Feature Story

Nuts and Jolts of Constructing Core Labs

April 2001
Anne Paxton

When you talk to people who have formed core laboratories, you might start thinking that renovating a 100,000-square-foot facility, taking testing from four different sites and performing it on one instrument, and transporting specimens all over the state are relatively trivial tasks. That’s because few core laboratory directors cite these as their challenges. What is hard, they say, is overcoming staff resistance, keeping the hospital administrators supportive, and getting different computer systems to talk to one another.

Whether a core laboratory brings together entities that are miles apart, across the street, or separated by partitions, the material barriers, it seems, are often easier to surmount than the invisible ones.

University Health Partners at the University of Oklahoma Health Sciences Center, for example, suffered major staff adjustment problems when it formed one core laboratory for two hospitals. "There were imaginary boundaries we had to break down," says K. Michael Parker, PhD, director of the University Health Partners core laboratory in Oklahoma City.

At UHP, coagulation, hematology, chemistry, immunoassay, and toxicology/therapeutic drug monitoring were combined under a single supervisor, and the staff was reduced from 56 to 33. "Our biggest challenge was cross-training people and getting them to think beyond their section." Three years after the process began, "We are just now getting people to say, ’I work in the core laboratory.’"

To cross-train staff, they started with scheduled time, then went to an intermediate approach of pairing people from different sections as "training buddies" to train each other as they had time. Eventually, they moved to computer-assisted training, which had the advantage of requiring only one person’s time. But the whole process took longer than planned. From time to time, depending on the daily workload, they had to suspend cross-training, and it was always difficult to restart.

"Cross-training is easy when there are plenty of people but harder as you reduce staff," Dr. Parker points out, adding that attrition was easy to count on. "You up the attrition rate when you start, because some people really don’t want to do it. They left and found other jobs. When somebody working in chemistry for years is told to start recognizing things under a microscope, it often exceeds their comfort level." On the other hand, he says, other people admitted they were bored with what they had been doing and welcomed the new challenge.

Most staff trained formally as generalists but had become specialists with years of experience. For the core lab they had to be reschooled as generalists, but this process can be taken too far, he warns. "One of the challenges of a core laboratory is not to make everybody a generalist. You still need a resident expert in each area. Our laboratory probably overestimated how extensively it could cross-train and, in the end, left an expert for each discipline."

It was the hospital administration, which compared staffing levels to industry benchmarks, that prompted the formation of the core lab. "When they had our laboratory beside some reference points, they thought we were overstaffed," Dr. Parker recalls. "The pressure was to reduce lab staff. They didn’t specify a core laboratory as the solution. But we said if you’ll support the renovation cost, we’ll compensate by reducing total FTEs. It was important very early to meet those targets, because it built our credibility with the administration."

While restructuring proceeds, he says, it’s also necessary to keep asking how processes can become more efficient. "You have to put reorganization and reengineering all together. Otherwise you’re using fewer people to do the same inefficient things." His laboratory found that reengineering workflow made it possible for fewer staff to handle more work. "Perhaps our greatest challenge was to form the core lab while continuing to provide all services," he says. "It’s been like changing tires on a car with the car still moving."

At Loyola University Medical Center, the core laboratory was designed three years ago to provide 80percent of the institution’s laboratory testing in one contiguous area 24 hours a day, seven days a week. George Krempel, associate vice president of health care at Loyola, said that at the time, the hospital was experiencing growing volume and shrinking staff. "As the medical technologist schools shut down, we were concerned about the turnover rate and our ability to fill positions with the organization."

With the typical section-dependent work areas dating from the hospital’s construction in 1969, the old laboratory structure employed a lot of duplicative staff, often with technologists busy in chemistry while those in hematology were idled. Meanwhile, the testing volume of the hospital, the 13 ambulatory care centers, and some 50 outside clients was growing by eight to 10 percent annually.

Krempel’s chief regret about the conversion was the speed of the transition and its effect on the roughly 200 FTEs who staff Loyola’s laboratories and blood bank. Between 1997 and 1999, while the core laboratory’s billables rose 18percent, paid staff hours dropped by 12 percent."We did it too fast," he admits. "We were way too aggressive, in hindsight, and it stressed the staff out."

Notwithstanding that mistake, he believes the staff actually likes the change. "When you’re segmented into chemistry and hematology, you’re limited in your ability to cover all weekends and still meet staff schedules. Once you expand the pool by cross-training and putting everything in one area, you can have a more flexible schedule, and more liberty with days off."

Carved out of the general chemistry laboratory in the lower level of the main hospital, the core laboratory has limited space, which has barred it from adding microbiology, even though Krempel would like to do so. The remaining 20 percent of the testing is done by the specials laboratory across the hallway, staffed with medical technologists who perform special chemistry, special coagulation, immunology, and flow cytometry 10hours a day.

Relocating to a larger space remains an option, Krempel says, but unlike many other core laboratories, Loyola’s is not supplemented by rapid-response laboratories in the hospital. The ER, half a block away, uses pneumatic tubes to transport blood for all its stat testing. "We’ve thought about moving more than once. But being an academic tertiary-care level 1 trauma center, we feel we need to keep the laboratory as central to the institution as possible."

In fact, his laboratory has been able to prove that the reduced turnaround time it provides has ended up shaving about eight minutes off the length of stay for patients waiting in the ER for laboratory work. "The ER docs love it," he says.

Demonstrating that kind of value is not that common in the laboratory world. "The laboratory is always asked to have faster TAT, a lower budget, and to perform new testing. But I think we add a lot of value to other departments." Efficient testing can mean shorter lengths of stay and greater patient satisfaction, and "it’s nice to have the ER recognize the laboratory for that," Krempel notes.

Rather than expand physically, the core laboratory plans to rely increasingly on automation. A preanalytical processor from Beckman Coulter gave it dollar savings in direct labor and in the ability to use more medical laboratory technicians than medical technologists to run the laboratory. Now the laboratory is installing the next version, the Power Processor. "We do about two million billables a year," Krempel notes. "With the growth we have and not being able to move, we feel we’ll be able to continue using the preanalytical line as a backbone to increase throughput and reduce turnaround time."

The core laboratory of Sisters of St.Francis Health System (SSFHS), which serves four hospitals on six campuses in Indiana and Illinois, began as a small pathology laboratory in a resort town on the southeastern shore of Lake Michigan. When that Michigan City laboratory was sold to SSFHS, the company turned to the consulting firm Coopers Lybrand (now PricewaterhouseCoopers), which advised moving the laboratory to a more central location and turning it into a core laboratory.

"The growth potential was pretty muted where the laboratory was. But the consultants looked at it and said, ’You have a jewel in the rough,’" recalls James C. Sparks, PhD. He was the person Sisters hired in 1998 to accomplish the conversion, and he oversaw the purchase and renovation of a former warehouse club in Hammond, Ind., as the new site for the core laboratory. Dr. Sparks is now president and CEO of the resulting SSFHS division, Alverno Clinical Laboratories, which just finished its second year. It handles 60 percent of the four hospitals’ testing volume—about 1.6 million tests annually.

"Everything that’s not time sensitive we’ve consolidated in the core lab," he says, noting that by getting rid of six to eight platforms for routine chemistry in the hospitals and turning to one fully automated, modular system in the core laboratory, "we can get big economies of scale," even in the first round of consolidation.

The laboratory saw a 300 percent growth in its outreach volume during fiscal year 2000. After hiring a marketing manager in November 2000, Alverno added a large number of new accounts and saw a 45 percent increase in testing in one month alone. "We were able to handle that growth spurt with no added technical FTEs. But in the preanalytical area, which is still pretty manual, we basically had all hands on deck for two months."

In Alverno’s experience, moving microbiology was the easiest because the tests are not time sensitive. But it was also the most difficult because of infection control issues at each hospital, Dr. Sparks says. "We’ve worked really hard with different pharmacy committees at the hospitals to get to a common antibiotic panel. We now have four different infection control committees within the six campuses, and we need to make sure what we do in the microbiology department supports all of these infection control activities."

Although the computer systems still haven’t been fully standardized, Alverno operates all the individual hospital laboratories and the core laboratory. "We did that for two reasons," Dr. Sparks says. First, staff may be transferred easily to different locations when needed. Second, "It gives us enough of a critical mass to qualify for different cost-saving insurance programs the state of Indiana allows for employers that meet specific minimum employee number criteria."

For UMASS Memorial Medical Center, created by a merger of two hospitals in Worcester, Mass., standardization in the core laboratory was a major problem when it was launched about a year ago. "That’s been the most difficult," says Guy M. Vallaro, PhD, vice chairman of hospital laboratories, operations, who cites cultural differences between the institutions. "One is an academic center, the other a community-based teaching hospital," he says.

Different procedures were performed, and different instruments were used. For example, endocrine tests were shifted to instruments with different reference ranges. "Some physicians are comfortable with the way things used to be done. They don’t like to get calls regarding panic values in the middle of the night because they feel it’s unnecessary, yet it’s been done this way at the university site for years," Dr. Vallaro says.

Occupying 13,000 square feet of newly renovated space, the core laboratory was designed to perform all testing from the university site and tests from Memorial when results can exceed a two-hour turnaround time. For example, Dr. Vallaro says, lithium was originally going to be done at the Memorial Hospital rapid-response lab, but it was shifted to the core laboratory. "It was going to require a separate instrument, and we made the decision that, due to the volume they received, they would be best served by doing all the lithiums here in toxicology." For occasional stats from Memorial, the laboratory uses taxis, which it finds more reasonable than calling a courier.

The laboratory, which processes about six million tests each year, likes to use the medical technologists it has to conduct esoteric tests to build its menu and add more outreach testing. Chronic understaffing has meant that taking on some potential outreach clients has been delayed, Dr. Vallaro says, but as new staff are added and a new Beckman Power Processor is installed that will automate front-end procedures starting this summer, he hopes the laboratory will be able to meet demand for its services.

UMASS Memorial has become known for its effort to make the transition to Web-based test ordering, which is already saving the laboratory time and money but has suffered enough snags that the project has become a long-range one. "We have 35 clients that order in through a software called Metricom that is Web-based, with 15 on the Web, and we are growing that area, but that’s still not the major way we get orders," Dr. Vallaro says.

Metricom’s Laboratory Requisitioning System gives the test registration an electronic head start, eliminating the need to write a specimen label and test requisition and include the patient demographics in each order. However, because of security issues, the orders are not sent in real time and do not cross instantaneously into the core laboratory LIS, so after coming in labeled with a Metricom bar code, they must be replaced with a Meditech bar code.

Web-based ordering is around the corner at Fletcher Allen Health Care, the academic medical center for the University of Vermont. Fletcher Allen has become a kind of core laboratory/regional reference laboratory for all the other hospitals in the state through a messenger model network set up by Mayo Medical Laboratories, says its director of business development Michael Gagnon, whose background is in pathology informatics.

Fletcher Allen, which has been recognized by the Council of Teaching Hospitals and Health Systems as having the lowest adjusted cost per discharged patient of any of the 130 U.S. academic medical centers, serves as the reference laboratory, and Mayo performs highly esoteric tests. MayoNet electronically links all 14 hospitals in the state and allows six of them to electronically order tests from Fletcher Allen or Mayo and have test results sent directly to their LIS.

Fletcher Allen has developed in the last 2 1/2 years a product for Web-based ordering called the Pathology Laboratory Information Network, or PathLine. Honored last fall at the University of Pittsburgh Medical Center’s Advancing Pathology Informatics, Imaging, and the Internet conference, PathLine was judged the best Web site in the category titled "advancing the field of pathology through computer technologies." "There are other commercial products similar to it, but ours is designed to be tightly integrated with our other laboratory, clinical, and administrative systems," Gagnon says.

When a patient is admitted in any way, even for outpatient laboratory work, that information is immediately downloaded into the system and updated in real time. Fletcher Allen owns about 200 physician practices and serves another 250 in its primary service area, and "they really want Web-based results reporting. We’re moving aggressively on that and will have it in the next three to six months. Clearly for doctors referring a tertiary-care patient to Fletcher Allen, they want to see the results from that patient after discharge."

Once the PathLine system is fully functional, he says, preanalytical errors such as incorrect diagnosis or billing information should drop dramatically. "We expect enormous savings in labor costs in the central receiving area," Gagnon says, predicting that the 25 FTEs who now staff receiving will be pared by half over the next two years.

The next big push will be to strengthen the electronic links between hospitals and create a statewide laboratory data repository that can aid disease management. Fletcher Allen has started to create a localized diabetes registry with the intention of making it statewide. A pilot program conducted in 1998 and 1999 demonstrated the feasibility of collecting hemoglobin A1C values on virtually all diabetics in the state and doing educational interventions with primary care physicians, internists, and endocrinologists. "We compared A1C results before and after those interventions, following a cohort of patients," Gagnon says. "We hope to expand the diabetes registry statewide with grant funding we’ve applied for."

While computerized databases become increasingly sophisticated, many core laboratories have found that when it comes to transporting specimens, the oldest technology still works the best. Loyola’s pneumatic tube system extends throughout the hospital, and with about 50 dropoff points, it can bring in specimens, including blood, blood gases, and tissues for frozen sections, from up to four blocks away, Krempel says.

In his view, the key to a successful pneumatic tube system is reliability. First, the plant operations must be committed to maintaining the resources to keep it running. Krempel compares the tube system to a computer network, and says both must be considered clinical tools. "You have to have enough backups to treat it as a clinical connection, not just a paper connection," Krempel says.

"Historically, hospitals put in a computer network only for e-mail, so if it went down, they just wouldn’t get e-mail for 10 hours." Networks and tube systems changed in the same way, he notes. Now institutions have to make sure the human and technical resources are dedicated to keep them operational all the time. At Loyola, the tube system has redundancy in blowers, and if a spill occurs at any time of the day or night, "somebody cleans it out and gets it running within an hour."

Blood gases, which have to be iced, pose special problems, and their transmission requires extra steps. At Loyola, a nurse usually calls from the ICU to notify the laboratory that one is being sent. "If we don’t get it in five minutes, we call back and see where it is."

University Health Partners in Oklahoma relies on an even older—but reliable—technology: a dumbwaiter. "We do allow blood gases to travel in the tube, but the majority come from the adult ER where we have dumbwaiter access because we’re right below the ER," says Dr. Parker. "We can get specimens faster that way."

Ask Dr. Vallaro at UMASS what the main unexpected problem has been in his core laboratory and he names blood gases, without hesitation. Since the core laboratory is located across the street from the university, the tube system runs under the pavement. "When you put a blood gas in and ship it across the street, the PO2 rises dramatically in some cases, probably because of agitation and the length it travels," he says.

While the core laboratory is working on a solution with the tube manufacturer, Translogic, at this point it is still running blood gases at the old facility. With only a few hundred blood gases a day, the staff have a lot of time on their hands, and his office has considered doing more testing there, such as basic electrolytes and CBCs for the emergency department. But the cost would be sizable, and Dr. Vallaro hopes that as there is improved "uptime" with the tube system, such a move can be avoided.

He considers the tube system the laboratory’s lifeline. "When it works, it’s great." But when an ER test gets stuck and sits in the tubes for hours, as happened recently, then there is pressure from some quarters to add a rapid-response laboratory at the university. "One thing we didn’t realize up front was the amount of effort that would be required to maintain the system," he says.

The carrier tubes, for instance, have two pieces of carpet around them that slowly wear away. "If they’re not replaced appropriately, they’ll get stuck in the tube system, which has happened." The laboratory recently hired engineers to maintain the tube system.

By contrast, the LIS has been 100 percent reliable, he says. "There were two issues when we moved here that we thought would either make or break us. The first was the tube system, and the second was the information system. We need it to work flawlessly, and it has."

The Sisters of St. Francis Health System’s opposite experience confirms that IS is central. "The biggest challenge we did not expect was to be able to provide rapid connectivity," says Dr. Sparks. "Through a variety of misunderstandings and miscommunications, we had expected the IS group that is part of SSFHS to be able to provide connectivity between two disparate LISs. The original plan was that connectivity would be established in May 1999. It actually occurred in September 2000. That was a significant and expensive delay."

While much of the lagtime can be blamed on Y2K concerns, which overwhelmed most IS staff and vendors, his caveat for anyone thinking of forming a core laboratory is this: "Make sure you have connectivity among all your systems before you put a brick in place in a new building."

University Health Partners’ relatively new, nearly $1 million pneumatic tube system connects University and Children’s hospitals and a third hospital to be added next year. But the tubes work better within hospitals—where the system has multiple "lanes"—than between them, Dr. Parker says.

The single tube connecting Children’s and University hospitals is like a country lane, where people have to pull over if they meet an oncoming car. "You can only send one bullet in either direction, so a lot of times those bullets wait in a queue for their opportunity to travel through the tube." The blood bank, for example, doesn’t send blood units from the adult hospital to Children’s simply because a wait in a queue for 20minutes or more can compromise their quality.

A new LIS was installed at the same time as the new tube system and shortly after the core lab was formed, because of communication problems with the hospital information system, Dr. Parker says. With the LIS and HIS now the same system, "the next stage for our core laboratory is to begin looking at service lines, making sure that there is a good flow and that the tests a particular clinic or service needs are accessible in some sort of group way for them."

"I think we’re very good analytically in the laboratory, but if a physician cannot easily order a test and get the result in a timely fashion, he really doubts that the quality of the lab is good," he notes. For example, if a neurosurgeon draws spinal fluid and wants both chemistry and cytology tests run, "in many instances it’s difficult because cytology is in a separate location, two floors up at UHP. We’re trying to make sure we have the right processes in place so we don’t hold the specimen up for glucose testing, delaying the cytology exam."

In fact, keeping clinicians happy with a new core laboratory is a constant concern. At UHP, the move to a core lab was fairly transparent to clinicians. In general, they don’t care much where the work is done as long as it preserves or improves turnaround time, Dr. Parker says. "One exception is within Children’s; the pediatricians want to feel like that’s a full operation and not just a skeleton, so we left microbiology and virology there." His department is now making a separate "infectious disease core laboratory" at Children’s by combining microbiology and virology, he adds, noting that it makes sense because, based on volumes, "much of our infectious disease testing originates in pediatrics."

Dr. Parker’s advice for those contemplating a move to a core laboratory: Always have the support of hospital administration, be good at communicating with staff why the move is good, and be persistent in pursuit of the goal. "Be prepared to deal with the naysayers and inevitable problems with cross-training," he adds. And "be prepared for the transition to go slower than you can imagine. There are all kinds of challenges to slow you down."

But the rewards outweigh the troubles in the end. "We could not have survived the reduction in staff without moving to the core lab," Dr. Parker says. Soon a third hospital laboratory will be incorporated into his core lab. "Reorganization and reengineering are ongoing processes. I don’t think we’ll ever be completely finished," he says. "There are always more changes that need to be made."

Anne Paxton is a freelance writer in Seattle.