August 2009
Feature Story

Karen Lusky

Proficiency testing referral is back in the spotlight with new regulatory and judicial developments. The issue at stake is whether it’s fair for the government to revoke a lab’s CLIA certificate when the laboratory had no intention of cheating on PT. Some argue that the severe penalty not only devastates well-meaning labs but also threatens patient access to care, especially in areas with limited health care services.

To date, the fate of labs in proficiency testing referral cases has hinged on the government’s interpretation of the word “intentionally,” says Anthony Killeen, MD, PhD, director of clinical pathology at the University of Minnesota Medical Center, Fairview, in Minneapolis. The CLIA statute says that a laboratory that intentionally refers its proficiency testing samples to another laboratory for analysis is subject to revocation of its CLIA certificate for at least one year.

“Administrative law judges and the Departmental Appeals Board have taken the word intentional to mean simply that you meant to send the PT sample to another lab,” Dr. Killeen says. That interpretation doesn’t address the lab’s motivation for referring the sample. “Nobody in the clinical lab profession would have much sympathy for labs that intentionally cheat on PT. Unfortunately, the net that catches the cheaters sometimes catches those who are not cheating,” he says.

The good news is that the Centers for Medicare and Medicaid Services plans to revisit the issue of intent in enforcing PT referral prohibitions. The agency is “looking to provide more clarity in the regulations about what does and does not constitute intentional PT referral,” says Judy Yost, MA, MT(ASCP), director of the CMS Division of Laboratory Services. The CMS and the Centers for Disease Control and Prevention are planning to convene a CLIAC work group to examine that and other issues. The group will include representatives from the government, PT providers, accrediting organizations, and CLIA-exempt states.

The CMS also appears to be backing off its black-and-white approach to PT referral (“PT referral raising red flags with CMS,” CAP TODAY, March 2008). “As we began to review all cases centrally in my office with a team of experts, and began to understand and delve further into the fine points of each case, we adjusted our thinking somewhat from that initial hard line,” Yost tells CAP TODAY.

“Centralized reviews and determinations have ensured more consistent outcomes,” she explains, and enabled the CMS to learn in more depth the circumstances that can “create an environment for PT referral, aside from true ‘cheating.’” This will make it possible for the CMS to develop “better future policies and regulations,” Yost adds.

Amy Daniels, MT(ASCP), investigation manager for the CAP Laboratory Accreditation Program, says more cases of PT referral have been reported to the College in the past three years than in prior periods. She attributes the increase to people being more aware of the PT referral prohibition. But “to date, I haven’t seen a single case of PT referral in a CAP-accredited lab where the lab intended to cheat,” Daniels says. And, she points out, “labs can make much more significant mistakes than inadvertent PT referral and not have their CLIA certificate revoked.”

CAP president Jared Schwartz, MD, PhD, says the CMS has received a lot of feedback from many organizations, including the CAP, on PT referral, prompting the agency to assemble a group to see if it can come up with a solution.

Yet regulators and stakeholders grappling with what constitutes intentional versus unintentional PT referral may have their work cut out for them. “The issue of whether PT referral is intentional,” says Jack Bierig, a partner with the Chicago law firm Sidley Austin, LLP, “is a complicated one.” There are at least three scenarios that raise questions.

The first occurs when a lab technician or technologist refers a PT specimen without the laboratory director’s knowledge. A director of a laboratory that loses its CLIA certificate for PT referral is banned from directing any lab for two years. In such cases, “there’s unquestionably intentionality on the part of the individual who made the referral,” Bierig says. “But should the intentionality of that individual be attributed to the laboratory director who didn’t know about the referral? The case law has been that you can hold the director responsible for the action of all laboratory personnel even if the director didn’t know about the referral.” In Bierig’s view, this approach is unfair—particularly where the director has put in place policies against PT referral and has sought to educate laboratory personnel on the importance of not referring PT specimens.

A second scenario arises when a laboratory sends a PT specimen to a pathologist who is associated with the laboratory in some capacity but is technically not an employee of the laboratory. On this point, Bierig cites a situation in which he represented a laboratory director whose practice had included a pathologist to whom the director routinely sent difficult cases. That pathologist retired from the practice but kept an independent contractor relationship. Notwithstanding that change, the laboratory director continued to send the more difficult cases to the now semiretired pathologist—as he had always done. Unfortunately for the director, he also sent to this pathologist PT specimens that he regarded as difficult. Since the pathologist was no longer an employee of the practice, the CMS took the position that an unlawful referral had been made, revoked the CLIA certificate of the lab, and forbade the lab director to direct another lab for two years.

“The director could not believe that he had to suffer such a draconian penalty for what he regarded as having acted properly and in good faith,” Bierig says. “And, at least in my judgment, it’s hard not to have a lot of sympathy for the well-intentioned director who had no idea that what he had done violated CLIA.”

The third situation is one in which a lab routinely sends out patient specimens to another lab for a certain aspect of the test, such as the technical component, and the laboratory does the same for PT specimens. The CLIA statute, in fact, includes langu­age directing a lab to treat PT samples as it does patient samples. “Where a laboratory routinely sends out patient specimens for some reason,” Bierig says, “there’s a tension between the statutory directive to give assurance that PT samples will be treated just like patient specimens and the statutory provision that says that a laboratory may never refer out a PT specimen. I’m not sure that Congress intended to punish a laboratory for sending out a PT specimen when it routinely refers out patient specimens in the same situation.” There’s no denying there was intentionality, he adds, “but whether it’s the kind of intentionality that Congress had envisioned in the statute is another question.”

Dr. Killeen agrees, recalling a case in which a lab identified an organism on a PT sample and sent the sample for subspeciation to another lab, as it would for a patient sample. “That was considered a prohibited referral,” he says. However, the first laboratory didn’t do that testing, raising another question: “How can you cheat at proficiency testing if you don’t do the testing? Is it even a proficiency testing sample within the meaning of CLIA if a lab doesn’t perform the testing?”

Current law already gives the CMS flexibility to be lenient when circumstances warrant, Bierig believes. He points out that the subsection of CLIA that provides for intermediate sanctions, Sec. 263a(h), provides in pertinent part that the agency “may impose intermediate sanctions in lieu of the actions authorized by subsection (i) of this section.” Subsection (i) is the provision that imposes the penalties of loss of the CLIA license and the prohibition on the ability to own or operate a laboratory. Thus, Bierig concludes that “where circumstances warrant, CMS has discretion to require a directed plan of correction, monetary penalties, and/or payment of the costs of on-site monitoring in lieu of the ‘atomic bomb’ remedies.” He urges the CMS to give consideration to these intermediate sanctions where there was no deliberate effort to cheat.

Apart from the CMS’ efforts to illuminate what’s meant by intentional PT referral, labs might be able to get signals from the court system. The news there is that, in June, Wade Pediatrics of Muskogee, Okla., lost its appeal challenging the revocation of its CLIA certificate for a year in a PT referral case before the U.S. Court of Appeals for the Tenth Circuit (Wade Pediatrics v. Department of Health and Human Services, Centers for Medicare and Medicaid Services). Though the case is not a typical example of unintentional PT referral, it does raise the issues of the lab’s motivation and what constitutes cheating.

According to the court, Wade Pediatrics, on two occasions, checked its PT results by first running the PT sample at nearby Muskogee Regional Medical Center’s laboratory. Wade did, however, submit its own PT answers to the state.

It was a CMS field investigator who directed Wade Pediatrics to work with the Muskogee hospital laboratory to improve its testing standards after Wade failed parts of two proficiency tests in 2005. The CMS temporarily limited the scope of Wade’s CLIA certificate, after which Wade submitted a remedial plan to the CMS in which it said it would “continue internal proficiency testing with assistance and support/guidance” from the Muskogee hospital, according to the court ruling.

“Eventually CMS got wind” that Wade had twice tested its PT samples at the hospital before submitting its PT results, and it revoked Wade’s CLIA certificate for one year, according to the court.

Lawyer Sarah Glick, of Scoggins & Cross in Oklahoma City, handled the Wade appeal and explains: “The Wade lab tech had been testing a representative portion of Wade Pediatrics’ patient samples at the hospital to ensure accuracy. The hospital tech observed the Wade tech doing the testing and offered guidance.” The Wade technician did the same for the proficiency testing sample but by that point [in 2006] had become proficient in doing the testing. Wade Pediatrics was operating on the notion that it was treating its patient and PT samples the same. In its appeal, Wade maintained it didn’t act in an attempt to cheat the PT program but as “part of a training program, undertaken in good faith, to confirm the accuracy of its own work,” according to the court ruling.

In its opinion, the court responded by saying, “Under the statute’s plain terms, any intentional ‘referral’ of a proficiency testing sample ‘for analysis’ in another lab is forbidden,” and “Simply put, Wade is responsible for its intended means, whatever its intended ends might have been.”

Wade Pediatrics persisted in pursuing the appeal despite the legal costs because its owner believed he had done the right thing. “It was a matter of principle,” Linda Scoggins, a principal of Scoggins & Cross, says.

Washington, DC, lawyer Peter Kazon, who has represented numerous labs in PT referral cases, says the Wade Pediatrics case, as he reads it, “is a pretty straightforward case of a lab checking its answer at another facility. That’s not to say that the person doing that acted with a bad purpose or that the basic PT provisions [requiring loss of a CLIA certificate for a long period] makes a lot of sense.”

The Wade Pediatrics ruling maintains the CMS’ perfect score in winning appeals in cases of PT referral. Now the lab community’s attention is on another appeal, by Victor Valley Community Hospital, Victorville, Calif., which faces loss of its CLIA certificate for one year because a lab staff person sent a PT specimen to Quest Diagnostics as part of the hospital’s protocol for handling a patient sample.

The case is being handled by Patric Hooper, of Hooper, Lundy & Bookman in Los Angeles, who represented the laboratories in the landmark lawsuit that halted the CMS’ competitive bidding demonstration in San Diego. Hooper says the Victor Valley case was, at CAP TODAY press time, pending before an administrative law judge, which is the first step in the appeals process.

“We had a live hearing before an ALJ in February 2009, and have been briefing the case since then. We expect a decision within 30 to 60 days. Our most recent round of briefs deal with the Wade Pediatrics decision out of the Tenth Circuit.”

Given past precedent, Hooper doesn’t think the ALJ is likely to rule in Victor Valley’s favor. “If he rules against us, we have one more administrative appeal step with the appellate panel [the Departmental Appeals Board], and then we’d go to the Ninth Circuit.” In theory, he adds, “if the Ninth Circuit disagreed with the Tenth Circuit, that could be a situation where there’s a conflict over how to interpret a statute, which is the kind of issue that the Supreme Court will agree to resolve in certain situations.”

If Victor Valley prevailed at the ALJ level, the CMS could appeal the ALJ’s decision, “so we’d have to go to the DAB,” Hooper explains. But he believes in continuing to challenge the case because “there was no cheating involved, and no harm.”

The CLIA statute does not specify that a lab’s CLIA certificate must be revoked for instances in which a laboratory communicates with another lab with a different CLIA number about its PT results before the PT event is over. But, Dr. Killeen, in an Archives of Pathology & Laboratory Medicine article chronicling PT referral cases from 1993 to 2006, says the CMS views specimen referral and result communication as being “effectively the same, and either can result in revocation of a CLIA certificate” (Arch Pathol Lab Med. 2009;133:979–982).

Daniels of the CAP’s Laboratory Accreditation Program knows of instances in which labs did not realize they were engaging in interlaboratory communication about PT results because they viewed themselves as all working within the same organization or system. “But where PT is involved, each lab with a CLIA number is a separate laboratory,” she says. For example, Daniels has seen cases in which an organization may have a parent laboratory and secondary labs with different CLIA numbers. And the person at the parent lab may be in charge of point-of-care testing for the entire system. If a secondary laboratory sends PT results to the POC coordinator at the parent lab and the coordinator sends the PT results to the PT provider, interlaboratory communication has taken place, Daniels cautions.

The upshot: “You cannot communicate results about PT across CLIA lines,” she says.

Other examples of interlaboratory communication that the Laboratory Accreditation Program has encountered include an instance in which some­one entered PT data into the information system so that the data became accessible to everyone within the hospital system. In another case, someone e-mailed a group of labs within a system about a problem with an instrument that was discovered with a PT sample.

CMS’ Yost says that by clarifying the PT referral regulations in terms of what constitutes intentional referral, the statute can remain as it is. Yet some in the lab community believe that the CLIA statute may require updating to address PT referral, among other issues.

“I would think that at some point it may be necessary to go back and revisit the statutory language,” says Dr. Killeen, who believes the intent of the statute was to prevent cheating on PT. “At the time CLIA was written, it was well known that some labs would send samples for comparison of PT results to another lab. That was why Congress wrote the no-referral language for PT samples.” It’s unlikely that Congress meant to outlaw referral of PT samples when patient samples are referred “for perfectly valid clinical or operational reasons,” Dr. Killeen says. In fact, he adds, “PT samples should follow the same path as patient samples if one were considering this from purely a quality assurance point of view.”

CLIA also hasn’t kept pace with laboratory technology in the area of digital pathology, raising the question of whether sending a digital image of a PT specimen to another lab constitutes PT referral, says lawyer Jane Pine Wood of McDonald Hopkins in Dennis, Mass. Wood, in fact, recently dealt with a PT referral case involving digital pathology that was “quietly resolved” at the final hour. In the case, the laboratory, which did only the technical component for the pro­ficiency test in question, sent a digital image of a molecular pathol­ogy specimen to a specialty pathology group out of state for professional interpretation, as it normally would a patient sample. Unsure how to handle the situation, the CLIA inspector contacted the CMS, which said it was a PT referral, Wood reports.

“CMS ended up withdrawing [the complaint] the day before the lab’s CLIA certificate was [to be] revoked,” Wood says.

In Dr. Killeen’s view, a strict liability standard is applied to laboratory directors based on the “captain of the ship” analogy whereby the captain is responsible for everything that goes on onboard. But “a captain has complete authority on the ship,” Dr. Killeen points out, whereas “the employed lab director … may not even have hiring and firing authority in a large organization.”

Enforcement of the prohibition against PT referral, many say, should be consistent with the intended effect on the quality of and on patient access to laboratory services.

“In plenty of these PT referral cases,” Kazon says, “ultimately the sanctions work to the disadvantage of patients and everyone else. You may have a hospital lab that, because of some error, can’t serve patients anymore.”

Wood agrees, noting that the goal is to “discourage sloppy quality and fraudulent behavior,” while also encouraging people to provide lab director services, especially in rural areas where it’s difficult to get a pathologist to serve as director, even without the liability related to PT referral. Many of the PT referral cases do occur in remote or rural hospitals where the lab director is not on site full time.

Wade Pediatrics fought the revocation of its CLIA license in part because the practice is in a rural area of Oklahoma where a lot of children are on Medicaid, Glick says. “Probably 75 percent of the [Wade] practice is Medicaid patients, and having the lab closed meant those patients had to have their lab work done elsewhere.”

CAP president Dr. Schwartz says, “The whole problem with all of health care is that everyone wants consistency, but situations are different, and people are different, and patients are different. There has to be some flexibility for special circumstances and situations. But often achieving flexibility isn’t easy.”

“One also has to look at the ultimate goal,” he adds, “and that should be that the lab has a system in place to ensure it gets the right results for a given patient.”

As the PT referral saga evolves in the regulatory and judiciary arenas, and perhaps eventually in Congress, Hooper cautions labs against complacency: “Labs are competitive and [loss of CLIA certification] for PT referral is one of those things that I’m not certain people are very concerned about—until it happens to them.”

A laboratory, Kazon says, “cannot overemphasize to its employees the importance of understanding that they should never, ever, ever refer a PT sample to another lab—period. People should know it’s a situation where even though they normally send a test to the lab down the street to do part of the test, if it’s a PT sample, don’t do it.”

Yost warns that labs that perform distributed testing or send reflex or confirmatory testing or consults to another lab must be especially cautious when training employees. “Many of the incidents occur in these situations,” she says.

She urges labs to have “robust policies and procedures in place to avoid any perception of cheating.”

“The devil,” she says, “is truly in the details.”