Perhaps a little like planning a house renovation or even a second marriage, the second time around can be more complex when it comes to PACS. Implementing a next-generation PACS can bring the same angina, but with calculated planning, the benefits are quite predictable. With concerns ranging from moving data from the old system to the new and avoiding system down time, replacing a PACS needs extensive planning and careful execution—and a team approach to assure success.

What can institutions looking to move to a next-generation PACS expect to find in the way of differences between new and old? Terence Matalon, MD, chairman of the department of radiology at the Albert Einstein Medical Center in Philadelphia, believes that one of the things that distinguishes a next-generation PACS is an ability to “accommodate different, multiple other applications. The openness of a system to work well and interface with other applications is being viewed as an essential necessity,” says Matalon. “Particularly since no single [PACS] vendor has been able to meet all the needs of the various institutions.”

Some of the applications, Matalon says, are related to things like Joint Commission requirements from the standpoint of critical test results reporting, while others might be related to issues surrounding peer review, or how well a voice product interacts with a radiology information system. What is critical is the system’s ability to interface with other systems across the enterprise, he says, “and a lot of companies that previously took more of a proprietary approach, now recognize that there are a lot of niche products that people may want that aren’t necessarily within [the vendors’] repertoires.”

Matalon adds that the incorporation of a robust web product also is the mark of a next-generation PACS.

For Mark Watts, corporate imaging system manager at IASIS Healthcare Corporation in Franklin, Tenn., the question isn’t only about what a next-generation PACS can offer, but what a first generation PACS can’t. “Our [previous] PACS product just never developed along with the changes in healthcare, and interoperability became an issue,” he says. “It was like driving an older car, the PACS didn’t seem to be able to grow with the needs of our organization.”

System reliability and stability

In early 2009, Iowa Health System was rapidly approaching the breaking point with its first-generation PACS. The health system, which comprises 34 hospitals and clinics and is divided into seven regions within the state, was experiencing some severe “reliability” issues, according to Joy Grosser, Iowa Health’s chief information officer. “Our system was growing and the PACS itself was becoming unstable,” says Grosser. “It wasn’t giving us enough uptime.”

“The system was often down, and that obviously impacts patient care,” says Tony Langenstein, IT directory of technology for Iowa Health. “We had asked for—and worked with the vendor on—enhancements, but those efforts just came to naught.” An example of a problem the system faced, he says, was an effort to improve the old PACS web product “that was just horrendous.” Iowa Health worked with the vendor to upgrade the product, but right in the middle of the process, Langenstein says, the vendor purchased an entirely different web product that negated all of the upgrade efforts.

While Iowa Health had identified the problems, and knew what it wanted to accomplish with the new generation PACS, it faced one major impediment—time. The old PACS had basically reached the end of its effective life and needed to be replaced immediately. “We had kind of painted ourselves into a corner,” Langenstein says. “We had waited so long that we were forced to do something”

According to Grosser, the system had been having internal discussions—involving physicians, administrative staff and IT personnel—to identify ongoing problems with the old PACS and to determine what a new PACS needed to provide. In the spring of 2009, the system decided to move—fast. And a selection process that began in May had, by the beginning of August, resulted in the selection of a product and the beginning of a system-wide installation.

“We wanted reliability, and a partnership that would allow us to see enhancements throughout the PACS product, something that was fast and would provide good web access right out of the box,” Langenstein says.

The expedited process—from sending out RFPs to the final PACS installation—took about six months, or about as long, Langenstein says, as the vendor usually spends to go live at one specific location. “This was described as a singular event,” he says.

The results? While it’s still early, Grosser says that one immediate benefit has been the ability of radiologists to customize hanging protocols and improve workflow. And Langenstein points out that the web product is vastly superior to that of the old PACS, which had a web product that was so slow that physicians refused to use it. “Now we have physicians using it, and they’re starting to like it,” he says. “So from a referring physician perspective, that will definitely improve workflow.” 

As for the main driver behind the decision to go to a next-generation PACS, Langenstein says that “although there may be some growing pains, from a reliability perspective I’m positive we’re going to see some huge benefits.”

Defining a Next-Generation PACS

Mark Watts, corporate imaging system manager at IASIS Healthcare Corporation in Franklin, Tenn., says a next-generation PACS should feature the following: Adaptive image loading: The first image of a data set is viewable before all the images are loaded On demand and user controlled compression ratio manipulation. Let the user decide how fast of a down load is needed for remote viewing Voice clips and documentation scanning to associate the images with an order and provide a “virtual radiologist consultation” for the referring physician with a recorded preliminary report via a portal or image enabled EMR HIPAA-compliant access tracking Interoperability with other clinical IT systems Web-based thin client for mass distribution of remote viewing capabilities Redundant hardware and remote support capabilities Clinical series mapping. Prior images of the same body part are presented for comparison regardless of modality Integrated 3D images and voice recognition for remote final reading Integrated communication, email voice over IP (VoIP) Critical test values reporting documentation Teaching file and peer review documentation

Solving the ‘niche’ problems While the implementation of a first-generation PACS can immediately provide obvious workflow and cost benefits (i.e., money saved by eliminating film), Watts and his colleagues at IASIS asked themselves what benefits their 16-hospital system would gain by installing a new generation PACS.

With a first-generation PACS, Watts says, users are immediately grateful for those tangible benefits that are seen quickly after deployment. But, with a new PACS, the prospective user needs to understand “all of those little niche problems” the previous PACS hasn’t been able to solve: “How will it incorporate digital mammography? If we want to have a voice clip associated with a study and we want the physician to be able to listen to it, how do enable that? How do we take a document from outside [the enterprise], and associate it with an image so that we can have a permanent medical record that justifies a payment for a procedure?”

Asking these questions and determining whether a new PACS will deliver the desired benefits is a critical step in determining what new generation PACS to deploy.

Another key question to ask, Watts says, is how successfully the current PACS is in meeting the healthcare organization’s needs. In IASIS’s case, an assessment of the system’s workflow situation as it existed with its first generation PACS suggested there were some problems. “We did a walk through at one hospital and it turned out we were still spending $40,000 a year on film there,” says Watts.” Now, those two things do not equate. If you have an effective PACS in place, why do you continue to produce films?”

A further look into the old PACS demonstrated that—much like the problem faced by Iowa Health—the web product associated with the PACS was inadequate. “The web product we were offering to clinicians wasn’t as robust as we wanted,” says Watts. “And every time they wanted to use the system, it would take 90 seconds for them to log in. So that dissuaded them from using the system.”

It’s these kinds of inefficiencies, says Watts, that indicate that any search for a next-generation PACS solution has to take into consideration that a PACS “just doesn’t exist solely within a radiology department, but has to be integrated with a portal, or an EMR that is image-enabled, or that has a web product that allows a specific customer to interact with that product.”

Like Iowa, the selection and installation process was expedited at IASIS, with the ultimate goal getting PACS up and running throughout the system within 60 days. In deploying the system to 16 different hospitals, it was important to maintain some flexibility while maintaining the system consistency that was necessary across the enterprise in order to achieve the anticipated benefits of the new PACS, Watts says.

While IASIS may have installed the same hardware  throughout the system, or set up and configured remote access the same way in each of its facilities, it could, for example, make allowances for radiologists who preferred a two-monitor workstation set-up, rather than one with a single split-screen monitor.

“I’m not going to take dual monitors and replace them with a single, if it means creating a workflow problem for the radiologist,” says Watts. “Part of my job is understanding how that transition [from old to next-generation PACS] works, and then make that customer experience just as good, or better.”

Migration challenges

One issue that is going to come up in the transition from an old to new PACS is that of data migration. The deployment of a first-generation PACS involves the relatively straightforward—if potentially lengthy—process of turning film into digital images for archival purposes. But, with a second-generation PACS, that patient history has to be transferred from one system to another, and, according to Watts, that transition has to be fairly quick and seamless if clinicians are going to have continued access to the information.

When considering the process of deploying a new generation PACS, hospitals also need to take into account that a slow transition can be costly. Watts says a slow transition means the facility will have to incur the expense and inefficiency of running two PACS at once until that information is transferred. 

For Iowa Health, the PACS installation process ran from August through December—an “unheard of” quick pace, according to Langenstein. And with over 60 terabytes of data to transfer, the biggest hurdle, by far, was that of data migration. The process was going so fast, Langenstein says, that the storage subsystem was unable to archive all of the data as fast as they were being migrated and started causing performance problems for the production PACS. The solution was to transfer two years of data during the installation process and then finish the data migration process after the last “go live” was completed in December.

As for IASIS, Watts says, “we were able to take one mission critical system out, and replace it with something more effective, while transferring 3.2 million studies and not having a second of down time.” All of which means, he says, that IASIS was in a better position to quickly realize the benefits that can accrue with the installation of new-generation PACS.

Benefits that include, for example, the ability of radiologists at one hospital in Utah to turn final reports around to emergency room physicians in an average of 14 minutes. “Twenty-four to 48 hours used to be the standard for turning out final reports,” Watts says. “Now by the time the patient gets to the CT scanner, has the images acquired and sent to the radiologist, and gets back on the gurney, there’s a final report in the PACS.”

Watts also has found that the installation of the new PACS has had an impact on radiologist job satisfaction, leading to a drop in radiologist turnover.

“We gave them a preferred tool to work with and they have responded by choosing to stay with our team versus testing the free agency market,” Watts says. “The grass isn’t necessarily greener on the other side of the fence.”