Interoperability and the healthcare enterprise

By Janine Broda, VP and General Manager, MEDxConnect Medical Division, Compressus, Inc.

Space station operations; air traffic control systems; electronic banking – major systems with many participants where standardization and interoperability is critical, right? Well, don’t you think that standardization and interoperability in healthcare would be just as critical? Unfortunately, if you attend healthcare IT meetings and review the industry, one thing is clear – while interoperability is the hot topic, people aren’t sure what interoperability means or how to achieve it.

Healthcare has made strides recently with the creation of DICOM standards, which are designed to bring standardization to the transmission of medical images and associated data, such as reports, notes, and patient information. The goals of DICOM are to achieve compatibility and to improve workflow efficiency between imaging systems and other information systems. This has largely been accomplished as a radiologist can now sit down at an Agfa workstation and pull up images acquired on a Toshiba CT system, for example.

However, despite the impact of DICOM and widespread adoption of PACS systems with over $2 billion spent annually on PACS implementations, interoperability is still a major issue. The reason is no surprise. Despite its progress and success – DICOM isn’t always DICOM. Because DICOM does not define system architecture or specify all functional requirements beyond transmission, compatibility often stops at the image review process.

The greatest source of interoperability problems within a Healthcare Enterprise is associated with the fact that different systems use proprietary data structures to store and move information. For example, PACS use DICOM for images, but often have proprietary structures for notes, overlays and other vital image criteria. This makes it difficult to impossible for other PACS or referral physician’s workstations to view the complete image file. Some RIS systems still use proprietary messages and files, making it arduous for other systems to receive critical information including patient history, orders and reports. It is these proprietary structures that limit interoperability and leads to the inefficient delivery of patient care.

Integration Over Consolidation

Now, many large venders – and even some health system CIOs – will argue that the solution is simple: consolidate to one PACS/RIS system and you won’t have any compatibility issues. While there are obvious commercial benefits to this approach, it is also a natural by-product of the communications problems inherent to traditional PACS implementations based on proprietary technologies and competitive pressures. This often forces medical facilities to settle for a sub-par solution with a single vendor providing all components or rely on stop gap measures to achieve common communications across all operational areas – many times in the form of moving personnel to follow the images from one workstation to another.

But, what happens when your hospital or health system adds a new radiology practice or imaging center that uses a different system? The answer is that you’re right back where you started – with components that only partially communicate. It would be naïve to think that in practice, any multi-site healthcare system could use a single system for any period of time.

As Health Systems grow in size and complexity, Enterprise Systems Integration will become increasingly vital to the successful management of patient care and the bottom line. Successful integration technologies must adhere to accepted medical and communications standards and efficiently leverage existing communications infrastructure to connect those systems and applications through standardized and proprietary formats. To help create an efficient interoperable information system for a healthcare enterprise, systems integration software solutions must provide seamless connectivity and interoperability between diagnostic acquisition devices, PACS, HIS, RIS, and their related information systems.

Invariably, when my company, Compressus, is brought in to develop a comprehensive integration solution, the site has struggled with a series of IT consultants to look at multiple sites using many different PACS/RIS systems and to link them into an efficient operating network. In our experience, to create a true communications hub the solution must support not only DICOM standards, but also HL7, XML, SSL, TLS and other standards to ensure HIPAA compliance in both procedural and security areas. Using the communications protocols and semantics of each participating information system, enables users to control the flow of images, reports, messages, standard patient demographics and other important data for diagnosis and treatment planning of a patient between otherwise independent and disparate systems.

This provides sites the freedom to choose the HIS/RIS/PACS systems based on feature preference rather than to avoid integration issues. In essence, by mapping the data elements and bridging isolated islands of data, the software mediates interoperability between the different applications and creates a virtual integrated information system that adheres to the IHE technical framework. This information can then be reliably used to manage pre-fetching rules, study distribution, and report creation and delivery. The result is a single enterprise-wide diagnostic solution from many disparate parts.

Another significant step to implementing an enterprise with robust connectivity and interoperability is developing a layer of workflow in the system architecture. A workflow engine in this layer, or application service bus, can be set-up by a skilled systems integrator to achieve automation of the workflow desired for the enterprise. The engine is “programmable” for workflow threads rather than “hard wired” like many workflow applications that can not be customized to fit the needs of an individual enterprise. As a result, this engine can be set-up to achieve IHE compliance for the IHE workflow threads and use cases. This type of Service Oriented Architecture is essential to the success of building an IHE enterprise.

Further, the workflow engine layer must also support intelligent routing of any object in the enterprise as prescribed by the customer. It must be programmable during set-up to achieve the customization desired – this includes standard rules-based routing of cases by physician, exam type of anatomical area, but must also include the ability to mediate a variety of file types that the customer desires to achieve an open system. Another crucial element of the workflow engine layer is the ability to mediate or normalize data. For example, the ability to present DICOM images as JPEG or other file type that the customer needs for purposes other than diagnosis.

Ultimately, the lack of interoperability between existing information and imaging systems is one of the most critical problems facing the diagnostic and management effectiveness of the healthcare profession today. To be successful, Healthcare Enterprise Systems must create one seamless integrated enterprise diagnostic network. Not only will this increase efficiency and save time and money, this level of interoperability is essential for eventual implementation of a standards based Electronic Health Record.