From left, Dan Singer, MD, Roberta (Bobbi) J. Miller, RTR, CRA, and Gary Gordon, RN, The Toledo Hospital, Toledo, Ohio.

In the course of sketching a plan to acquire and deploy PACS, hospitals sometimes make the mistake of assuming that the introduction of this technology will automatically eliminate all workflow problems associated with film use. At The Toledo Hospital in Toledo, Ohio, such assumptions were not permitted. The 717-bed tertiary care facility, which acquired PACS in 2002, wisely girded itself for PACS installation by first taking pains to resolve the productivity, efficiency, and quality-control issues inherent to its analog-based operations, says Dan Singer, MD, FACR, medical director of radiology and diagnostic imaging. "Anything less would have been counterproductive," he says.

Gary Gordon, RN, senior vice president and chief operating officer, agrees. "We took the position that, if processes we had in place were not optimized in a manual environment, they certainly wouldn't be any more optimized by switching over to an electronic environment," he says.

Roberta (Bobbi) J. Miller, RTR, CRA, radiology administrative director, was hired 2 years ago to improve processes and prepare for PACS. She says, "I'm a firm believer that you shouldn't start implementing any kind of computerization unless and until you have an efficiently functioning film fileroom. Otherwise, you end up with a garbage in, garbage out scenario."

Miller knows what she is talking about: "Our film fileroom was the area where we had the greatest need for more pre-PACS efficiency. We felt that, once the fileroom was put in optimal condition, our conversion to PACS would go smoother, faster, and generally be more successful."

That prediction was proven right.

ROOM FOR IMPROVEMENT

The Toledo Hospital is one of the city's oldest such facilities, having opened in the late 1870s (although not until 1931 did it move to its current location). Within the past decade, Toledo Hospital joined 11 other hospitals in the region to form the ProMedica Health System, for which Toledo Hospital serves as the adult and pediatric tertiary referral center, thus attaining for itself a market reach extending into 23 Ohio counties plus several more in southeastern Michigan.

Thanks to the hard work of its 800-member medical staff and 4,500 employees, Toledo Hospital today claims a reputation for top-notch neurosurgery, vascular, cardiology, and cardiac surgery services. An on-campus children's hospital (featuring a 60-bed neonatal ICU) means Toledo Hospital also is well known for its pediatrics, obstetrics, and fetal medicine.

Flow chart representing processes prior to the implementation of PACS. PACS processes diverge at the file room depending on availability and form of priors.

Some 25,000 inpatient admissions are handled at the hospital in a typical year's span. For the radiology department in 2002, that translated to orders for 240,000 imaging procedures. This year, the department projects it will perform about 1,000 more such studies, a sign that business in general for both the hospital and the department is up slightly.

Much of this diagnostic work is conducted in a quartet of x-ray rooms, a trio of fluoroscopy berths, a pair of emergency department (ED) direct radiography (DR) rooms, three CT scanner areas, six ultrasound closets, seven nuclear medicine rooms, and two MRI suites. Additional work takes place in imaging sections at 11 outpatient centers, most of which are equipped with plain-film radiography and mammography machines (except for one that also has CT, fluoroscopy, ultrasound, and, soon, MRI).

Given the volume and variety of imaging studies generated at Toledo Hospital and the on- and off-campus distances film files must travel, it perhaps was inevitable that the enterprise would find PACS of great interest. When Miller arrived at Toledo Hospital in March 2001 to lay a proper foundation for the PACS that was soon coming, the first thing she did was assess the strengths of the manual film production, collection, reading, reporting, distribution, and archiving practices then in place.

"Much of what I found was good," she recalls. "The technical staff, in particularthey were talented, dedicated people."

There was, of course, room for improvement. And it was in the film fileroom that Miller would uncover the greatest opportunities to render improvements.

"Mainly, I discovered that workflow was not as efficient as it could have been," she says. "Too many steps required to accomplish tasks, for example."

Among Miller's goals was to increase the speed at which records could be retrieved in the fileroom so they would more expeditiously reach physicians requesting them. An envisioned benefit of this would be shortened length of stay, since having records available faster meant that physicians could intervene with patients earlierand earlier intervention might well result in patients getting well enough for discharge sooner, she says.

One minor sticking point in mapping out what to do about fileroom efficiency was the matter of how many months of film archive capacity should be maintained on campus at all times. Senior decision-makers projected a financial benefit by moving to off-site storage all films older than 2 months. This, they thought, would free up space on campus for income-producing uses. However,  Singer disagreed, saying that off-site storage of all images older than 2 months would generate significant expense.

"It would require us to make extensive use of couriers," he remembers cautioning. Singer's idea was to maintain an on-campus archive of films up to 12 months old. "Far fewer courier trips per day would be needed," he says. "That would mean a very substantial cost savings right there alone."

Singer also said that retaining on-site films up to 12 months old would result in faster turnaround of requests from referring physicians and radiologists seeking prior images. Eventually, top administrators and Singer agreed to preserve the space in question for fileroom purposes.

CHARTING THE FLOW

One of Miller's first moves in attempting to boost fileroom efficiency entailed a customer service survey of referring physician satisfaction. In so doing, she hoped to gain a better understanding of the strengths and shortcomings of the fileroom as perceived by those outside the department.

Miller based those reorganization plans to a significant degree on established workflow patterns. "I made an analysis of the steps each staffer took in order to create a file jacket, load the jacket with content, put it away, label the content, field requests for retrieval, pull requested files, deliver them, bring them back to the fileroom, and return them to their proper resting place," she says.

Her analysis included making flowcharts of those steps. From those charts, she was able to clearly see which steps could be eliminated, which could be pared, and which needed no modification. She also could identify places where entirely new steps could be inserted either to form shortcuts or to accommodate the performance of new or revamped tasks.

To support the new workflow she developed, Miller specified that extra shelving be installed to help maximize the amount of fileroom space available for storing jackets and other records.

"We revamped the physical positioning of furniture and fixtures so we would have more wall space for more shelving," Miller explains.

A number of small desks scattered alongside the walls and in corners were removed. In their place, a single large work surface was set up as an island in the center of the room. The island was custom-built by the hospital's plant-maintenance craftsmen.

"On one side of the island's countertop, we would later install information system workstations for staff to use to look up file locations and file availability," says Miller. "On the other side of the island would be staff whose job is to assemble jackets and make sure the contents are in correct order for hanging on an alternator."

Once the workflow redesign was fully mapped, the fileroom employees were given 3 weeks of intensive training so as to be able to perform their job correctly. Afterward, however, each staffer also received a 3-week training rotation in every other job assignment in the room.

"The purpose of this was to help the staffers understand their assigned duties in the larger context of everyone else's assigned duties," says Miller. "This showed them how even a minor mistake at their end could have an adverse ripple effect throughout the entire room. It helped them be more diligent in their work as well as encouraged them to function more cohesively as a team."

Importantly, Miller made it a point to keep the staff informed at every step of the way before and during the fileroom reorganization. Communicating with them took time and effort, but proved itself an investment that paid huge dividends.

"No one felt disaffected by the changes," she says. "I explained to them what I was finding in my analysis, what I thought needed to be changed, what needed to be kept as is. I also sought their input and feedback. All of this helped keep up morale. The last thing I wanted to see was morale sink during this time of change, which is something you have to carefully guard against whenever you take away the familiar and replace it with something new."

Morale in general was kept high by staging small, in-office celebrations each time a major milestone had been reached. For example, the installation of the countertop island and getting it operational was commemorated with a luau.

"I brought in a potted palm tree, we played island music, had tropical-themed food," she says. "It was no big deal, but it made a dramatic difference in the way the staff felt about the pains we were going through during this time of change."

THEN CAME PACS

A mere month after starting the fileroom reorganization, Miller began pulling together teams of personnel to serve as members of PACS selection and implementation committees. The PACS team included the IT operations manager, the IT networking technician, the PACS/clinical information manager, the RIS administrator, a radiologist, the radiology administrator from nearby Flowers Hospital, which shared the same PACS, and Miller. So that each member could do the best job possible in that capacity, Miller supplied them with a wealth of information about PACS technologies old, new, and still out on the horizon. She also provided background information explaining what PACS would mean to the hospital.

"I lobbied the hospital's information technology department to be centrally involved because I knew we could not do a project the size and magnitude of PACS without their help," Miller says. "It was absolutely crucial to have them with us, right from the very inception of the project."

After narrowing the field of prospective vendors to the seven best, the PACS selection team drafted and issued RFPs. When the replies came back, three vendors were eliminated from the list. PACS team members then visited installed sites recommended by each of the four vendors still in the running. A year-long examination and comparison of performance, costs, and architecture features brought the hospital to its ultimate winning bidder.

"Once our vendor was selected, implementation went quickly," says Miller.

Already installed at the hospital was a very large, tape-drive data archive to facilitate development of an electronic patient record system. This expensive archiveacquired about a year earlierwas only being utilized to 25% of its potential.

"By incorporating it in our PACS project, we saved money at the same time we increased the value of the hospital's previous investment," she says. "The tape-drive data archive required some upgrading in order to work with PACS. Even so, the cost of upgrading was minor compared to what the hospital would have spent on an all-new archive system."

The PACS team harnessed the underused archive by licensing an additional 3,000 cells and adding 5 tape drives dedicated to PACS. Two servers were added in front of the long-term archive for the PACS data.

Network infrastructure too was already in place, another savings. Reundant gigabit lines run to each of the three hospitals and two imaging centers served by the PACS, configured in a circle to provide backup in the event that one line gets cut. "We and our PACS vendor brought in a network validation company to validate the network to make sure it would be able to handle the PACS image files," Miller divulges. "It found that all we needed was some minor tweaking of the network, some switch changes, and the addition of a few redundancy points."

PACS went live first in the ED. That was just days before Christmas, 2002. Two months later, PACS was brought up throughout the main radiology department. Less than 30 days afterward, the entire hospital was online with PACS.

BIG WAIT LOSS

According to Gordon, the first phase of the PACS implementation is now complete.

"It went very smoothly and is now deployed throughout the hospital," he says. "We expect to start phase two in 2004. At that time, we'll begin to extend PACS throughout the system to off-site locations, including physician's offices and homes."

Originally, Miller planned a full phase-out of film and the fileroom by gradually converting all past historical records to an electronic format. She changed her mind after colleagues across the country who had tread that same path warned her about the drawbacks to that approach.

"It was economically infeasible, first of all," she says. "The manhours required to digitize the films would have been enormous and the number of digitizers needed to do that work would have amounted to a very costly investment. Plus, digitized films are not as flexible or as high quality as images that begin life as digital output. So, film is going to be with us for quite a while. It was a good thing, then, that we undertook the reorganization of the fileroom."

The plan for film now is to select just certain categories for digitizing, which will occur only after they have been read by the radiologists. The remainder will stay on the shelves for 10 years to satisfy HIPAA requirements and then be destroyed, Miller notes.

So far, everything is working close to how Miller, Singer, and Gordon envisioned. The one disappointment is that business is not more brisk.

"The reorganized fileroom plus PACS make it possible for us to handle appreciably more volume than ever, but the demand has not materialized," says Miller. "That's because the market here is so very competitive."

The upside is that the gains in efficiency have allowed the radiology department to decrease backlog and open more time slots on the patient schedule, which minimizes the chances that patients will choose to take their imaging business elsewhere. Miller notes that waiting time for MRI appointments when she first arrived at Toledo Hospital ran 21 to 30 days. Now it is closer to 5. CT studies entailed a wait of nearly 2 weeks; now patients enjoy next-day appointments.

"Even if we had not gone to PACS, a reorganization of our fileroom would still have been most advantageous for us," Miller contends. "Fortunately, fileroom reorganization is not rocket science. It is simply about building a good, strong foundation, paying attention to details, and teaching personnel to perform the tasks in a uniform way."

Rich Smith is a contributing writer for Decisions in Imaging Economics.