Toward True Globalization: The Air Force and PACS
Sharing images across any health care enterprise represents a challenge, but doing so across the Pacific Rim was the dilemma faced in 2003 by the US Air Force. Taking up this challenge were Lt Col Grant Tibbetts, MD, now radiology consultant to the surgeon general, and Tom Lewis, the director of the Air Force PACS Office. “The largest hospital in the area was at Elmendorf Air Force Base in Alaska,” Lewis recalls, “so at the time, we had to fly radiologists from there to Japan or Korea when they were needed. Solo radiologists within the Western Pacific needed support when they went on vacation or were out of pocket for various other reasons. That’s what led us into trying to get a PACS solution that would allow a global enterprise view and sharing of the workload.” At that time, the Air Force had implemented PACS at numerous sites around the country, some of which were set up to communicate with smaller sites using a hub-and-spoke approach designed to create efficiencies in staffing. In the Pacific, out of nine imaging sites, there were radiologists at four, with strategic realities often leading to staff reduction at smaller sites. During radiologist absences, these were each supported by the only site that had more than one radiologist: Elmendorf Air Force Base. The Elmendorf PACS was several years old and difficult to support by 2003, Tibbetts says, and its continued operation was viewed as an increasing risk. Meanwhile, other sites in the Pacific were in the queue to get their first PACS. “The Air Force had a few projects on the books, spread out over a few years for budgetary reasons, that would have brought PACS to those sites that didn’t have it,” Tibbetts says. “I needed to get Elmendorf moved forward in priority, and combining the projects was a way to do this. Better, our idea to implement an enterprise-wide PACS across the Pacific not only saved us a lot of money, but also offered us the opportunity to connect our sites in a way that the Department of Defense (DoD) never had before.” Unique Needs Ten proposals for new PACS were initially received, but Tibbetts, Lewis, and their team of clinical, IT, PACS, and administrative consultants soon truncated that list. “We quickly whittled down to two capable vendors,” Tibbetts says. “The others simply weren’t ready to meet our security requirements, and were less eager to embark on a challenging project spread across 65 million square miles of water, nine sites, and three countries.” The first obstacle faced by any vendor to the Air Force is the DoD’s stringent security requirements. FUJIFILM Medical Systems USA, Stamford, Conn, worked with the Air Force to qualify their Synapse PACS platform for the project, becoming the second qualified vendor on the Air Force network. Both vendors continue to be important partners today, Tibbetts says, and several other vendors have approached security testing favorably in recent years. After security, another requirement of the prospective Pacific PACS was access to relevant prior studies across the enterprise. “We were attracted by the ability to serve these up in a compressed format,” Tibbetts says. “The final two PACS vendors we considered both offered selectable wavelet compression. We knew that might be very beneficial for what we were trying to do in a sometimes bandwidth-constrained environment.” Tibbetts later went on to write supporting policy allowing for and establishing limits on compression use in DoD for primary interpretation. Tibbetts and Lewis wanted a user interface that was intuitive, and one with which the Air Force’s IT professionals could become quickly familiar—“for better or worse,” Tibbetts says. “Windows is not the be-all, end-all for everyone, but it’s simple and immediately familiar to most, quickly giving you basic functionality with little training. We really liked the fact that the clinicians’ interface was the same, too. At the time, Synapse was one of few products offering this. Today, almost all PACS systems mirror most of the radiologist’s interfaces, providing the same look and feel for their clinician counterparts.” Finally, the team knew that it needed the ability to build a worklist that could be shared across all nine sites—another uncommon characteristic of PACS platforms at the time. “If I was covering a site other than my own, I could bring it up in a combined worklist, or open up another instance of Synapse, which thicker clients didn’t allow,” Tibbetts says. “This functionality gave us more options. We were enthusiastic about the freedom to share work both ways and cover calls across time zones.” Patients on the Move Air Force patients are constantly on the move, both within the Pacific Rim and around the world. With respect to the Pacific region, Tibbetts and Lewis thought it was important to be able to access relevant prior studies from any base, which was first done in the Pacific. Now, the Air Force’s PACS team is pursuing a global archive for all Air Force bases; soon, all medical facilities in the Air Force will feed their images into five regional archives. “That way, each PACS only has to reach out to five locations in searching for priors,” Lewis explains. “We can’t ask it to query more than 70 independent sites, and we don’t want to maintain that many archives. We saw the future and decided to limit it to five.” This helps patients and providers, too. “It’s very common anywhere in the United States that a patient seen in two different hospitals in the same town doesn’t have prior images available, and it can be a hassle to get them,” Lewis says. “Instead, imaging is often repeated. We have had the same problem, but spread around the globe.” Meanwhile, radiologists wrestle with more than 60 functionally distinct RIS platforms across Air Force facilities. Fortunately, completed reports can be seen by clinicians anywhere in the DoD using the current Air Force electronic medical record, which has only been fully implemented for three years. “The central database is slow, but it’s functional,” Tibbetts says. “It would be useful to have the image to go with the report, a tool we have in the works, but sometimes, words alone are enough to avoid repeating studies this way. A picture’s worth a thousand words, but a hundred words are still better than none. This was a huge step.” Toward Full Interoperability With so many RIS platforms and a handful of different vendors’ PACS to bring together across DoD, the service’s PACS teams till have their work cut out for them. Tibbetts says, “We’d like to link with our Navy and Army colleagues’ facilities more extensively, not just in RIS, archiving and clinical image access, but in workload sharing across all sites. We still have security, administrative, and technical challenges to overcome, but there are regional or vendor-based opportunities short of this global goal that are working toward these horizons. In the Pacific, for example, joining our Synapse sites together with those the Navy more recently implemented is something I’m very interested in doing to mitigate resource constraints and improve patient and clinical care.” Lewis notes that enabling the electronic health record to handle images is a parallel effort that might bring the DoD much closer to this goal. Tibbetts, too, emphasizes not only the resource benefits of interoperability, but also the improvements in patient care that it will bring. “Sometimes, tracking down images can be a big challenge,” he says. “We’re further down that road, in some ways, than our US civilian counterparts, but we’re still fighting for more. If we’re going to do the best possible care, we need to know what was there before and what has changed, avoiding reimaging the patient when we can, reducing cumulative radiation dose, and ultimately, improving diagnostic power.” Cat Vasko is editor of and associate editor of Radiology Business Journal.