The Children’s Hospital of Philadelphia (CHOP) was an early pioneer in establishing an enterprise image repository; its enterprise medical storage platform went live in 2005, and its vendor-neutral archive (VNA) went live in 2009, followed by an enterprise viewer in 2010. Chris Tomlinson, MBA, has helped to lead the charge in his role as administrative director for the department of radiology. “We were early adopters in the world of VNAs, and today everyone is talking about them,” he notes. “Everybody’s talking about them—but from the perspective of people who are doing it, what does an enterprise archive really look like?”
Like many in the field of imaging informatics, Tomlinson takes issue with the blanket term “vendor-neutral archive,” and in a talk at the 2012 meeting of the Society for Imaging Informatics in Medicine (SIIM), he proposed that the term “PACS neutral archive,” or PNA, be utilized instead. “Even though this starts in radiology, it’s an enterprise platform and something the enterprise owns,” he says. “Radiology is almost all DICOM, and it would be a very radiology-centric view to assume the whole enterprise will be DICOM as well.”
Stepwise Process
CHOP’s image management challenges are similar to those faced by most health care organizations: imaging is being used more regularly, in more departments, and image datasets are growing exponentially. “The number of studies is increasing, image resolution is increasing, and the number of images per study is increasing,” Tomlinson says. “Now these little silos, the departmental PACS systems, are running out of space, and from an IT perspective, the question becomes how you can best support all of them.”
The solution commonly referred to as the VNA can help, Tomlinson says, by funneling the data stored on all of these systems’ servers into one archive that is centrally supported and administered. At CHOP, implementing such a system resulted in demonstrable savings, but, Tomlinson cautions, “There’s still a vendor there with whom you have to partner. It’s not like you’re using some shareware to manage your images. There is a vendor there—you’re just segmenting the storage and ownership of the images away from your PACS vendors.”
The result, at CHOP, has been what Tomlinson calls a “stepwise process” toward enterprise ownership of medical image data. “There’s a lot of marketing spin out there about VNAs as a silver bullet. They’re not,” he says. “There are still a lot of challenges associated with enterprise archiving because it’s not all that well developed yet.”
Remaining Hurdles
For instance, Tomlinson says, specialty areas with their own PACS solutions, like cardiology, still do not have the equivalent of RIS systems to complement their image management. “The RIS manages the way things are stored by issuing accession numbers,” he explains. “In cardiology, they don’t really have anything like that, so people are having to key information into the modalities by hand. When you move out of radiology and don’t have the RIS to automate that, how do you organize the data?”
Health care organizations also need to look ahead to avoid facing similar challenges in the future, Tomlinson says. They need to prepare for the demand for multimedia archiving and be poised to participate in regional/national and large government projects. “You need an information lifecycle management policy,” he explains. “You have to ensure your image management layer is using XDS or some kind of standard so that you can prepare to interoperate across health systems when the time comes.”
Perhaps most importantly, attention must be paid to the enterprise methodology for archiving; the infrastructure utilized will have a downstream impact on everything from IT solution speed to collaboration between specialties to the integrity of data. “Think of your enterprise methodology as a bunch of planes landing at different times with different knowledge,” Tomlinson says. “Radiology is coming first, then cardiology, then OB/GYN, and so on. We have to coordinate that into one strategy.”
Data Ownership
The challenges remain extensive, but Tomlinson is clear on one point: health care organizations can no longer afford to cede ownership of their image data to vendors. Keeping data in silos is costly, he says, and impedes optimal care: “The big goal is for all of your providers to have access to all of the images. If somebody wants to see the imaging lifecycle of the patient from the EMR, they should have a way to do that.” The only way, he says, is to leverage an enterprise viewer showing images from a VNA.
This imperative is reflected in the growing market for VNAs. According to the 2012 InMedica World Market for Medical Enterprise Data Storage report, the VNA market is expected to grow by 49.9% over the next five years, while the PACS market will grow by just 5.2%. “Most people have already been married and divorced once when it comes to their PACS, and they are constantly migrating that data,” Tomlinson says. “It quickly becomes untenable.”
For these reasons, Tomlinson believes it is imperative to be clear about what a VNA is and is not. A true enterprise image archive, he says, would provide a single repository for all images in the enterprise on any storage hardware; provide standardized metadata for every image to support anytime, anywhere, any-provider viewing; and provide flexible rules for image lifecycle management to support carrying regulatory and clinical requirements.
“The market and the drivers have changed,” Tomlinson concludes. “Some of the basic tenets we were working with three or four years ago aren’t true anymore. Vendors are selling these products now, and we need to start setting expectations.”
Cat Vasko is editor of HealthIT Executive Forum.