Part II: Exploring PACS secrets presents the second part in a series by PACS consultant Michael J. Cannavo exploring commonly accepted PACS theories -- as well as industry secrets your PACS vendor might not want you to know.

Like two kids fighting over the same toy, PACS has created a battlefield between radiology and hospital information technology (IT) departments in recent years. Numerous articles have been written about who owns PACS -- radiology or IT -- from both a decision-making and support perspective. The reality, however, is actually a contradiction: No one owns PACS and everyone owns PACS.

Over five years ago, most of the PACS decisions were primarily radiology-centric and devoid of IT’s input. In a knee-jerk reaction to prevent this from happening again, IT has more often than not vowed to "take back the night" and bring PACS under their domain. After all, they say, PACS is a clinical system, and clinical systems fall under IT’s umbrella.

What many entities fail to consider is that the applications software is what really makes PACS work or not work, not the hardware. That requires input from people savvy in global radiology operations, including contributions from radiologists who will be using the system, input from techs who will be interfacing with the system daily, and various other radiology staff.

Since the software runs on hardware, this requires a partnership between radiology and IT. Unfortunately, neither entity is usually willing to give up control of the project, with each having various reasons why they feel they need total control.

This puts hospital administration in the position of playing "good mom, bad mom" and choosing sides. The easiest answer of all is to make the decision-making process a joint decision, assigning weight factors so that no one entity can have total project control.

With our clients, we typically break down the decision-making process to 40% radiology, 30% IT, 20% purchasing/materials management, and 10% administrative, with the final decision requiring at least 51% of the vote. This requires at least three of the different entities to agree on the choice.

Each entity should address the areas of their own expertise, with radiology focusing on software functionality and application, IT focusing on hardware maintenance and support (including networks), purchasing addressing contracting and terms and conditions, and administration responsible for global systems design, relationships, and similar big-picture issues.

Each group’s area should also be broken down into subsections, addressing everything from software graphical user interface (GUI) ease of use to hardware maintenance, integration strategies, and upgrade paths, etc., with all areas numerically weighted and scored.

Many observers will say that this division isn’t fair given that PACS is a radiology system and that radiology is its primary user. While this is true, PACS directly affects every hospital department, from call-center support for the primary care provider (PCP) who can’t figure out how to use Web browser software to the IT or biomedical departments that may be acting as second-tier hardware support prior to initiating a call for service to the vendor (either remote telephone support or on-site).

PACS impact on IT

PACS has a significant impact on the hospital's IT department. Not only does it impact systems at the infrastructure level (local- and wide-area network specifically), but also from a support standpoint, both service and training as well. More and more IT departments are offering to both buy and support PACS hardware, with vendors surprisingly going along with the plan.

The one exception to this is core system servers, which the PACS vendor wishes to retain control of for obvious reasons like uptime guarantees, remote troubleshooting, etc. This is a dramatic change that has taken place in the past year or so, with vendors migrating from offering only turnkey solutions to now providing software-only configurations and having the facility pay a small integration fee for the software applications.

Several factors have driven this change. Hospitals have routinely been told the hardware markup is only 3%-5% from what the vendor pays, but often times it's close to 30%-40% or more. With hardware making up 35% of the PACS cost on average (with software accounting for 50%, and integration and training 15%), a facility can save $90,000 to $100,000 in software costs for every $1 million it spends.

That in itself is significant savings, yet it is far from being the biggest benefit. Most systems hardware can be added to existing service contracts, as virtually all PACS hardware consists of over-the-counter products. Most PCs sold today carry 2-3 year warranties on them, so a facility can factor in savings of two years’ worth of service contracts into the mix, as system hardware will be covered by the manufacturer's warranty.

Furthermore, because additional equipment is typically being piggybacked onto an existing hardware support agreement, adding equipment may even drive overall service costs down for the entire hospital. At 14%-16% of the list price, this can save an additional $30,000 to $50,000 per year in service costs per $1 million spent.

Some hospitals even serve as a service depot for a manufacturer's brand of PCs (Compaq, Dell, Gateway, etc.) and can provide cheaper and better service. Even "non-standard" PC components like displays and display cards can be purchased through many of these providers, or, worst case, simply addressed by having at least one hot spare available to exchange.

Hospital IT departments need to look at both their resources and resource utilization before embarking on this approach. Without adequate resources, this tactic can backfire in a big way.

Regardless of supporting system hardware or not, IT also needs to examine the depth and availability of resources for supporting PACS. PACS requires additional network drops to be run, power outlets to be expanded, and rooms redesigned. Lighting also needs to be changed (overhead lights are replaced by indirect lights), walls typically are painted a dark color (so monitor reflections aren’t seen) and several other related tasks must be performed, all either coordinated through or done by IT.

IT is also responsible for the integration of PACS with other clinical systems. This is a sizeable project, and can take months to do properly. It also requires a clear project plan delineating who is responsible for what tasks, including the PACS vendor, the hospital IT and radiology departments, and also the RIS vendor.

Getting cooperation from the RIS vendor and PACS vendor is no small task, especially if the RIS vendor bid on and lost the PACS deal. It can be done, however. Again, it all needs to be laid out on paper ahead of time with agreed-upon implementation time frames

Unfortunately, penalties for failure to meet implementation time frames in this area are difficult to put together, as there are a plethora of variables that fall into play and a failure by one party to meet a deadline impacts everyone else’s. Still, with a proper project plan with reasonable timelines, it's doable.

ROI facts and fallacies

It's fascinating to look at all the return on investment (ROI) models out there relating to PACS. There are at least 10 PACS myths, yet the most damaging ones relate to ROI.

PACS can yield an ROI in one of two ways: expense reduction or productivity gains. I would venture to say that 90% of all models focus on reduction in expenses, yet better than 60% of the ROI actually comes from increased productivity due to process changes from PACS.

Tangible factors such as film reduction need to combine with intangible yet documented savings relating to process changes to make up the ROI. A properly designed ROI analysis also factors in both types of factors and includes added expenses such as service contract costs (if elected), FTE costs, internal charge backs, department renovations, and the like.

Virtually every model talks about film reduction from Day 1. Will PACS reduce film from Day 1? Yes. Will it be 100%? No. Typically, film savings are phased in over a two- to three-year period using 80%, 90%, and 95% reductions in film use or thereabouts. Why not 100%? Because that’s unrealistic.

Few facilities are ready to make the jump to digital mammography, so at least 5%-7% of the film budget needs to be reserved for mammography film alone. To achieve even an 80% filmless rating or higher (even grudgingly so), the facility also needs to define an etched-in-stone date when filming will be turned off.

The use of both Web servers and CDs with embedded DICOM viewers will help to significantly reduce the amount of filming, but it's always best to maintain a percentage of film in your budget until Year 3 just in case.

Folder costs and labeling also play a role in the ROI, but not a dramatic one. The same can be said with mailing and processing of outside films. FTE reductions in ROI models typically are shown in the form of bodies, not total operating dollars.

Unfortunately, PACS will probably increase the FTE operating budget unless there are specific areas directly related to PACS that will be eliminated (i.e. dedicated film hangers). Some file-room clerks will be eliminated, but for every two file-room clerks you eliminate, their salary is offset by a single systems administrator.

Storage costs too are usually factored in and also added with a cost value. Unfortunately, with the exception of film storage space used in the main radiology department, most of the space used for storing films is considered dead space anyway and has negligible value (the basement storage area, etc.).

Off-site storage and courier costs certainly can (and should) play into the mix, but it’s a very fine line one walks to include space savings, even if the space used for film storage within the department can be changed from non-revenue-generating to revenue-generating space.

PACS itself won’t impact report turnaround time, because the bottleneck typically isn’t the image viewing, it’s the signing of reports after they have been dictated. This is primarily a RIS function, not PACS, although certainly PACS can augment turnaround time.

It's important to note that just because you save time and effort with PACS and can potentially perform more studies with the same number of bodies doesn’t mean the studies from the PCPs will automatically come to you. Marketing costs to attract PCPs to the facility are almost always excluded from ROI models, yet are an equally critical component. Typically, 10% of the net projected savings should be set aside for marketing alone.

Support costs, too, are almost never factored in. These issues were addressed in Part I of the series, but not only do you have service contract costs and time and materials costs, but internal charge backs as well for IT and biomedical support.

Ultimately, an ROI for PACS isn’t what's really required to gain acceptance by hospital administration, but rather a PACS business case. While a ROI may address only initial costs, hard benefits, and projected payback period, a cost model takes into account costs, benefits (hard and soft), five-year departmental growth by modality, competitive and reimbursement-related issues, ongoing costs including upgrades, as well as three-, five-, and seven-year ROI including various "what if" scenarios. This gives a facility a much better "feel good" than just seeing number crunching on paper.

By Michael J. Cannavo contributing writer
July 6, 2004

For "Part I: Exploring PACS secrets," click here.

In "Part III: Exploring PACS secrets," Mr. Cannavo will tackle topics such as working with PACS consultants and PACS contract tips.

Michael J. Cannavo is a leading PACS consultant and has authored over 250 articles on PACS technology in the past 15 years. He can be reached via e-mail at [email protected].

The comments and observations expressed herein do not necessarily reflect the opinions of, nor should they be construed as an endorsement or admonishment of any particular vendor. Rather, they should be taken as the personal observations of a guy who has, by his own account, been in this industry way too long.

Related Reading

Part I: Exploring PACS secrets, May 14, 2004

PACS still a bridesmaid at HIMSS, February 27, 2004

The year of the Uni-PACS: A view from the RSNA floor, December 11, 2003

The PACSman’s opinionated view from RSNA 2003, December 4, 2003

The PACSman’s opinionated view from RSNA 2002, December 4, 2002

Copyright © 2004

Page 1 of 775
Next Page