two organizations that utilize this network have any process interoperability.

There are three IT approaches that can support active and powerful classes of interoperable organizational processes. In fact, the decision to pursue these approaches is based on a desire to have some significant level of interorganizational process interoperability. These approaches may have more value potential to the participating organizations than the "simple" exchange of clinical data between them.

The first approach involves one organization extending its systems into another. For example, a joint venture to provide cancer care may involve one organization, a cancer center, extending its oncology electronic medical record (EMR) data into another organization (e.g., a community hospital) for use by clinicians at both organizations. The ongoing evolution and management of this shared application becomes a shared responsibility. This kind of collaboration is deeper than the exchange of data; it is likely to involve using common care protocols, and is predicated on two organizations (or more) engaging in a substantive partnership surrounding a joint approach to delivering a clinical service.

The second approach centers on two or more organizations using the exchanged data to address problems or opportunities of mutual interest. For example, a provider and payer may leverage the exchange to integrate data from medication claims with the provider's EMR data. Merging these data enables both organizations to examine medication compliance and misuse issues and jointly develop strategies to improve care. The exchange of data is a necessary step in creating a new shared process.

In the third approach, two or more organizations may leverage an exchange as one tool in a joint effort to improve the effectiveness and efficiency of an existing interorganizational process. For example, claims submission using an insurance transaction exchange is a tactic that two organizations may use to improve a critical process that binds them. In the examples above, the exchange of data is a tool to further a significant joint undertaking with the potential to advance an aspect of care. The focus of the initiative is primarily on a process and secondarily on IT.

The Means to an End

Healthcare information exchange and interoperability is a fundamental requirement for the nation's healthcare system to derive the societal benefits promised by the adoption of EHRs. The promulgation of interoperable EHRs will enable the nation and its stakeholder organizations to take a major step forward in efforts to improve the quality, safety, and efficiency of care. As implementation of the national healthcare IT agenda progresses, it is important to remember that care process interoperability is the goal. Achieving this goal will lead to several IT approaches designed to achieve interorganizational process improvement.


John Glaser, PhD, is CEO, Siemens Healthcare Health Services, Malvern, Pa. (john.glaser@siemens.com).



 

Publication Date: Tuesday, November 01, 2011

Login Required

If you are an existing member, please log in below. Username and password are required.

Username:

Password:

Forgot User Name?
Forgot Password?







Close

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:

Email:

   Become an HFMA member instead