Rush-Copley Medical Center in Aurora, Ill., epitomizes the busy community hospital committed to leading edge technology. Six years ago, the 183-bed hospital decided to deploy PACS. In addition to PACS, the hospital was in a massive acquisition phase, deploying or planning to deploy several other technologies including voice recognition, clinical documentation, advanced radiology and cardiology imaging, wireless applications and computerized physician order entry.
The hospital’s position as a technology leader made for a tight budget. PACS represented a major capital outlay, so the Rush-Copley leadership decided to think creatively and eventually decided to go with GE Healthcare’s Centricity PACS ASP model.
Under the ASP model, GE Healthcare assumes the burden of long-term image storage; the vendor also assumes responsibility for disaster recovery and redundancy and data migration. The Rush-Copley decision was driven by financial considerations, explains CIO Dennis DeMasie. “PACS is a significant capital expense, and a lot of our money was going into bricks and mortar. We needed a different funding model; ASP allows the hospital to pay for the capabilities on a monthly basis without owning the hardware or the product itself. We basically expense the system as we use it.”
In the six years since embracing ASP, Rush-Copley has discovered that financial necessity is a winning combination, with ASP delivering numerous benefits. In addition to eliminating the initial hefty outlay, ASP:
- Removes long-term archive management duties from the local site;
- Incorporates technology upgrades and data migration into PACS funding; and
- Provides a robust disaster recovery strategy.
The ASP model under the microscope
The ASP model differs from the traditional buy-and-own image archive plan, but it is a model that works and enables efficient, cost-effective imaging operations.
When a technologist acquires an image at Rush-Copley, the image is stored in an onsite, short-term archive. Images are available for immediate review by radiologists. The radiology department utilizes a 4 terabyte (TB) RAID for short-term local storage. “The RAID holds two to three years worth of images,” says Tom Markuszewski, director of imaging. The other, equally important piece of the puzzle, is long-term storage.
The long-term image storage strategy means that PACS sends the image sets to two separate, offsite data centers in Chicago and California, which serve as the long-term archive and disaster recovery solution. The benefits of this model are significant, particularly when it comes to long-term storage.
Hospitals and imaging centers are required to retain radiology studies for a minimum of seven years. What’s more, file size continues to grow exponentially as sites deploy new systems such as 64-slice CT and digital mammography, both of which can produce very large data files. The upshot? Long-term digital image storage is a challenge on multiple fronts — including finances, planning, staffing and management. The ASP model shifts this burden to the vendor. “Our long-term storage is infinite, yet we pay only for the storage we use. We know that we will never run out of storage. The challenges associated with long-term storage remain invisible to us,” explains Markuszewski.
That flexible long-term archive associated with ASP has come in handy at Rush-Copley Medical Center, which has seen its image volume increase from 40,000 to 120,000 studies annually since deploying PACS in 2001.
The disaster recovery bonus
Disaster recovery and PACS implementation travel hand in hand. Healthcare facilities are required to implement a redundant archive; a single long-term, on-site archive does not suffice. The challenge is immediate with a PACS deployment, says Markuszewski. “Hospitals don’t have the luxury of implementing a PACS one day and identifying a disaster recovery solution at a later date. The disaster recovery solution must be identified from day one.”
Developing an internal disaster recovery strategy represents a tremendous undertaking as the hospital must purchase, site and manage the storage solution and ensure that images can be recalled readily in the event of a disaster. “It is very difficult for hospitals to provide a level of service that matches that of a vendor,” explains DeMasie, “GE can put together economics and scale to handle disaster recovery for our system and several others. Our partners at GE handle disaster recovery much more effectively and much more thoroughly than any individual hospital could.”
ASP: Myths vs. reality
Some facilities shy away from ASP because of a few lingering myths. Two primary myths that may make ASP a tough sell at some hospitals are easily dispelled.
The first relates to cost; it may appear that an ASP solution costs more than an onsite storage system. “The notion that ASP costs more than a non-ASP model is untrue,” asserts DeMasie. But sites need to realize that ASP pulls double duty; it provides both a long-term archive and robust disaster recovery solution. In addition, there are hidden additional costs that can arise as hospitals assume long-term archiving and disaster recovery responsibilities. “When hospitals consider all of the factors and costs involved in disaster recovery and long-term archiving, ASP and site-base systems are about even. The added bonus on the ASP side is that it is a complete turnkey operation; hospitals don’t need to invest in in-house labor, equipment or maintenance,” explains Markuszewski.
DeMasie adds another economic consideration to the equation. Budgeting for PACS through the operating budget enabled Rush-Copley Medical Center to implement leading-edge technology without spending capital, so the hospital deployed PACS years earlier than it could have under a traditional ownership model. Plus, the ASP model may leave room in the imaging budget for other capital purchases.
Related economic issues include data migration and technology updates. The unfortunate reality of any PACS is that it will need to be replaced at some future point in time. Replacement can become tricky if the hospital wants to switch from one vendor’s solution to another as migrating a long-term archive is a complex undertaking that adds to the cost of a replacement system. “On-site, long-term archive decisions sometimes make it very prohibitive to change PACS vendors,” confirms Markuszewski. ASP stores an unlimited bank of DICOM images to eliminate concerns associated with moving the long-term archive. The ASP model also incorporates technology updates, so moving from a tape archive to a spinning disk or from one type of disk to another is a cost-neutral process.
The other myth surrounding ASP relates to image retrieval time. That is, radiologists may reject the model, believing that it may take too long to retrieve older studies from the offsite long-term archive. Rush-Copley, like other ASP sites, has found that retrieval time meets radiologists’ demands. Developments in the telecommunications world, including speed of lines and bandwidth, combined with a solid pre-fetch algorithm make access speed a non-issue, says DeMasie. At Rush-Copley, radiologists must pull images older than two and a half years from the long-term archive. The radiology department set rules and algorithms to automatically pre-fetch needed images from the long-term archive, so older studies are pre-fetched and wait in the background for the radiologist. For example, if a radiologist requires the most recent CT and the study is four years old, the algorithm pre-fetches the study as soon as the current CT is complete. “The study is in the queue and waiting for the radiologist,” sums Markuszewski. “As long as sites size the short-term archive appropriately and set well-defined pre-fetching algorithms image retrieval is satisfactory,” he says. “Your PACS vendor should have a formula that recognizes average file sizes for each modality. The institution provides the projected annual volumes for each modality, for the next five years. The vendor inserts the institution’s volume projections into the formula in order to appropriately size the short-term archive.”
The ASP checklist
Deploying PACS and ASP is a major decision; hospitals and imaging centers need to thoroughly assess the pros and cons of ASP and other models before committing to a model. Decision-makers should weigh the following considerations:
- What does the site gain by deploying PACS through the operating budget? Can it deploy digital image management sooner? Does it stretch capital dollars and allow for other critical investments?
- How robust is the disaster recovery strategy? “Take a site visit and look under the cover of the disaster recovery plan [and compare it to internal structures],” recommends DeMasie.
- Does the hospital have access to the staffing resources needed to manage the long-term archive and disaster recovery?
Words from the wise
ASP has been a success at Rush-Copley Medical Center, enabling the leading-edge community hospital to maximize its budget and deploy state-of-the-art systems in a cost- and time-efficient manner. The long-term archive is robust and elastic, able to grow with the medical center and keep up with its increasing needs. The disaster recovery plan is solid, and requires minimal internal planning or management. In fact, the medical center not only plans to remain with an ASP model for PACS, but also is considering the ASP model as it upgrades cardiology imaging systems and other major clinical systems currently hosted in-house.