August 2012
Feature Story

Karen Titus

T.S. Eliot lied.

It’s not just the journey that matters (as he’s often credited with saying). So does the arrival. And so does settling in comfortably at the destination.

Just ask Greg Sossaman, MD, who in recent years has overseen the addition of several community hospital laboratories to New Orleans-based Ochsner Health System. That meant standardizing patient testing across sites that had, until then, been marching to their own brass bands.

Becoming standardized has presented the expected challenges and led to one unexpected one. Staying standardized, it turns out, is as easy to achieve as perpetual motion. “It takes more work to stay that way. A lot more work,” says Dr. Sossaman, chair, pathology and laboratory medicine. “And that’s something we didn’t realize.”

The journey as well as the arrival has been filled with discoveries and questions, says Dr. Sossaman, who spoke about Ochsner’s experiences at the 2012 Executive War College in New Orleans this spring and in a recent CAP TODAY interview.

“I don’t have a lot of answers,” says Dr. Sossaman. “I just have a lot of, ‘Well, this is what we did.’ I don’t know if it was the right thing.”

He’s being modest. Whatever the challenges, Ochsner continues to build on decades of success.

Ochsner was founded as a private clinic some 70 years ago. As the clinic evolved, its five founders started a foundation, which built a hospital.

The influence of those early years continues, with Ochsner simultaneously supporting hospitals and clinics. Early on, the clinic supported the hospital, says Dr. Sossaman, while the reverse is true today, thanks to changes in reimbursement. Ochsner Medical Center, with its main campus a 450-bed hospital in New Orleans, might be considered the flagship, but the system is spread across eight hospitals, 35 health centers, and close to 850 physicians, all of whom are employed by the clinic and who represent more than 90 medical specialties and subspecialties.

The laboratory has always been centralized, says Dr. Sossaman, with the core lab located on the second floor of the hospital’s main campus. The LIS, in another historical nod, has been in place for nearly three decades. Originally a Lab Force product (the vendor is no longer in business), it has been sustained over the years through Ochsner’s purchase of the code and support from in-house programming staff. “Major portions of it have been completely rewritten,” says Dr. Sossaman.

Over the years Ochsner remained a fairly self-contained organization, though growth was not unknown. The primary care provider network expanded, as did investment in internal medicine and primary care. During this period, Ochsner used what Dr. Sossaman calls a cookie-cutter approach to help new clinics coordinate with the centralized laboratory. The template was particularly useful when the powers that be forgot to tell the core lab they were opening a new facility, “which actually happened a few times,” Dr. Sossaman says.

Into this model—homegrown, tightly knit, and with a high degree of harmonization thanks to use of a common LIS and EMR—came several new hospitals. Blame, or thank, Hurricane Katrina, which devastated the area in 2005. Ochsner’s hospital survived fairly intact, thanks to its relatively high elevation and location next to the levee. The ED never closed, nor did the hospital, and patients did not have to be relocated. But the hospital did suffer $27 million in property damage, and 4,000 of its 8,000 employees either relocated or were displaced by the storm.

That could have signaled Ochsner leaders to hunker down, count their blessings, and try to return to normal as quickly as possible. Who would blame them?

Instead, “We used the never-let-a-good-crisis-go-to-waste” approach, says Dr. Sossaman. That meant purchasing several hospitals—each with its own ways of doing things—and bringing them into the Ochsner fold. At the same time, Ochsner has been overhauling its own homegrown systems. In other words—standardization × 2.

If that sounds like an undertaking that can be done only with a plan, think again. “There was an institutional urgency to convert, and we didn’t have the luxury of time to adhere to a traditional plan,” Dr. Sossaman says. They first converted their IT infrastructure, then began to build on that. Rather than a controlled entry, Ochsner cannonballed into the pool.

Certainly other institutions—Dr. Sossaman cites Mayo Clinic as an example par excellence—have taken the obvious approach: sticking to a methodical plan of first standardizing instruments, followed by processes, procedures, and competencies for technologists, then wrapping it all up with a standardized information system.

Ochsner had to do without such obvious logistics. “We did it contrary to popular wisdom,” says Dr. Sossaman. But it appears to be working.

In October 2006, Ochsner purchased three New Orleans community hospitals from Tenet Healthcare Corp. In 2008, it added a fourth hospital.

Then the real work began. Ochsner’s leaders decided they could save money by having all the hospitals on the same IT platforms.

But it didn’t make sense to keep the homegrown LIS or EMR over the long run. “Once we extended across multiple hospitals, it was too laborious, too time-consuming, to administer it over the entire system,” Dr. Sossaman says.

In fact, he adds, even without the Tenet influx, Ochsner would likely have moved off its legacy systems sooner rather than later. As the years go by, upgrades tend to be limited to what current staff can handle, he notes; furthermore, it’s hard to take advantage of major changes in software with a legacy system. Imagine trying to load an iPhone app onto a 1970s mobile phone. “As the newer systems have evolved, they’re much more robust,” Dr. Sossaman says, pointing to their data mining and data reporting features.

At the same time Ochsner decided to purchase a new LIS (from SCC Soft Computer), it also decided to switch to the Epic EMR. Dr. Sossaman offers a somewhat inelegant formula to describe Ochsner’s goal: “We were hoping integration plus standardization would equal harmony for us.”

His terms are a little idiosyncratic, perhaps not surprising for someone who works in a system long dominated by homegrown solutions. He defines standardization as using similar platforms and instrumentation, methods, reagents, procedures—“every-thing,” he says—in the same fashion. Harmonization, in his view, enables a laboratory to generate results that can be combined, without any additional manipulation, with results from another laboratory.

The integration piece “is the highest level of standardization,” he says. With it, physicians can use the same LIS or middleware, for example, with the same rules, across an entire system.

Adding the new hospitals has meant both standardization and integration over time, Dr. Sossaman says. First, there was the not-so-little matter of integrating the Tenet hospitals to Ochsner’s homegrown systems.

Only after this began did standardization get underway. This, too, was affected by the hurricane. “We were not in a very capital-rich environment after Katrina,” he says. All departments were struggling for capital, and the core lab was several years behind, he says, in terms of needed instrumentation. By the time capital did become available, the lab lagged on automation and middleware throughout the system, including the former Tenet hospitals.

They’ve since brought in new instrumentation (including chemistry, hematology, and coagulation analyzers), middleware, and robotics, with an eye toward consolidation. For additional cost savings, Ochsner has standardized basic kits, such as pregnancy tests and strep kits.

From a starting point of 42 analyzers across 19 platforms, Ochsner now has 31 analyzers across nine platforms. It has a robotic line at the main campus and standardized instrumentation at all the hospitals. This has made it possible for the lab to consolidate and reduce the use of reagents, proficiency testing material, calibrators, and controls, and decrease the time spent on maintaining so-called extra equipment. Consolidating immunoassay instrumentation has allowed the lab to bring current send-out tests in-house, as well as decrease the need for batching and the time spent on processing specimens.

As of mid-summer, they were still building the new LIS. Epic was being rolled out throughout the institution. “We’ll begin converting labs to the new LIS once all facilities are live on Epic,” Dr. Sossaman says. “Again, it’s a rolling time frame.”

It’s been tantamount to cleaning house and putting on a new addition, both at the same time. “We didn’t want to let the energy of changing to a new computer system go to waste,” is how Dr. Sossaman explains Ochsner’s extensive “renovation.”

The biggest hurdle has been determining which tests could be grouped within one display in the EMR. Would certain tests match up from one platform to another? Reference ranges were another concern. Critical values had to be aligned across the system and approved locally by the medical staff. That meant Dr. Sossaman and colleagues visited with the executive medical committees at each site and spent time educating physicians about the need for system standardization.

Dr. Sossaman is the first to admit he didn’t try to do it all himself. He worked closely with his medical directors, who, being mostly employed physicians, were able to answer other physicians’ concerns.

Questions were plentiful. “You have to overcommunicate, which is tough to do across a system,” Dr. Sossaman says. “Physicians don’t read e-mails, they don’t read newsletters, they don’t read memos.” Having medical directors maintain a connection at each local site helped.

His goal now is to examine test menus and utilization, which has proved to be harder than he anticipated. His first step was a typical one: starting a utilization committee. His first finding was typical too: There’s no one-size-fits-all solution.

Getting data has been difficult, especially using the legacy information systems. Finding out who’s ordering what has been nearly impossible, he says. The send-out volume is particularly nettlesome, and costs continue to climb. And, as an academic institution, Ochsner must contend with test ordering from residents and academicians. “It’s very difficult to get your arms around that,” Dr. Sossaman says. For reasons ranging from technical to educational to political, tackling utilization has been slow to get traction. “A lot slower than I thought it would be,” he says.

If anything, the little things tripped him up. When the former Tenet hospitals were being moved to Ochsner’s legacy LIS, says Dr. Sossaman, he and his colleagues heard several common refrains: “Oh, we forgot to tell you about this test,” “We need this test code,” and ”Oh, we do this differently than you, and we forgot to tell you.”

But, he notes, these little reveals can also be an advantage. Dr. Sossaman describes it as hitting the reset button for every process as systems are upgraded. “All these workarounds fall out,” he says, recalling staff consternation when they realized the new system didn’t accommodate their “fixes.”

Such little things are probably the biggest part of any IT implementation, but would have been impossible to anticipate, he says. “Nobody would have fessed up.”

Standardization was also slowed by Ochsner’s early laissez-faire attitude toward Tenet practices. “I wouldn’t say we came in with kid gloves on, but we tried to respect their local autonomy,” says Dr. Sossaman, who compares it to the balance between federal and states’ rights. As with any core lab with an eye to standardization, he says, “there’s a delicate balance between local autonomy and the need for central decisionmaking.”

Perhaps the biggest surprise has been the energy it’s taken to tread water—to stay standardized, that is. From where he stands, the journey, for all its challenges, may have been easier than the arrival.

It is, says Dr. Sossaman, a case study in unintended consequences, akin to other great leaps forward, such as e-mail. “It seems like standardization would make your life easier,” he says, punctuating this sentence with laughter, “but in a lot of ways it makes it harder.”

With standardization, “You almost need a system to maintain your system,” he says; if he were master of Downtown Abbey, he’d be hiring staff to keep track of the servants. At Ochsner, it’s meant, among other things, continuing to stay on top of critical values. “It keeps coming up and up and up, again and again,” he says. “I didn’t realize how much time I was going to have to spend at a system level, going over those decisions.”

So standardization doesn’t run on autopilot? Not in the least, Dr. Sossaman says. “It’s like the assumption that computers are going to make our lives easier.”

Not that he prefers a nonstandardized world (or, for that matter, a world sans computers and e-mail). “But if you’re going to become a system, you have to have an advantage somewhere,” be it cost savings, wider market share, or improved processes. “Otherwise, it’s just more work.”

Or, as a certain poet might have put it, labs might find their efforts ending with a whimper.