Strategic planning key to PACS installations at imaging centers

Multi-imaging centers demand a short-term return on investment and accept only hard-dollar cost justifications -- a tough bar to clear for any systems implementation. With a PACS installation, these requirements can become even more challenging.

With proper planning and oversight, however, these goals can be attained, said Gary Reed, president and CEO of Integration Resources, a Lebanon, NJ-based RIS and PACS integration, management, and consulting firm. "A successful outcome to a PACS installation is proportional to the degree of specification," he said.

Reed, speaking at the American Healthcare Radiology Administrators (AHRA) conference in Nashville earlier this month, observed that as more and more hospitals embrace PACS, for-profit multi-imaging centers are starting to follow their lead. Unlike the caregiver focus of hospitals, imaging centers are knowledge-based service businesses, he said. As such, their processes emphasize a timely distribution of radiologists’ reports and images over a wide geographic area. And, with approximately 1,800 imaging centers in the U.S. merging and consolidating on an almost daily basis, Reed said, competition is fierce and margins are slim.

To keep costs low, many imaging centers have an installed base of used equipment. While serviceable for their current needs, this older equipment has to be DICOM-compliant to successfully integrate into a PACS. "You need to get a mandate to leverage information technologies to improve productivity and profitability -- and that requires strategic planning at the senior management level," advised Reed.

There are nine desired outcomes that Reed believes define PACS strategic planning for imaging centers. They include giving radiologists the ability to:

  • Access worklists and studies from any site.
  • Send selected images to a referring physician, along with a report.
  • Autoroute a subset of historical studies to specific workstations.
  • Access comparative studies by query.
  • Reduce file space at imaging centers and convert to productive space
  • Download patient data at the modality and track exam status.
  • Perform QA and peer-review overreads.
  • Balance radiologists’ workloads.
  • Provide overreads and subspecialty consultations via Web technology

To realize these goals, the existing imaging environment must be analyzed, and the center's business processes examined for areas that a PACS will make redundant. These should include courier expense, film storage and retrieval, film and supply costs, radiologist travel time, and data entry, Reed said.

Equipment assets must be inventoried by vendor and model number at every site to determine which machines will need to be replaced and which can be integrated into a PACS. Not only must all the image output equipment be DICOM compliant, the equipment must also support the DICOM services of storage class and storage commitment user, modality worklist user, print user, and performed procedure step provider, he said. Exam volumes and output per-machine should also be scrutinized -- not only by day, week, month, and year -- but by site and modality as well.

The center’s existing network infrastructure must also be assessed in detail. The assessment needs to account for current network traffic and load balancing, as well as projected data throughput from each machine in each center; and for a 24-hour period, weekly, monthly, and annually. From this analysis, telecommunications and PACS requirements can be specified, and cost estimates for these services and systems can be generated.

Archive capacity and retrieval play a crucial role in any PACS; for a geographically diffuse multi-imaging center, the image archive is arguably the most vital piece of the network. Because an archive is so critical to business success, Reed maintains that a multi-imaging center should specify a reliability uptime rate of 99.9% -- a figure he believes can only be achieved with a Unix-based system.

The center's managers will need to be involved in the archive architecture design, because only they can determine their radiologists’ need for near-term (RAID) vs. long-term (tape or other permanent storage media) image access. Management will also need to take into consideration the costs associated to converting their historical studies to digital media. In addition, redundancy and fail-over mechanisms will need to be explored for both the archive as well as the telecommunications portions of the PACS.

Imaging centers must insist on acceptance testing for requirements as part of the contractual obligation of all their vendors’ products in a PACS. Because a PACS network is comprised of many parts -- network, telecommunications, archive, database, DICOM, imaging modalities, and viewing stations -- downtime in one part of the system may well lead to downtime in other areas. Since this is not an acceptable option, imaging centers have to clearly obligate vendors to implementation, support, and uptime parameters, with financial penalties associated to their failure.

Reed noted that centers should expect a six-month crunch time when a PACS is first adopted because there will be a blend of film and digital imaging in the workflow. A good solution, according to Reed, is to start digitizing film first and to bring the archive system online early in the process.

Multi-imaging centers can offset the capital expense of PACS acquisition by aggressive implementation.

"The sooner you turn off the costs to film, the sooner you will see system amortization with a PACS. If you have both (PACS and film) running simultaneously, then both are cost centers," Reed said.

By Jonathan S. Batchelor staff writer
August 21, 2000

Let know what you think about this story.

Copyright © 2000

Page 1 of 775
Next Page