College of American Pathologists
Printable Version

  Speaking volumes about
  IDN successes


cap today

July 2004
Cover Story

Anne Paxton

Rolling up clinics, hospitals, laboratories, and other facilities into integrated delivery networks seems like a sure ticket to slashing redundancy and streamlining operations. And the payoff can be especially enticing in big-ticket purchases like laboratory instrumentation. But integrated delivery network labs may have plenty of pitfalls to dodge before integrated instrumentation starts brightening the bottom line.

Park Nicollet Health Services might be a model for laboratory instrumentation re-engineering in IDNs. A vertically integrated system, it performs about 3.5 million tests a year in St. Louis Park, Minn., a suburb of Minneapolis.

In 1990, when its main hospital, Methodist, integrated with 25 neighborhood clinics that already had their own core laboratory, "there weren’t many instrumentation changes," says Dwight Grotte, MD, Park Nicollet’s medical director of laboratories. "It was merely a matter of closing down their core laboratory and sending the work over here. We just had to expand what we had here at the hospital."

Since then, Park Nicollet has standardized all of its analyzers and installed a Lab-InterLink front-end automation system to speed throughput and ease the workload. "Everyone in the IDN has the same analyzers, and it helps a good deal because the manufacturers price products based on testing volume, so if we can demonstrate a large volume they can adjust the price."

That’s how it’s supposed to work. But IDNs aren’t automatically an ideal structure for centralizing laboratory services and standardizing instruments, warns Stan Schofield, MHA, MT(ASCP), president of NorDx, a laboratory network for the Maine Health System in Scarborough. "Lots of IDNs have come together, but whether there are regional laboratories to support them is a different issue."

For example, Sutter Health System in northern California is "a beautiful IDN," he says, "but they don’t have a core or regional laboratory structure because it’s too diverse and geographically separated and the service demands are too great."

He points to North Shore Long Island Jewish Health System in Lake Success, NY, on the other hand, as an IDN with a regional laboratory serving all areas successfully and competing well against the large commercial laboratories.

Laboratory testing for nine of the 18 hospitals of North Shore Long Island Jewish—the third largest non-secular health system in the country—has been integrated since 1998 into a core facility. With revenue of more than $40 million per year, the core laboratory can take advantage of total laboratory automation robotics systems as well as instrument standardization.

Nevertheless, across-the-board standardization is a work in progress, says Robert Stallone, North Shore’s vice president of laboratories. For nine of North Shore’s hospitals, the core lab is doing 65 percent of their work. "The reason not all hospitals are on board yet is in New York there are issues relating to contracts and unions, as well as traffic and distance from the core facility," he says.

North Shore’s system has a test volume of about 14 million a year, including anatomic pathology and blood banking procedures. However, not all of that has been integrated. But it is North Shore’s integration and standardization of laboratory services thus far that have made it possible to capture significant business from commercial laboratories. "It takes a solid three years and a large financial investment to develop an infrastructure competitive with commercial laboratories," Stallone says. "But now we’ve built it."

He credits North Shore’s willingness to spend more than 10 percent of the core laboratory’s revenue on an information technology infrastructure. "That’s the type of investment typically hospitals will not make, and why commercial laboratories have been successful and hospital laboratories have not," Stallone says.

He and his colleagues defined a limited number of tests, about 50, that needed to remain in the hospitals for immediate care. "The goal of our hospital labs is to have minimal turnaround time, not to reduce costs or make money," he says. So North Shore established four workstations—chemistry, hematology, coagulation, and basic immunochemistry—and used that model at every hospital with the same equipment. "That made it a minimal-staffing model to achieve a rapid response," Stallone says. Even lower staffing levels are possible at the core laboratory because of workstation consolidation through total laboratory automation, he says.

"Since we do standardize instrumentation, we’ve been very successful in negotiation, and many times we actually do not go through the GPO but do purchasing ourselves, saying ’I need 36 of these instruments.’ With that type of volume and buying power, weable to negotiate."

He prefers to negotiate a guaranteed cost per billable. "In other words," he explains, "I’m going to run 1,000 tests this month. I’ll pay you X dollars per billable test, and that excludes all variables associated with the cost of service, cost of improper utilization or leakage, problems with the instrument, consumables that weren’t planned for—all of that is built in. It’s guaranteed by the vendor."

If your laboratory network is large, he notes, "you have more leverage to do this. It’s something many laboratories like because it shares the risk with the vendor—but many vendors may not want to do it unless you’re large enough."

Sentara Health System, which performs about 6 million laboratory tests per year in the Hampton Roads region of Virginia, meets that standard. For the last six years, with its 70 sites of care including six acute-care hospitals and 25 primary care practices, Sentara has ranked among the top 10 most integrated health care systems in the country.

A re-engineering initiative at Sentara in 1996 charged the central Laboratory Services with showing that its costs were competitive. "We essentially had to bid for our own business," says Diana Cline, MBA, MT (ASCP), clinical specialist for chemistry. "Our costs were very good, but we still had to show systems savings," and standardization was one way to help achieve those savings.

When they started, they had a combination of Beckman Coulter, Roche Diagnostics, and Johnson and Johnson general chemistry equipment, and different immunoassay equipment across the system. Cline’s task was to standardize chemistry. "In 1997 we standardized with Bayer for im mu no assay and Beckman for the community hospitals—with the exception of the reference laboratory," she says. There, they stayed with the Hitachi 747 to meet their throughput needs.

In 1999 a chemistry team composed of senior technologists, a clinical pathologist, and a clinical specialist focused on finding one vendor for general chemistry. That was about the time that Stephanie Spingarn, MD, director of clinical pathology, arrived at Sentara.

"The process for that was very interesting," Dr. Spingarn recalls. "Diana was selecting equipment for the entire system. The selection team had a checklist of questions to quiz all the vendors on test menus, methods, machine operations, disposables, etc. There were lots of very technical questions, and it was a lot of hard work to narrow down what we thought one vendor could do for the entire system."

At the time Sentara’s Laboratory Services had to replace 13 general chemistry analyzers. Norfolk General does a high volume of testing because it houses the reference lab, and the volume at the sister sites—each of which has a customized rapid-response menu—varies from low to moderate. "We had to pick a vendor that could handle both small and large volumes," Dr. Spingarn says. As medical director, her job was to make sure the methods chosen were optimal for their patient population.

To make the decision as objective as possible, the Sentara instrument selection team used a grading system for each vendor included in the evaluation. Each instrument system was graded on its quality—ease of use, throughput, technologist needs, and need for maintenance, for example—with cost taking a back seat. Says Cline, "We knew that if we defined quality up front, cost would follow because of our volume."

Once the grading was complete, the committee members were each asked to vote for the vendors they preferred. "The vote matched our scorecard," Cline says, "so that validated our grading system." The results of the grading gave the team objective criteria to supply the vendors that weren’t selected. "It was a big contract and people wanted to know why they weren’t chosen," she says.

Cline eventually selected 13 Roche Hitachi analyzers. "We brought two major volumes together—the reference laboratory volume, which had Roche equipment, and the volumes from the rest of the system using Beckman equipment." After bringing in the 13 new analyzers, they were able to reduce the costs per reportable to below their original costs. "The cost was lower because we looked at a system contract instead of looking at individual contracts," she says.

In addition to having saved money, Sentara now has a single reference range for the entire health system. "It is truly standardized," Dr. Spingarn says, "so if a patient gets a glucose at one facility, the next month at a second facility they’ll get comparable answers."

There are other benefits, too. "We developed one set of procedures that could be used across the system and staff members are able to work at any facility with limited training," Cline says. Reagents and supplies are shared from site to site, and they’re able to participate in a system CAP inspection instead of having multiple inspections per year. Quality control is also standardized. "The use of QC results is one way to maintain harmonization of all the instruments in the system," Cline says.

In tandem with the instrument standardization, Sentara Laboratory Services introduced front-end auto mation for general chemistry. Its front-end aliquoter/ archiver from Roche, the VS II, is now located at the reference laboratory. Initially there were problems, Cline says, as the laboratory brought automation to its processing department. Ultimately, "the aliquoter improved quality by decreasing mislabeled aliquoted specimens. The archiver has saved time finding specimens, which in the past meant looking through seven or eight racks of 100 specimens per rack. Now they just look at the computer and it takes them directly to the slot where the specimen is located," she says.

Centralization can bring all of an IDN’s testing to one site, but you have to balance that with the physicians’ service expectations, Dr. Spingarn stresses. "I think re-engineering here was very enlightened because they constructed a rapid-response menu, with the help of the medical staff, that is available at each hospital." Point-of-care instruments were standardized as well. A separate team handles POC testing for all of the hospitals, and the hospitals acquire their instruments as a group.

Before re-engineering, the hospitals had laboratory directors at every site, and each one made separate instrument selections. "Today," Dr. Spingarn says, "we have one clinical specialist across the system for each discipline who is charged with making instrument selections based on system needs."

In mid-size cities, networks or buying groups are becoming much more common. They make it possible for some of the smaller community hospital laboratories to enjoy contracts they might not otherwise have access to, says Susan Mammina, MT (ASCP) MM, laboratory administrative director of Spectrum Health Hospitals Laboratory, Grand Rapids, Mich.

The three laboratory sites of Spectrum Health Hospitals have grown significantly in the past five years and have standardized chemistry analyzers to Beckman’s LX20 and Access, Mammina says. But the much larger Spectrum Health Regional Hospital Network has gone further. It includes 10 independent community hospitals and the nine hospitals owned by Spectrum Health System.

"We’re working with this regional network to provide contracts for group volume discounts," she says. Spectrum Health System corporate vice president of strategy and business development, John Mosley, encouraged other hospitals in the community to join the network to facilitate networking, participate in group purchasing, help keep patient care in the community, and standardize the process of admitting or transferring patients to Spectrum Health facilities.

"Our volumes together support better pricing, and if we can share discounts with community hospitals, it helps their performance," Mammina says. Spectrum Health Hospital, a nonprofit organization, views the community hospitals as valued partners. "We work together with the regional network members as a reference lab and as a network for selecting equipment and signing contracts for blood products suppliers," she says.

The regional network has been supportive to many of the smaller hospitals’ laboratories. "When we selected Beckman as our prime vendor for chemistry, Beckman was very smart," Mammina says. "The vendor was willing to go to other regional hospitals and offer the same pricing it gave Spectrum Health to use their equipment, and it gives those hospitals an opportunity to get much better pricing than they would ordinarily." Through group purchasing firm VHA Novation, the SpectrumHealth Hospital Network is also exploring pricing with Dade Behring, BioM8Erieux, and Becton Dickinson for microbiology instruments.

The same kind of affiliation is used to structure Sacred Heart Medical Center and Pathology Associates Medical Laboratories in Spo kane, Wash. Both are integrated with a network of five hospitals in Seattle, three other Seattle testing centers, and another 12 hospitals and testing centers throughout the west, says Lawrence Killingsworth, PhD, DABCC, chief science and technical officer for the laboratory network.

The other hospitals are partners. "But we don’t own them," he says. That makes his job of standardizing and assuring comparable results from all testing sites somewhat tougher. "We can’t say, ’OK, hospital X, you have to buy Roche or Beckman analyzers.’ But we do say you have to get the same results as everyone else."

When a new partner comes in, standardizing its laboratory with the core laboratory is fairly easy with analytes like therapeutic drugs, but more difficult with other tests, such as some enzymes and coagulation tests. "It turns out that almost all our hematology testing sites just happen to have the same brand of analyzers. But we have chemistry analyzers from six different vendors and every possible combination of coagulation equipment and reagents you can imagine," Dr. Killingsworth says.

Operationally, they function as a single laboratory. "But we haven’t gone to vendors and said, ’Treat us all as one buying group,’ because we’re 20 different hospitals and testing sites, with multiple owners. We’ve had vendors who’ve said, ’We don’t care, we’ll treat you as a buying group.’ But this laboratory network isn’t particularly about buying power. It’s about consistently high-quality results throughout the system," he says.

Laboratories that focus on certain specialized areas of testing may not combine easily with others in an IDN, posing another potential obstacle to standardized instrumentation. Erlanger Health System for example, consists of four hospitals in the southeastern part of Tennessee. The core laboratory in Chattanooga runs about 1 million tests per year, says chief of pathology Edward P. Fody, MD, and does as much group contracting as it can when purchasing instruments.

"We get instruments for the smaller laboratories, too. You try to make it the same across the board; that’s our goal," he says. But the independent pediatric laboratory in Erlanger Medical Center is probably the least integrated. "It’s really a fifth laboratory, and their patients don’t overlap very much with the other hospital laboratories. So we’re least concerned about standardization there," Dr. Fody says.

For some, integrated delivery networks have proved to be the ideal vehicle for shrinking instrumentation costs. Schofield’s laboratory network, NorDx, is an example. Four entities had to be consolidated to form NorDx: the laboratories at Maine Medical Center and Brighton Medical Center, a small regional reference laboratory called Northern Diagnostics, and a research institute.

"We started off with probably 25 major analyzer platforms from 15 different companies," Schofield says. "And when we came together in 1997 we started immediately moving toward standardization. We needed flexibility for staff to use the same equipment, and we needed leverage in purchasing power with vendors because we were paying way too much money."

"Every organization had its personal bias and experience with certain equipment. But through a long process of bidding, we narrowed the field, selected Roche as the vendor for all sites, and installed the instruments in 1999 and 2000."

The result: seven years of 12 to 15 percent growth each year without added staff. "We reduced instrument count, reduced workstations, reduced our service costs, reduced our cost per test, and improved not only our service but also our turnaround time to clients, and we’ve made our staff happier with state-of-the art instrumentation."

The pressure of competition from the large commercial laboratories keeps the process in motion. "We have to compete against Quest and Lab Corp and they’re very efficient with much larger organizations, so we can’t run a sloppy or loose, non-productive operation. The workload and workflow are designed to free the instruments and minimize excess staff and time and motion that don’t result in productive effort by the staff."

The core laboratory was built without walls, and thus it’s been able to adapt quickly to changing needs. "There were sections and divisions and departments, but the walls are open modular design, and when we want to reconfigure the laboratory we literally can, with a pair of pliers, a wrench, and a screwdriver, reassemble the modular partitions. We’ve done it several times as we redevelop instrument platforms and reconfigure workflow."

Computerized processes such as autoverification—though implementation hasn’t been smooth-have helped ease the staff’s workload by allowing normal results to be released without a delay for manual re-checking, he says.

But Schofield is most enthused about the leverage gained from group contracting. "If I were just going out for chemistry instruments and my past chemistry purchases were $1.2 million, the best I could do would be to keep it there or drop it a little. But the pricing concessions get bigger if there’s more business.

"So suddenly, now that I’m going to buy hematology and immunoassay instruments too—at big savings as well—I might get chemistry down to $800,000 and end up saving just from this one company a total of $600,000 to $700,000 a year."

"We were one of the big organizations in the country that had standardized on one brand," he says, "and while it’s risky, as long as it’s a good brand and you can work with the vendor, then you have much greater purchasing power. They recognize that as a loyalty-type issue."

Can other laboratories match those results?

Says Sentara’s Dr. Spingarn: "Somebody reading this article might say, we don’t have those client volumes and we can’t achieve those savings. But other systems can do it depending on how many sites they have. Even in a small hospital, if you go in and present competition, and make vendors vie for the business, you will save."

Anne Paxton is a writer in Seattle.