CEO SUMMARY: Revocation of a lab’s CLIA license as penalty for inadvertent errors in handling proficiency tests (PT) is not a new problem. As explained here, most clinical laboratories have appropriate protocols for handling PT samples. But, when errors occur, the Centers for Medicare & Medicaid Services (CMS), based on its current interpretation of the CLIA law, does not distinguish between an intentional effort to cheat the proficiency test process and an inadvertent PT error.
HOW FEDERAL OFFICIALS REGULATE proficiency testing (PT) has been the source of debate and controversy since the inception of the Clinical Laboratory Improvement Amendments (CLIA) statute. It is a story with two sides.
On one side, officials at the Centers for Medicare & Medicaid Services (CMS) are tasked by the law with regulating clinical laboratory testing in a manner that protects the health and safety of patients. Assuring that labs operate a competent proficiency testing program as specified by CLIA is part of this responsibility.
On the other side, clinical lab professionals regularly take issue with two aspects of how CMS regulates PT as part of CLIA. First, it is observed that CMS officials interpret the CLIA language in a most restrictive manner. Second, the enforcement actions levied by CMS as a consequence of its interpretation—including revocation of a lab’s CLIA license—can be overly harsh.
Some critics go further, stating that, when it comes to proficiency testing, both the CMS interpretation of the CLIA statute and its enforcement actions are not what Congress intended when it passed the CLIA law. That is why news that the laboratory of the Ohio State University Wexner Medical Center (OSUWMC) faces revocation of its CLIA license in the wake of self-disclosures of errors in the referral of PT samples again brings these issues front and center.
Explanation of Mistakes
Because it is a story with two sides, THE DARK REPORT sought an expert who could explain the issues. What follows are the comments from an individual who has managed clinical laboratories and who has participated, at the national level, in different aspects of policy-making and interaction with federal agencies in matters relating to laboratory testing. To encourage full candor, THE DARK REPORT agreed not to identify this individual. The person was able to review the letter sent by CMS to OSUWMC and the response to this letter sent by the OSUWMC laboratory director.
“To those on the outside, I’m sure it looks like there’s no logic behind the sanctions that CMS issues for unintentional PT violations,” said this former lab director. “But the law is proscriptive on how these cases should be handled. CMS would like to be more flexible but there is currently no flexibility in the law. And, there is no way to stop the process once it starts.
“The CMS letter notes that the PT errors at OSUWMC constituted ‘…immediate jeopardy and is likely to cause serious harm to the individuals served by your laboratory,’” observed the source. “But in fact, no patient was harmed or would have been harmed by this unintended error! Referring the specimen to another lab for further testing was, in fact, the standard of quality patient care for this lab. Theoretically, not referring the specimen would put patients at risk—not the other way around.
“Therefore, where is the perspective?” asked the expert. “When you say, ‘immediate jeopardy,’ that should mean that patients were and are, truly, in danger. But in this case, that’s not even close to being true.
“There was no risk of harming a patient,” observed the lab director. “So we have to wonder where the immediate jeopardy was. It would be different if that laboratory normally ran the additional testing on its own for all patients and referred just the PT specimen. However, that is not the case in this issue involving OSUWMC. “Did the OSUWMC lab make a mistake? Yes, it did,” added the lab director.
“And both the language of the CLIA law and the way the regulations are written specify that—when a lab sends a PT specimen out to another lab—that action by itself qualifies as putting patient care in immediate jeopardy.
“It is essential to acknowledge that, if a lab sent a PT test out in an effort to cheat—in other words, to make sure it got the right result—severe sanctions, would be required,” said the expert. “Because in this instance, there was intent to deceive.
“But if the lab follows the same protocol with the PT sample as it does with a patient specimen and sends out only a portion of a test and only at a certain stage—which is what OSUWMC did—then CMS should know that the lab is simply following its own procedures,” observed the expert. “In this case, the fact that the PT sample went out is clearly an unintentional error.
“However, under its current interpretation of the CLIA law, CMS views this one inadvertent act as being akin to cheating,” the former lab director explained. “Keep in mind that those in the regulatory business may see only the bad side of labs. If you only see the bad side of the clinical laboratories, it is understandable that you may start to question the integrity of every process.
Circumstances Of The Case
“Now, there are circumstances in this case that seem to support the actions of CMS when you consider that lab sent up to six proficiency test samples to another lab,” stated the expert. “A lab could make a mistake once and maybe twice. But after three or more times, it should surprise no one that regulators conclude that those events represent a potential problem.
“In addition, in the OSUWMC response letter, there is not a full explanation about why just those specific specimens were referred to another lab,” observed the lab director. “The letter discussed how lab staff are instructed to handle PT specimens. But the lab did not fill in the details for the surveyors about what exactly happened regarding these PT samples.
“In fact, a lot of people in the lab understood that the PT specimens should not go out but they still went out,” continued the expert. “The lack of a full explanation of what happened with those six samples likely raised questions in the surveyors’ minds.
“But now—in the defense of OSUWMC—it reported that, from October 2010 to September 2011, its lab performed 9.2 million patient tests and 9,200 proficiency tests,” said the expert. “The laws of probability tell you that you are going to make a mistake or have human error when such a huge number of proficiency tests are regularly run in a major clinical laboratory.
How to Define ‘Intent’
“What we have in this case clearly demonstrates the problem CMS has in enforcing the rules as written,” the former lab director continued. “CMS is aware of this issue and has discussed ways to address the PT referral problem.
“CMS officials know they need to figure out a way not to penalize laboratories where PT samples were sent out inadvertently,” noted the expert. “But they have yet to come up with a solution that makes the language of CLIA unambiguous.
“The problem stems from language in the law,” said the lab director. “The law says that any laboratory that intentionally refers its PT samples to another labora- tory for analysis will have its certificate revoked for at least one year and be subject to fines and penalties as well. And the CLIA regulations included stringent penalties for this ‘intentional PT referral.’
“To correct this problem of confusing regulations, we understand that CMS is considering whether and how to revise the definition of ‘intentional PT referral,’” commented the expert. “Now that Congress is addressing it as well, we might see some resolution in the coming months.”
Every Lab Has Procedures for Proficiency Tests, But There Are Several Ways for Mistakes to Occur
IDEALLY, ANY CLINICAL LAB SCIENTIST running a proficiency test (PT) does not know it’s a PT sample. The PT sample should be indistinguishable from any other specimen in the lab.
“PT samples should be totally blinded to the staff,” stated an experienced lab director. “The lab director or the compliance officer or both will know that a PT sample is in the lab that day. But they can’t tell the bench techs anything.
“In most labs, the PT sample has a bar code label and that label may have a generic name, such as John Doe or Jane Doe,” noted the lab director. “This is consistent with policies for live samples where a positive patient identification has yet to be made. “Some labs put only a unique patient identifier, a test number, and a bar code on every sample, PT or otherwise,” continued the expert. “Labs have many ways to make the PT samples anonymous.
“So if the results of the PT sample are such that the next step would be to send it out, the sample would either go into the refrigerator or go to some area of the lab for processing,” he added. “Whoever handles the send outs would take it from there.
“All the while, the compliance officer would monitor the send out area and refrigerators to make sure the PT sample doesn’t go out,” the expert said. “The compliance manager would pick it up and make a log notation that the PT sample testing was complete and that the lab had followed its procedures just as it would for a live patient specimen.
“For certain tests, the lab has told CLIA that it performs only the initial screen and a different laboratory does the rest of the test,” stated the lab director. “The lab would thus report a positive or negative—or report the presence of a certain result, but not the level.
“Despite all these procedures, mistakes in handling the PT sample can happen,” added the expert. “A PT sample will be referred out because: 1) it was a mistake that the compliance officer didn’t catch; or, 2) a failure to train staff properly; or, 3) due to high staff turnover.
“Or, 4) it could be that a lab sends the PT sample to another lab in the same system but because those two labs have different CLIA numbers, that would be a PT violation,” con- cluded the lab director. “This illustrates a few reasons why a PT sample gets referred out and many of them are just simple mistakes. It is also why CMS needs to have some discretion for cases such as at OSUWMC where there was no intent to subvert the PT process.”