The kindest cuts of all: labor savings via standardization
For the Veterans Health Administration network in the Great Lakes area known as VISN 12, standardizing lab instruments is nothing new. In fact, laboratory managers there are old hands at it. They standardized chemistry analyzers in 1998, then moved on to coagulation, urinalysis, point-of-care glucose, blood culture, blood gases, and, finally, hematology.
But they still find themselves amazed at the dramatic labor savings they achieve simply by installing the same instruments in all facilities in the network.
One year after VISN 12 standardized hematology analyzers to the Abbott Cell-Dyn, the pattern has held for one of the lab’s most sought-after goals: reducing the number of manual differentials. "When the Cell-Dyn analyzers were implemented, we saw an immediate reduction in the amount of time spent at the microscope performing manual differentials, verifying platelet counts, and assessing RBC morphology," reports Susan Ayres, MT(ASCP), manager of the Madison, Wis., VHA laboratory part of the VISN 12 network.
For even greater savings, the VISN 12 hematology technical committee, under the direction of John L. Kennedy, MD, worked to refine blood smear review criteria. Using the criteria and the new instruments, the eight medical centers in the network collectively cut the number of manual differentials performed each month from 4,640 to 2,860 between September 2000 and last May. That’s an estimated labor savings of 149 hours per month.
The Veterans Health Administration was one of the leading beneficiaries of the "reinventing government" concept that re-oriented federal agencies to think of themselves as productive businesses. The VHA’s 170 medical centers were organized into 22 veterans integrated service networks, or VISNs, in 1995.
VISN 12, the Great Lakes Health Care System, consists of VHA facilities in Michigan’s Upper Peninsula, all of Wisconsin, and the greater Chicago area—eight of them medical centers, plus three outpatient clinics with on-site laboratories and 25 smaller outpatient clinics.
VISN 12’s Pathology and Laboratory Medicine Service became a "product line" in 1996. That was when it began to operate as a single, integrated administrative and business entity, with the goal of realizing savings through economies-of-scale purchases and consistent instrumentation. In 1998, VISN 12 was the first network to undergo the CAP’s new network
VISN 12 was also one of the first VHA networks to do a significant amount of standardizing. "We were early on in the curve," says Bruce E. Dunn, MD, chief pathologist for VISN 12. But the financial benefits swiftly became apparent. "The dollar savings have been very significant. In 1996 our fiscal year budget was $32 million. Last year [fiscal 2001] we spent $28.7 million. So when you take into account five years of inflation and cost-of-living increases, we’ve done a lot."
Ayres, who has been with VHA during the entire consolidation, says that when the VISNs were set up, arrangements for laboratory services were left up to each network. "Somewhat strangely for the government, the VISNs were not given any direct guidance on how they were supposed to do things. So some actually contracted out most of the laboratory to private companies like Quest and just kept stat labs in the hospitals. Our administrative directors came up with the plan to keep the lab within the VA but consolidate testing to the core labs. We thought we could do it better than a contractor."
She credits much of the success of the standardizations to technical committees whose effectiveness has evolved since the first team was formed for chemistry. "The first standardization was more of a forced standardization," in which the laboratories had little input. "It was a big time of major change, and we were being resistant back in the beginning," she explains—partly because the various laboratories weren’t "psychologically onboard" the network concept. Many protested at the idea of switching their chemistry analyzer.
"There wasn’t a whole lot of buy-in initially. It really wasn’t until we got the chemistry analyzer and started seeing it was less work, and there were helpful people at the other sites, that we started getting more enthusiastic." As other standardizations proceeded, and since the technical committees for each field were set up to include representatives from each hospital or clinic, all the facilities started to feel more ownership of the standardization process.
The result was a relatively smooth transition to standardized instruments in chemistry (Dade Dimension), coagulation (Dade/Sysmex CA Series), urinalysis (Bayer Clinitek/Atlas Series), glucose meters (Abbott), blood culture (Organon Teknika), and blood gas (Bayer). "With each one, it was less work. We would bring in the analyzers, do lots of sharing, face-to-face meetings, and monthly teleconferences. By the time we got to hematology, standardization was totally collaborative," Ayres says.
"Before the VISN 12 lab product line was established, there was no interaction between our VA labs, absolutely none. Now we collaborate on everything."
Thomas O’Donohue, DLM (ASCP), administrative director of VISN 12’s southern tier, which includes four hospitals in Chicago and several community-based outpatient clinics, says the hematology standardization ran into a few rough spots at first. "The two core labs had the same vendor’s analyzer, the other six major facilities had a mix, so there was no consensus on the best equipment. Each site had its own favorite, and there were a lot of allegiances built up to specific instruments."
Even though chemistry had been successfully standardized, Dr. Dunn recalls, "there was a push for a variety of reasons not to standardize hematology completely. Our two large sites were used to using the equipment they liked. But this was one time I said we’re not going to have different analyzers."
Says O’Donohue, "We developed criteria for a request for proposals for an analyzer, with the idea we would all standardize on one specific vendor with different models appropriate for the size and needs of different sites," which ranged from urban and highly affiliated to small and remote. After the technical committee reviewed bids and selected analyzers it felt met the technical criteria, the administrative group did a final evaluation. "We ultimately went with the most economical," O’Donohue says.
As it turned out, the Cell-Dyn also had technology that could better identify differentials that needed manual screening versus instrument—an important advantage. The difference between automation and technology is significant, O’Donohue explains: "Automation can move the specimens through the instrument faster, but technology can reduce labor because of the lower number of differentials that would be performed."
Benefits became apparent even before the equipment was installed. "Obviously, we achieved some economies by volume, by getting standardized instruments for all facilities and the lowest price per reportable," O’Donohue says. "By combining all volumes into a single acquisition, and homogenizing the cost per reportable for all sites, in every case the savings have been anywhere from four to 18 percent off what individual labs would have paid."
"If Iron Mountain, which is located in the remote Upper Peninsula of Michigan and has only 11 technologists, went out on its own to get a hematology analyzer, they would probably be paying $7 to $8 per test. And they could not even afford the technology gained. This way, the larger facilities are underwriting technologies at the smaller facilities." Three sites, in fact, were able to add backup hematology analyzers with the money they saved in the purchasing process.
But the most dramatic savings stemmed from the reduction in the number of manual differentials. This was due to the analyzer’s capability to flag with greater accuracy the differentials that needed manual screening, and to the standardization committee, which looked at each facility’s percentage of manual to automated differentials and established a "best practice," so that facilities could shoot for a target. "The committee was able to realize some facilities were doing far in excess of what we considered prudent, and they collectively set a new goal," O’Donohue says.
"We realized almost immediate savings in labor," O’Donohue reports, noting that the rates at almost all the facilities have dropped steadily to about 9.5 percent. VISN 12 has set a target of six percent. "The estimated savings then will be $140,000 a year in labor," he says.
"The labor savings let us juggle people’s time around differently," Ayres explains. "Just in my own experience, we had always had at least one and a half FTEs assigned to the differentials bench. When we started with Cell-Dyn in 1995, it dropped to less than half an FTE needed at the differentials bench during the day shift. So it was not that we got rid of an employee, but we found less and less time was needed to verify CBC results at the microscope, so that person could be working in chemistry or drawing blood or doing a number of other things."
"We used to be extremely conservative in our review criteria," she adds. "In the past, there were many more results that the laboratory would look at under a microscope to confirm. But with five years of experience with Cell-Dyn, the hematology technical review committee decided the accuracy of the instrument was good enough that many results don’t have to be checked anymore."
"They also made the decision that if a doctor orders just a platelet, and we’re getting a white blood cell flag, we don’t have to look and check out an abnormal white blood cell flag. That’s probably more in line with what the rest of the world is doing with Medicare compliance," Ayres says.
Not every facility in the network has made the same progress. There is one outlier that continues to do manual review of about 40 percent of the CBCs because of computer and personnel issues, Ayres says. "We could bring them down to about eight percent where the rest of us are. They’re not there yet, but they are working on a performance improvement plan."
Another problem for some facilities was having to de-automate. "Our two core labs had more automation than they needed," Ayres notes. "They had done their own hematology standardization between themselves, and the automation they had is what you’d see in a large reference laboratory, where a track system takes specimens from the analyzer to the slide maker/stainer."
"It was really nice automation, but for the volume of CBCs they were doing, it was overkill and costing them money." Reference laboratories doing a thousand CBCs a day need that type of automation, she says, "but our larger labs do 150 to 250 per day."
"We’ve seen it throughout the country and in our network," O’Donohue says. "Some sites acquire technology and automation beyond their needs. It used to be part of the laboratory culture to get the latest and greatest. But many networks are stepping back and looking at robotics and preanalytical processing and total laboratory automation, and wondering if the up-front cost provides them with value. Many times the savings aren’t there in the long run."
The ease of adopting standard instrumentation in hematology was reflected in the rapid timeline, Ayres adds. Although only her facility, William S. Middleton Memorial Veterans Affairs Hospital, already had the Cell-Dyn, and a long learning curve was predicted for the other seven facilities, "Within six weeks we actually went live." That compares to an average of eight to 10 weeks to set up and complete training.
"We did fewer correlations than what you would do if you brought an analyzer on all by yourself," she says. "We were able to share procedure manuals, reference range information, training documents, and correlation and QC data."
All of the standardizations, in fact, have been successful. "There’s no question about the financial success," O’Donohue says. "Any vendor who’s placing 10 to 20 pieces of equipment at once is obviously able to reduce their prices because of the volume of this joint venture." But other benefits and savings can be overlooked, including shared lot numbers, standard training checklists, vendor-sponsored user groups, and sharing of staff and reagents.
"One of the things we learned from the hematology solicitation was the technical experts, that is the technologists on the bench, many times have different views of the value of a piece of equipment. Administrators may tend to just look for the best value or the lowest price."
In this case, O’Donohue adds, "It just so happened that we got the cheapest price. But if we’d gone out and looked solely for the cheapest price, we might have ended up with a different analyzer."
VHA patients don’t know, of course, whether instruments are standardized. But standardized reagent and reference ranges benefit them by allowing veterans to travel from one site to another within the network without introducing variable interpretations of laboratory data—an important assurance for veterans because so many of them travel to multiple sites and are on warfarin, Ayres says.
System inspections have also become easier. "We can demonstrate procedure manuals, correlation studies, and so on that can be done at one site and then modified," O’Donohue says. "When inspectors come in, they can look at the procedure manuals at one of the main core laboratories and not at each site. It cuts down on the amount of time both inspectors and inspectees have to spend."
Similarly, VISN 12 has been able to standardize quality assurance, continuing education, and safety programs. Staffing is more flexible too. The three facilities within Chicago have been able to rotate technologists on occasion, although the physical distance between sites is a barrier to doing more of that, O’Donohue says.
One obstacle that standardization project leaders have had to surmount is the close relationship between VISN 12 laboratories and their medical school affiliates. "It’s different at every site," Ayres says. "Here at Madison my anatomic pathology is totally integrated with the University of Wisconsin, but in the clinical laboratory it’s not a big deal."
"Before we became a network, we tried to align ourselves with the university, so some of the relationships are entrenched," she adds. Says O’Donohue: "Many sites are so highly affiliated, they are physically adjacent or connected, they share residents and house staff, and many shared instrumentation and techs, and they have staff members who are also on staff at the university." Nevertheless, encouraged by the collaborative process of the technical committees, most of them agreed to do what was good for the whole network rather than individual facilities.
The medical school factor is likely to complicate standardization in microbiology. "It’s the only area we haven’t done," says Dr. Dunn, who notes that the medical schools could present additional challenges because they seem much more interested in microbiology than in hematology and chemistry. In addition, "they want a say in what’s going on," he says.
Once skeptical of standardization, Dr. Dunn has become one of its biggest supporters. "The more we’ve gone forward, I’ve recognized the advantages of standardizing." They now have input from 11 different sites (counting the three small outpatient labs). "The sharing of information has been a big plus," Dr. Dunn says. "We now only have to write a single laboratory manual, with minor modifications at some sites." Problem solving is also easier. "One site can ask the others if they’ve seen a certain problem, and if they have, then they may have a solution," he notes.
The normal pattern is for the big institution to give the benefit of its knowledge to the smaller one. But at Edward Hines Jr. Hospital, Hines, Ill., Dr. Dunn found he was going through a procedure that had been written by a supervisor at tiny Iron Mountain VA Medical Center. "It had actually migrated from a small site to a large one," he says. The lesson is clear: "We’ve learned we have to take into account the knowledge we really do have around the network."
"It’s actually very gratifying to see the rate of manual differentials has gone down at so many sites, because they’ve all shared information and learned from each other, rather than my dictating to them," Dr. Dunn says. He prefers "allowing people to see the benefits of change rather than forcing it on them."
As a future goal, VISN 12 hopes to convince clinicians to stop performing scans for all hematology/oncology patients. "At Madison, we currently scan the slide for abnormal cells for any patient in hematology," Ayres says. "I don’t think it’s standard at other institutions and we’re considering altering it. That’s another benefit of having a collaborative relationship with other sites, so we can go to the hematologists and say, ’There is consensus within the network that there is no clinical rationale for performing this analysis.’ This gives us leverage in our utilization management efforts."
Dr. Dunn cautions, however, that the sites will probably have to move toward this goal at their own pace. In an ambiguous situation, "you’d want to go on the side of doing the differential rather than not doing it. I would never want to dictate to people not to do the scans in that situation."
O’Donohue says the key to a successful standardization is getting the buy-in of people at the technical bench level. "Getting them involved reaps real benefits—some of which were not predictable initially," he says. The payback will make it worthwhile, he says, adding that what facilities can gain from standardization is long term. "The byproducts of consolidation don’t ever really stop," he says.
Anne Paxton is a writer in Seattle.