PAC Pays (Really)

Timing may not be everything, but it certainly played a major role in several facilities’ implementations of picture archiving and communications systems (PACS). When the facilities were moving to new sites, the time was right for them to incorporate PACS into their radiology departments.

For others, going filmless was a matter of coming up with a PACS plan with some built-in financial flexibility.

For an early PACS adopter like the VA Maryland Health Care System, the numbers are in and financial justification abounds. No one at the facility envisioned when they approached the government in 1991 with the need for PACS for its new VA facility in downtown Baltimore that financial savings would result. At the time, clinical justification (higher quality and more accessible images) was sufficient reason to make the investment. They didn’t try to justify it economically. In fact, those who championed PACS for the VA never expected to see cost justification. Their best guess was that PACS would be more expensive than a conventional film-based department.

Armed only with patient-care rationale and the facility’s own misconceptions regarding cost, the PACS implementation proceeded, patient care progressed and the overall cost of running the radiology department became significantly lower than originally anticipated.

Some questions remain, however: Are the Baltimore VA Medical Center’s findings unique to this particular teaching, high-volume, highly expert environment? Or can others truly duplicate the increased efficiencies and improved outcomes that translate into savings? And, what roles do depreciation, capitation and fee for service play in a facility’s financial results when it comes to PACS? Read on and you’ll see.

The Real Numbers
The Baltimore VA, which went live with PACS six and a half years ago and performs approximately 85,000 radiological studies per year at the 300-bed hospital, collaborated with Johns Hopkins University’s School of Public Health (Baltimore) on a formal study that looked at the numbers for personnel, supplies, assessment of space, and procedure volumes. The purpose of the study was to provide a detailed economic analysis of the costs and benefits of the filmless operation (including the radiology and nuclear medicine departments) based on data collected at the Baltimore VA Medical Center. The study would determine whether crossover or “break-even” points exist and the volume of studies required for a PACS to become cost effective.

The study compared the Baltimore VA’s data with that from the film-based Philadelphia VA Medical Center as well as with data from the national VA system. It revealed that PACS delivers increased efficiencies, improves patient care and a few surprises – not the least of which is that more radiological studies result from the availability of PACS.

“We tried to assess what our cost would have been had we stayed in a film-based environment,” explains Eliot Siegel, director of imaging and associate professor of radiology at the University of Maryland School of Medicine. “We factored in that all departments had become somewhat more productive over the last few years as places became leaner and meaner.”

Taking the productivity factor into account, the VA tried to figure out what its costs would have been had they not made the transition to PACS and compared those costs to the current costs with PACS. Siegel said they essentially estimated what the savings were, and actually tended to underestimate the savings. For example, they did not include soft savings in clinician time, increased throughput in the ER and clinics, potential for reductions in patient length of stay and faster report turnaround time (1 to 2 hours until a report is typed into the hospital information system/radiology information system), which may improve patient care.

“Our report turnaround time dropped from two days essentially to two hours,” Siegel says. “And the time from when we do a study to when we read it has dropped [between] 15 to 25 minutes in general. So there are all sorts of soft benefits that you’d think would translate into economic benefits, but it’s impossible to assign a dollar figure, even though we spent a lot of time, effort and dollars in trying to do that.”

Personnel, both radiologists and technologists, became much more efficient, creating major savings. The average clinician at the facility estimates that he or she saves 50 to 60 minutes per day because of efficiency gained by the PACS.

The bottom line for the Baltimore VA is that the institution is saving about 25 percent per unit study by operating with soft copy, which translates into between $800,000 to $1 million per year because of the transition to the then $7-million dollar PACS. (Siegel estimates that if the facility were to buy the system today, the cost would be approximately $2.8 million.) Their breakeven point was 38,000 studies per year. Doing more than 38,000 studies makes it less expensive to operate in a filmless environment. At about 85,000 studies, they are approximately 25 percent cheaper per unit study in comparison to a film-based department, depending on which depreciation model you use.

The crossover point for the 8.8-year depreciation takes place at about 29,000 studies per year. The unit cost, depending on how many studies you’re doing, drops from about $62 in a filmless department with a volume of about 20,000 to about $34 if you’re doing 100,000 studies annually.

Using the five-year depreciation model, the crossover point takes place at about 38,000 studies. The unit cost drops from about $80 if you’re doing fewer than 20,000 studies per year to about $47 doing 100,000 studies. The reason for that drop is that the film unit costs drop from about $63 to $59 going from around 20,000 to 100,000 studies. So the film cost stays relatively even, even if the volume increases.

The depreciation demon
The Hopkins study took into account both depreciation and the facility’s service contract on its PACS, using a conservative approach to estimate its savings.

If you look at a five-year depreciation on the PACS bought in 1992, essentially by 1997 the value of the system would have depreciated down to zero. But the system with its replacement and upgrades of monitors, software and other equipment has limited the depreciation and is similar to what it would be if it had been purchased two months ago. “The PACS we have today is as functional,fast and productive as if we had bought it last month, so in some ways we are double counting the expenses because we’re counting the maintenance and we’re counting depreciation of the system down to zero,” Siegel explains.

Considering the major expenses (the depreciation and the maintenance contract) Siegel says is another indication that their model leans to the conservative side, tending to underestimate the savings from PACS rather than overestimate them.

Volume speaks volumes
If a facility’s volume is small, its up-front equipment expenses for PACS is still relatively high. If a hospital’s volume increases, its unit cost stays about the same. Storage is relatively inexpensive. “[The computer system] doesn’t care whether you’re storing 5,000 or 50,000 studies, so the cost of the PACS doesn’t really vary that much with the volume of studies that you’re doing,” Siegel says. “The cost of film, the cost of people and everything in a conventional department varies pretty much proportionately with the volume you’re doing. Whereas with the PACS, once you’ve got the equipment, increases in volume don’t result in significant increases of cost to run the department.”

Volume is an important element for consideration by those facilities in a capitated environment investigating going filmless. The Baltimore VA found that utilization of radiology services seems to have increased out of proportion to the increase seen by similar hospitals across the country.

“One of the possibilities – although it’s impossible to establish cause and effect, but it’s certainly very tempting to consider the possibility – is that the introduction of PACS and the ease of clinical access may result in doctors ordering more studies than they have previously,” Siegel says.

So the number of studies ordered per in-patient admission or per out-patient visit has increased out of proportion, not only to what it had been previously, but also to the changes seen in comparable hospitals across the entire country. If utilization of radiology services increases, that’s good news economically in a fee-for-service environment. However, if a facility operates in a largely managed care or capitated environment, increased utilization of radiology services essentially loses money.

“The economic implications of increased utilization, which may or may not occur in other PACS environments outside of ours is a significant one,” Siegel explains. “So if you’re thinking of making a transition to PACS, one of the things you have to think about that you might not have otherwise is a possible change in actual utilization of radiology services. You make it so easy to be able to order and access images that there is a potential to have an increase in the number of studies ordered. Even an increase as little as 5 or 10 percent has a major financial implication, either good or bad.”

New facility = easy justification
For the Kelsey-Seybold Clinic (Houston), moving to a new facility was the catalyst for instituting a PACS. The large multispecialty, out-patient clinic covers the greater Houston area and offers radiology services in 17 sites, performing 215,000 procedures annually. The new building forced the issue of whether to stay with film and spend $1.9 million per year, or go with a PACS. If they wanted to do same-day imaging, PACS was the solution, saving the clinic $665,000 in the first year. (See chart on page 49 for subsequent year savings based on a 5 percent growth rate and 3 percent inflation rate.)

The clinic would have had to spend $900,000 in year one for processors and printers just to support film. It was in the process of developing a new capital budget, looking at the cost of new vs. replacement devices. “It wasn’t as though we were staying in the same place and our equipment was useable,” Fisherman says. “It was not salvageable, and we were going to have to throw some things away.”

The clinic’s quantifiable expenses for film-based technology included film, chemicals, supplies, service, transcription, light boxes, alternators, etc.; depreciation on new equipment; and full-time employees (FTEs) associated with film management.

Like the Baltimore VA, Kelsey-Seybold took a conservative approach in its analysis. It did not include potential savings that were not quantified, such as courier service for film transportation, film room space that was going to be used by the clinic for other purposes anyway, changes in physician or staff productivity or savings it couldn’t quantify on new modalities purchased for the new building.

“There was some cost that we didn’t consider that perhaps should be considered, including the cost of digitizing old films and the cost of printing and sending films to outside hospitals and physicians,” Fisherman says. “We did not figure into the equation training time on the equipment for radiologists and other physicians because we didn’t think we had hard figures.”

The analysis, which should include only costs that can be quantified, should be developed in collaboration with information systems, the finance department and clinical staff. Kelsey-Seybold’s communication system helped keep cost down with its existent T1 line that connected systems among other clinics. Other factors included the volume to make the project profitable, whether to buy or lease equipment based on cash-flow estimates and securing the support of key physicians.

Kelsey-Seybold’s PACS team planned, installed and implemented its PACS in five months. And thus, it has reduced its turnaround time on studies from 100 hours to 24 hours or less.

The pay-per-procedure approach
For St. Francis Health System (Pittsburgh), the lack of capital to purchase its own PACS would not preclude the system from going filmless in two of their
hospitals. David Lackner, M.D., chairman of radiology for St. Francis Medical Center (Pittsburgh), the system’s tertiary care center, and St. Francis-Cranberry, a 35-bed hospital located 25 miles north of Pittsburgh, began the PACS planning process in May 1994.

At St. Francis Medical Center, the busy intensive care unit (ICU) had a long tradition of copying films for every ICU portable chest X-ray, with radiology retaining the original for interpretation and storage. The process gave Lackner some ammunition for his cost justification for PACS.

“We repeated that [copying] process 13,000 to 14,000 times a year, and right off the bat, knowing that and the cost of the copying, chemistries and labor involved, gave me a little head start on doing a cost justification,” Lackner explains.

At the time the PACS plan began, the health system’s financial status was good, but then capital budgets started shrinking at St. Francis and elsewhere, making it more of a struggle to find the dollars to replace equipment. “It became increasingly apparent as time went on that unless we did a successful cost justification ? with something that was close to operating budget neutral, this just wasn’t going to happen,” Lackner says.

The hospital system was dealing with several large projects, including new operating rooms at the medical center, ICUs and new cath labs, and the expansion of an out-patient clinic into a hospital – all of which significantly increased the bond debt. Lackner learned the bond debt could go no higher.

“So we looked for a way to do some sort of off-the-books lease,” Lackner adds. “If you meet certain criteria with your lease arrangement, it can be carried as an ‘off-the-books’ lease, i.e., you show your bond holder no new capital debt. So that was the premise we got to about this time last year.”

When St. Francis broached the subject, the PACS team knew they would have to come up with a creative financing plan. Inspired by a shared-risk plan in the hospital’s lab, Lackner and his colleagues talked to vendors. “They said in their experience, shared-risk agreements were a risk for the vendor and sharing upside or profit potential for the hospital,” Lackner explains. “They kind of viewed them as one-way streets, but we kept persisting to come up with a way to do a shared-risk agreement where the vendor has downside protection and the hospital has upside potential.

Eastman Kodak Co.’s Cemax-Icon unit (Fremont, Calif.) and St. Francis put their collective heads together and came up with a creative financing plan that satisfied both sides of the table. After the introduction of the idea and several months of refining the arrangement and establishing appropriate risk corridors, St. Francis had its PACS plan.

But figuring out a fair pay-per-procedure fee would prove to be somewhat elusive and arrived only after a detailed engineering study in collaboration with Cemax-Icon, determining the health system’s exact costs per procedure. “It’s not so easy for a radiology department to figure out how much it costs to produce your product or what your overhead is, where all the charge-backs are, whether they are accurate and fair, how to pay for the lights, the floor space, the film storage, etc.,” Lackner says.

Working with its vendor, St. Francis came up with a detailed economic analysis, including how long it took the file room to find films, factored in wages and benefits, and came up with a number and then basically cut all the numbers in half to make it a conservative estimate. “We looked at the film dollars, the chemistry dollars and came up with definite consumables per case, such as supplies, film jackets, inserts, etc.,” Lackner says. “That part is the easy part of the equation. We [also] came up with the human component – the human productivity or salary component. We figured, depending on how you count, it costs us between $15 and $18 a film study to produce what we do.”

Lackner says those figures can be argued up or down significantly in either direction, but they needed something, some basis to work from that was specific to the radiology department and was reasonable, if not exactly correct. The process took approximately two years to complete. “None of this cost per procedure would have been valid had we not had something to compare it to,” Lackner says. “We wanted to have that baseline and we do.”

St. Francis guarantees Cemax-Icon a minimum payment per month based on a relatively high percentage of its current volume. Taking 85,000 as the current volume and establishing that as the 100 percent baseline, the health system is guaranteeing Cemax-Icon that the system will continue to do that much or nearly that much volume. It is a high percentage of that number, so even if St. Francis falls below that floor, Cemax-Icon will still be paid as if the number were at the floor. And once the system goes above 100 percent of current volume, the health system’s cost per procedure goes down.

“Then when we go above another level, we pay nothing additional, and that’s the upside benefit or potential benefit for St. Francis,” Lackner explains. “So the vendor has downside protection, and the health system has upside potential.”