July 2010
Feature Story

Karen Titus

Patrick O’Sullivan, MS, MT(ASCP) SBB, could be a miracle worker.

O’Sullivan, who’s the director of the Florida Hospital Orlando Laboratory, tells the story of how Florida Hospital (part of the Adventist Health System) acquired a new hospital in 2000. The 300-bed facility had its own independent, fully functioning lab. O’Sullivan (who was LIS manager for the Florida Hospital system at the time) and his colleagues quickly found a way to save the lab $1 million a year.

“Immediately we closed the micro lab and took away all the esoteric testing,” says O’Sullivan. Thus stripped of these tests, the lab needed far less space than it once used. It also no longer needed the personnel who’d been running those tests, though several employees, along with the shuttered testing, were moved to the core lab site at the Orlando hospital.

“The savings were just incredible,” he says. At the core lab, “We can do things better, faster, cheaper. And the added volume from the hospital was a drop in the bucket.”

If this all sounds very eat-or-be-eaten, well, it is. Depending on where you sit on the food chain, it’s either an inspiring tale, or one that prompts you to start sleeping with the lights on, lest someone turn them off permanently to save money.

Though the Orlando lab is part of a fairly large integrated health delivery network, its story is not unique to IDNs. All laboratories, including smaller and non-networked labs, face similar prospects and perils, especially in the current era of health care reform. Even a storefront psychic could predict where reimbursements and payments are going. Says O’Sullivan, bluntly: “They’re not going up.”

Too, the mantra of better-faster-cheaper has remained omnipresent in health care. At the same time, service levels must remain high. And like the oil spill in the Gulf, none of this is going away. “You can’t not think about it,” O’Sullivan says.

When O’Sullivan thinks about these issues, he sees daylight, not darkness, not just for networks like his, but also for medium-sized—three hospitals, say—systems with around a thousand beds and a few hundred outreach clients. Such systems tend to grow through acquisitions, with each hospital boasting its own full-service lab. “There’s opportunity to centralize a lot of work,” O’Sullivan says, speaking from experience.

The model these IDNs use is commonly called a core lab, though there’s plenty of dithering about what that means, exactly, and whether “core” is the mot juste.

“Are we a core lab?” asks Priscilla Cherry, MBA, MT(ASCP), president, laboratory services, Fairview Health Services, Minneapolis. “Hmmm. How do we define ourselves?” Fairview has seven hospitals and 41 clinics; the specialized testing—cytogenetics, cytology, molecular, microbiology, histology—has been centralized at one site, at the University of Minnesota Medical Center. At Fairview, Cherry and others refer to this as the Special Diagnostics Laboratories.

O’Sullivan uses the word “core” with great frequency. But, he notes, those in the Florida Hospital system simply refer to the core lab site as “Orlando.”

At Sutter Health the core lab is known as the “shared” lab. “We’re in California, OK?” says David Velasquez, CLS, MT(ASCP), administrative director of Sutter Health, Livermore, Calif. “So you have to be politically correct, and make sure everyone ‘shares’ in the experience,” he says with a laugh, before noting that the meaning does run deeper. The shared lab, a stand-alone facility, is seen as an extension of the 31 Sutter affiliates it serves.

Each lab traveled its own path toward centralized testing. Florida Hospital went from four hospitals to seven hospitals between 1990 and 2000, acquiring two and building a third. Fairview’s specialized lab traces its origins to 1997, when the system merged with the University of Minnesota’s medical facilities.

Sutter Health’s affiliates include small, medium, and large hospitals across Northern California. The affiliates also include quite a few foundation models, Velasquez says—these are essentially outpatient clinics, some of which have their own laboratories.

Sutter’s genesis dates back to about 1998, when Sutter’s laboratories first began to function as a system. One of the first laboratory endeavors was to create a handful of regional core labs. One was at the Mills-Peninsula affiliate, where Velasquez was working. He helped expand the test menu and brought in testing from affiliates assigned to the region.

The broad menu appealed to affiliates outside the region, and eventually the lab became a victim of its own success. “We got to the point where we outgrew ourselves as a hospital-based core lab,” Velasquez says.

In 2006, pathologist Ron Workman, MD, vice president of system laboratory operations at Sutter, approached Velasquez with a bigger vision—to move the core lab out of the hospital and into a more central location, with an expanded menu. After much planning and a lengthy approval process from Sutter Health, the green light was given. “Pretty much all of 2008 was consumed with designing and building,” Velasquez says. The new lab, which opened in December 2008, was created in an existing shell of a building and rebuilt to lab specifications.

Sutter Health owns the lab; specialized testing is no longer linked to certain affiliate labs, in other words. That, says Velasquez, has broken down boundaries, making it easy for all the affiliates to send work to the lab.

Each of the affiliates continues to do some of its own testing. “Absolutely,” says Velasquez. “Every one of our affiliates has a different business model.” In general, the hospitals take care of their own inpatient testing to the extent their own menus permit. For tests not on their menus, the test will either get sent to the shared lab or to a reference lab (usually the former).

The shared lab performs basic molecular testing, including CF, HPV, GC/Chlamydia, thrombolytics, and HIV and HCV viral load. It also has a full-service microbiology lab, which is no small feat, says Velasquez. “In Nor­thern California, microbiology is a very difficult endeavor, mainly because of the labor shortage,” he says. (This has to do with the state’s stringent licensing laws, he says.) “Most of our clinics have given up their microbiology laboratories, and it’s centralized here.” Some hospitals have done the same, “mainly the more remote facilities that just cannot find microbiologists.”

Proof that centralization doesn’t mean automatic Darwinian slaughter, Sutter Health was able to keep jobs. When the lab left Mills-Peninsula, many of the lab employees moved to the Livermore site as well. Even that was insufficient, and the shared lab recruited additional staff, many of whom came from a recently closed LabCorp facility.

All three labs have found ample rewards with their core models.

Sutter Health’s goal in setting up the Livermore lab was to reduce dependence on commercial labs, Velasquez says. The combined volume of work going outside the system was significant, even with the earlier regional core labs.

Much of that high-volume work has been brought in-house, Velasquez says, and the menu has been expanded aggressively. “We’re getting to the point where we have to sit down and look at a make-buy decision on every test that we bring in now,” Velasquez says. Easy efficiencies have given way to less obvious ones.

The shared lab can capture economies of scale, but it’s not necessarily the one calling the shots. Affiliates make their own decisions about whether they want to run tests in-house or send them to Livermore. Frequently this is a staffing issue, especially at some of the more remote sites, which have a hard time recruiting CLSs, says Velasquez. Some labs may find it easier to start sending work to the shared lab when a CLS retires, rather than try to hire a replacement.

So is this venture a money-making one, or a money-saving one? Both, says Velasquez. He notes that California is an anti-markup state; however, Sutter also fits within the safe harbor of an integrated health network. When the shared lab brings a test in-house, the affiliates can bill at their own rates, which would be greater than what a commercial lab would charge them, says Velasquez. “So it does generate revenue for the affiliates.”

The shared lab operates on a break-even model: Testing at the Livermore site is done at cost to the affiliates. Velasquez says this fits the model of the shared lab being an extension of the other sites. “In essence if they run a test on-site or if they run it here [in Livermore], they own the test.” The financial benefit includes both a purchased services cost savings and a net revenue component that varies, depending on the outpatient percentage and payer mix, Velasquez says.

Sutter Health—along with all its California brethren—has another unique circumstance to take into account. All hospitals in California are required to meet a seismic standard that was enacted after the Loma Prieta earthquake in 1989. Older hospitals that don’t meet the standard can either retrofit or rebuild their facilities.

Most are choosing to rebuild, according to Velasquez, and in doing so they are designing their laboratories as immediate response labs. Sutter Health itself is rebuilding several hospitals. When they open, Velasquez says, “They will have a small footprint of a laboratory inside of them.” All nonessential testing will be done at the shared lab, which saves money. But smaller also translates into “less expensive to build.”

At Fairview, Cherry reports, savings have come from the sheer size of the network—vendors are happy to make deals with someone who speaks for seven hospitals and 41 clinics. By standardizing major instrumentation in one area, Cherry says, the system saved an additional $395,000 off an already discounted price. In chemistry, they were able to save an additional $406,000 off an already discounted price.

Standardizing instrumentation has also meant savings in labor—when labs are running the same machines, fewer people are needed to produce correlation studies and write procedures. Though the cost savings here are harder to calculate, Cherry estimates the figure is easily into the $200,000 range.

These savings are easily apparent, but Cherry and her colleagues have to work at showing administrators the whole financial picture. Current benchmarking tools “don’t do us justice,” she says, explaining that they can’t account for the constant transit between facilities or for the seemingly lopsided billing arrangements. Administrators may wonder why expenses are heavy at one site and light at another, viewing the disparity as a reason to demand huge budget cuts. “They don’t see that we’re sending everything to that one site,” Cherry says. “You have to look at the value of the whole service, versus picking apart the pieces.”

O’Sullivan concedes he and his Orlando colleagues have had to convince administrators that the model works financially, that it benefits the organization as a whole. Along the way, they’ve tried different ways of monitoring financials.

One approach is to transfer the revenue and the expense to the core lab. Another model keeps the revenue at the ordering center and ships the expense to the core lab. This creates an apparent loser, if you look at the small picture only. “You have to look at it as, what’s good for all of us?” O’Sullivan says. The other option is to keep the revenue at the originating facility, the expense at the core lab, and then transfer a cost-per-test from the ordering facility to the core lab. “That’s the model we have right now,” O’Sullivan says.

The worst things labs can do, O’Sullivan says—noting that he’s seen this occur in other systems—is to keep so focused on the bottom line at individual facilities that testing gets sent outside the system. “That’s the politics of the situation.”

At Orlando, says O’Sullivan, the CFOs decide how they want to divide the money. “But we all know what’s good for the family. You want to keep the money in the family.”

If administrators have a hard time seeing the financials of the core lab, clinicians have a hard time seeing the core lab period. “It’s invisible,” says Velasquez. In a survey sent to clinicians about a year after the venture opened, recipients were asked (among other things) about the shared lab. The answer from most, says Velasquez, was, “What’s the shared lab?”

“I don’t think our clinicians know how it’s set up,” Cherry says of the Fairview specialized lab, although the testing site is indicated on the lab report. Nor do they need to, she says. “They just know that when they sign on to their electronic health record, the results are there.”

What do Florida Hospital clinicians think of the setup? It’s unseen, says O’Sullivan. “They order it. It gets done.”

This invisibility is remarkable in a sense, given the transit of tests from one site to another.

When Fairview’s specialized lab took root, it made sense to keep the esoteric testing at the university site. Thus began the great migration, though occasionally it might be called the Trial of Tears. Some tests have moved from hospitals to the specialized lab. Others have moved back. And yet others have returned to the fold from reference labs.

Cherry meets regularly with her lab directors to figure out which tests should be run where. Tests have their own life cycle. They might start out being specialized, and then over time they become so rote they land on a basic chemistry analyzer. Ferritin has been one such test, TSH another.

Cherry and her colleagues keep a sharp eye on the tests going out the door. Once the system has sufficient volume for a test, it usually makes sense to bring it back in-house, to the specialized lab. In addition to the monthly volume reports, lab management does a deeper data dive annually to see what might be reeled back in. Recently, Fairview learned from its reference lab, ARUP, that its orders for vitamin B-12 testing were astonishingly high. This was eventually traced to the bariatric physicians, who didn’t realize the B-12 test was expensive, that it was being sent out, or that a water-soluble vitamin test would suffice.

The detailed vetting helps Cherry and her colleagues get their arms around a huge process. “Resources are so tight. We had to structure things, instead of allowing everybody to just bring in tests willy-nilly.”

When making the decision, O’Sullivan and his colleagues have plenty to consider. Do the technologists at the remote site have the expertise to run the assay? What about hidden costs: reagents, controls, CAP proficiency testing? If these are not cost-prohibitive, the preference is to push the tests back to the local level.

It’s like one big board game, trying to figure out where to move what pieces, and when. There’s nothing static about volumes, prices, personnel, or service. Vitamin D testing was recently brought in-house, to Orlando, when demand for it exploded. So were allergy and aspergillus tests. “And we’re constantly looking for molecular-type testing to bring in, such as herpes simplex virus by PCR,” O’Sullivan says. The HSV move was a service issue—faster results let physicians start antiretroviral therapy sooner. H1N1 came in-house because it’s an expensive send-out, and demand was high.

O’Sullivan recalls a conversation he had at one site regarding blood gases. The respiratory therapy department could perform the testing on a dedicated instrument, but did so infrequently. “I said, ‘What does it cost you to run a blood gas on this machine?’ And the manager thought, and said, ‘About $10,000 to do each one.’” An exaggeration—but after factoring in cost of owning the machine, running controls for sporadic tests, and idle time, it turned out to be more cost-effective to run the test on a point-of-care device, despite its apparent higher cost-per-test. “You’ve got to weigh all these considerations,” O’Sullivan says.

When Cherry talks about the Fairview system, she makes it all sound simple. Don’t be fooled. “We have to work at being a system,” she says. “It doesn’t run by itself, I’ll tell you that.”

It’s not unusual, in a system the size of Fairview, for individuals to try to blaze their own trails. “Somebody will go off in a different direction and try to do something completely different, and we’ve got to b-r-i-n-g them back in. People like to head out and d-o-o-o-o-o something,” says Cherry, unpacking her sentences with gusto. The next thing you know, she says, the menu is overflowing with tests that have “absolutely zero utilization.”

The management team meets every two weeks, face-to-face. “We figure out how to solve problems together, and how to respond as one voice.”

That united front helps when one lab digs in its heels. The specialized laboratory wanted to hang on to ferritin testing, she recalls. “They kept saying, ‘We have all the volume in one place, and we can keep costs down, etc. etc.’ And I said, ‘It’s. Moved. On.’ We talked about the fact that it’s matured, that it can be put on a routine instrument. And since we’re using the same instruments, the reference ranges are the same.”

O’Sullivan agrees there can be resistance when it comes time to move a test. “There’s a labor component, and there’s a cost component. And then there’s someone saying, ‘I’ve been doing that test for 10 years, 50 years.’” Before a test can be moved to the core lab, the system lab team has to talk the matter through and show how the move will save money, improve service, or both.

And sometimes you have to “reallocate staffing,” he continues. O’Sullivan likes to talk about this as being good for “the family.” It’s a sentiment echoed by leaders at other IDNs as well. And other labs, as O’Sullivan points out, may find it difficult, or unlikely, to keep going it alone. Centralizing work, by choice or circumstances, can have tremendous payoff. Some families are like the Waltons. But then, the Sopranos were a family, too.

Related Links

• Fairview’s new and replacement technology/test request form (PDF, 964 KB)