Baystate Health: System-wide PACS Replacement Yields Tight Integration

The seamless interface of information systems within an enterprise is the Holy Grail in health systems for good reason: Tight integrations can yield big benefits. After replacing its PACS and tightly interfacing the new system with its electronic medical record (EMR), Springfield, Mass.-based Baystate Health is reaping the benefits of streamlined workflow and improved patient care across its four hospitals, more than 10,000 employees, cancer center and heart and vascular center.

Baystate Medical Center is the western campus of Tufts University School of Medicine and the only Level I trauma center and pediatric trauma center in western Massachusetts. The system's radiology departments and freestanding imaging center are staffed by 45 radiologists and 16 residents who perform ~350,000 imaging examinations annually.

Two years ago, Kal Dulaimy, MD, chief of diagnostic imaging and medical director of radiology informatics, at Baystate Medical Center, the system’s 700-bed flagship hospital, and colleagues took a long, hard look at Baystate’s former PACS—and they did not like what they saw. It did not integrate with the speech recognition system, making bi-directional voice dictation impossible. Unsatisfactory measurement tools, right–left orientation issues and a degree of difficulty in manipulating CT images were topped off with integration issues with one radiography vendor’s technology.

“We had workarounds for some of these deficits, but they were not enough,” Dulaimy says. “Our workflow was cumbersome, and report turnaround times were increasing rather than decreasing. We knew we could address our problems with a new PACS, but we also knew that in order to do this fully, the PACS had to support a smooth, tight integration with the RIS and the EMR.”

After a comprehensive evaluation, Baystate Health selected Synapse PACS from FUJIFILM Medical Systems U.S.A., Inc., Stamford, Conn. A key component of the solution is an enterprise image viewer that enables clinicians from throughout the health care system to access patient images no matter where they happen to be working, with the same tools and with a consistent degree of functionality.  

Primary among the reasons for choosing Synapse were ease of integration with the existing RIS and speech-recognition solutions and more functionality than any other option under consideration. Fujifilm’s strong focus on image processing and viewing also factored into the decision.

Virtual desktop infrastructure

The technology enabler of Baystate Health’s systems integration is virtual desktop infrastructure (VDI), which allows access to and management of images, data and voice via a single sign-on and click.

“VDI provided us with a visual integration of Cerner Millennium RadNet® with our PACS workstations,” Dulaimy explains. “This integration enabled RadNet functionality to display in context when a patient and action is selected from the PACS workstation.  Our radiology team has access to vital clinical information, including Technical Comments, Online Requisition, and the Electronic Medical Record (EMR).”

Improved workflow and, consequently, faster study turnaround times, have resulted, Dulaimy says. “Our RIS is part of our EMR,” he explains. “Our goal, which we have attained, was to use integration to establish smooth, bi-directional processing from the PACS to the EMR and (by extension), the RIS—with the one click making it as easy as possible. With this kind of processing, the workflow has to be better.”

The integration of PACS/RIS/speech recognition also affords the hospital system far more functionality and flexibility than ever before. Notably, Dulaimy points out, the viewer component supports enterprise-wide visibility of all images “in a single place and under one view—something that of course, we never had in the past.”

Similarly, the tight systems interface and bi-directional processing capability resulting from the project yield radiologists the level of functionality required to structure studies and create reports. For example, in setting up protocols, radiologists can leverage the one-click entrée into a unified view of the systems to quickly put their fingers on data such as patient allergies. The functionality of the integrated configuration is such that radiologists have the option to utilize a PACS-driven worklist if they prefer, without losing any capabilities tied to the RIS.

The configuration has paved the way for support of all modalities from different resources; for example, images from other PACS, such as the one used in Baystate Health’s CVIS services department, can now be accessed with no difficulty.

Integration benefits

The tight integration between information systems has many positive implications for patient care and safety. With a unified view of all patient data and images, including EMR data, there is less likelihood that a critical piece of information that could compromise patient safety (for example, study priors that reveal a certain condition or an allergy noted in an EMR) might be overlooked. Faster study turnaround ups the patient care ante as well.

With the new implementation, Baystate switched from a RIS-driven workflow to a PACS-driven workflow, and there was some concern about workflow impact. Virtual desktop integration enabled the radiology department to switch seamlessly to a PACS-driven worklist with no impact on our workflow,” Dulaimy says.

Dulaimy attributes the success of Baystate Health’s EMR/RIS/PACS/speech recognition integration project not only to the “great fit” between its components, but also to practices followed by all constituents beginning at its inception.

For example, during the pre-implementation phase, representatives of Baystate Health, Fujifilm Medical Systems and Cerner met regularly to discuss concerns and questions, provide any necessary feedback and collaborate on solutions to problems. Such meetings are still held periodically.

“Technology is wonderful, and so is solutions integration, but communication and cooperation are key to getting the work done—and benefiting from it,” Dulaimy concludes.

Julie Ritzer Ross is a contributing writer for Radinformatics.