CEO SUMMARY: As happens now and again, a rather typical example of an error in lab testing has made the nightly news in Indianapolis because of one justifiably irate patient who got a false positive test report for an STD. One pathologist, asked by THE DARK REPORT to assess the public information about this episode for lessons to be learned, pointed out that there are at least four major systemic problems in today’s lab testing environment that contribute to these episodes. But fixes to these systemic problems are not likely to happen soon.
EACH TIME A CLINICAL LABORATORY determines that it has reported inaccurate lab test results for one or more patients, it is required to take specific actions to correct the situation and minimize patient harm. It must also conduct a root cause analysis to determine the source of the error and fix those problems.
On April 4, 2011, Mid America Clinical Laboratories (MACL) in Indianapolis, Indiana, issued a public statement acknowledging that it had reported false positive results for Chlamydia tests. The statement then described how the laboratory had notified the physicians and patients involved. MACL offered free retesting and other compensation for expenses incurred by patients as a result of the false positive tests.
MACL’s statement also described certain of the findings of its root cause analysis, along with steps it was taking to correct the problem and prevent similar problems in the future. The lab test errors involved a molecular test for Chlamydia that is based on Strand Displacement Amplification (SDA), a methodology cleared for market by the FDA. Television news coverage of the incident showed a BD Viper instrument in the laboratory and stated that MACL had reported false positive results for Chlamydia on eight patients. (See pages 9-10.)
As this issue of THE DARK REPORT went to press, Mid America Clinical Laboratories provided a detailed state- ment about this matter in response to our requests for information. MACL’s statement will be published in full in the May 23 issue of THE DARK REPORT.
Preventing Lab Errors
To provide other pathologists and lab administrators with useful insights about this laboratory error, THE DARK REPORT asked laboratory experts to review the first MACL public statement and the public news coverage of the episode. Their assessments brought out additional issues that were not recognized or discussed in MACL’s statement about the findings of its root cause analysis.
In Canada, Michael A. Noble, M.D., FRCPC, has been a leader in developing proficiency testing programs and helping laboratories improve quality. He is the Professor and Chair, Clinical Microbiology Proficiency Testing program and, Program Office for Laboratory Quality Management in the Department of Pathology and Laboratory Medicine, at the University of British Columbia, in Vancouver.
Noble went straight to the heart of the matter. He asserts that, involved in every case of laboratory error, there are systemic issues that often go unrecognized or unacknowledged. Noble thinks these systemic issues likely played some role in the false positive results acknowledged by MACL, but he wanted to emphasize that systemic problems in laboratory testing are an industrywide issue that doesn’t get adequate attention.
Noble identified four specific ways that systemic problems can contribute to false positive test results and other types of laboratory errors.
“The first systemic problem that comes into play involves the limitations and failures of the equipment and diagnostic analyzers used in today’s modern clinical laboratory,” stated Noble. “Lab directors know that lab equipment fails.
“Yet this information is not normally disclosed to physicians or to patients,” he observed, “despite the fact that such disclosure would be instructive for parties reviewing test results, particularly physicians and patients.
“Second, the errors at MACL likely could not have been prevented by accreditation or with proficiency testing,” declared Noble. “The punitive nature of proficiency testing does little to encourage labs to identify some of the problems that lead to false positive results.
“Third, there are systemic problems in healthcare, in medical malpractice, and in the clinical practice of medicine by physicians that likely contributed to the error rate MACL experienced,” added Noble. “The fourth systemic problem is in the lab itself, because, when errors occur, many laboratories do not respond quickly enough.”
Noble provided details about each of these systemic problems that exist today in laboratory medicine. “In my view, the number one systemic problem involves the equipment used in laboratories today,” he commented.
More Complexity in Labs
“As our laboratories continue to acquire and use more sophisticated diagnostic technologies and more automated testing, we add complexity,” he explained. “The nature of automation is that it increases the likelihood that our labs will have accidents, errors, and slips that go by—often unnoticed until after the fact.
“We need to recognize that every piece of machinery will fail,” declared Noble. “And they typically fail in a way that does not trigger an alarm. That is the norm.
“In Canada, pathologists have had problems with similar machines used by MACL for its molecular testing,” he stated. “In these situations, it was determined that the problem was a carry over problem within the sample dilution system. Carry over and contamination problems within automated analyzers is a recurrent theme. This is one type of recurring lab problem that accreditation and proficiency testing won’t stop.
“In addition, we know that the manufacturers of these analyzers can provide information on the false positive rate for each machine that they make,” he added. “In fact, most analyzers tend to have a false positive rate of about 1 in 1,000; some are higher and some are lower. Knowing the false positive rate within a reasonable level of confidence would allow a pathologist to add that information to each report, thereby providing context for each result.
“But we don’t do that,” Nobel continued. “We say the result our lab is reporting was positive or negative as if we have 100% confidence, which we don’t have. It’s dishonest not to share that information with our physician clients and with patients. We assume that physicians and patients understand this point. But they don’t.”
Noble’s second systemic problem involves the issues associated with accreditation and proficiency testing. “Reviewing the facts of the MACL case from the outside looking in, we don’t know all the details about what happened,” he said. “But my view is an error of this sort would not be prevented because of the lab’s accreditation and proficiency testing activities.
“This should be troubling to all laboratory professionals,” continued Noble. “The primary goals of accreditation and proficiency testing programs are to help laboratories drive out sources of errors and continually raise the accuracy and quality of their laboratory testing services.
“Yet, we can all recognize ways that these program requirements discourage full disclosure of problems,” he said. “In this regard, accreditation and proficiency testing play a systemic role in contributing to errors and inaccuracies in laboratory test results.
“The third systemic issue involves the litigious environment that exists here in Canada and in the United States,” he said. “One consequence of being litigious is that physicians do more testing than they would normally.
“Physicians practice defensive medicine because they don’t want to be identified as having missed something by failing to test,” he explained. “On the one hand, these litigious tendencies force overuse. But on the other hand, it also forces under-disclosure because the threat of litigation inhibits labs and manufacturers from admitting mistakes. The failure to admit mistakes creates additional problems because all labs lose the opportunity to learn from these mistakes.
“The overuse issue contributes to the systemic problem because you have large number of generally healthy people being tested on lab analyzers that were never designed for testing well people,” Noble continued. “Assume we have a piece of equipment with (let’s pick an arbitrary rate) a 1 in 1,000 chance of producing a false positive. If you test 1,000 patients each week, you have a probability of one person getting a false positive result every week.
“But then if you make it popular to do chlamydia testing and do one million patients per week, you will pick up 1,000 false positives every week,” Noble said. “Now you have 1,000 wives accusing their husbands of infidelity, and of that number, you could have 50 divorces! In part, the problem is that the lab testing errors don’t get picked up quickly enough. But the bigger problem is that most of those patients who were tested should not have been tested in the first place.
“But they get tested for two main reasons,” he said. “First, clinicians are afraid they will be sued if they don’t test. Second, it’s a financial ‘problem,’ because many labs have invested huge amounts of money to install this expensive equipment. The only way to recoup the cost of those machines is to test hundreds of thousands of well people who end up with negative results. In this regard, we have set up a system of lab testing that is destined to bite us.”
Noble then shifted to his fourth systemic issue, the slow reaction time of laboratories to problems in the process of lab testing. “This fourth systemic issue is that most labs don’t have a way to pick up false positives quickly when they occur,” he observed. “That means operations continue because most labs are blind to these problems.
“In the typical clinical lab, it is not until there are five, 10, 15, or 20 false positives that someone in the lab might say, ‘Maybe we have a problem,’” Noble said. “Conversely, a laboratory with a good qual- ity management system (QMS) tends to have a very active incident-reporting process in place. This helps the lab be much quicker at early detection of problems.
A Quality Lab Operation
“From what we know about MACL, there are indications that this is a quality lab,” Noble continued. “The fact that they got to eight results and recognized that they should investigate is a good thing. They didn’t wait until they got to 1,000 possibly inaccurate test results before they investigated.
“Still, the lab did report false positive results to those eight patients. This demonstrates a fact known by all pathologists and laboratory professionals: when the number is small, it is difficult to distinguish a positive result from a false positive result without re-testing,” explained Noble.
THE DARK REPORT observes that there are useful lessons to be learned from this case. Dr. Noble’s insights about the systemic issues that contribute to laboratory errors demonstrate how difficult it is for individual labs to eliminate all sources of errors.
Another Opinion about the Lab Error
UPON REVIEWING THE PUBLIC INFORMATION about the false positive results reported by Mid America Clinical Laboratories (MACL), one lab expert posed a basic question about how the test results were reviewed before they were released.
“The first flag for me that there would be a problem with this batch of tests is the number of positives that came up on this test run,” stated an experienced medical technologist (MT) who has extensive experience as a Lean, Six Sigma, and process improvement consultant. “MACL stated that ‘because the error was found quickly, it only affected 2% of the tests performed during this time period.’”
“I would want to know why the laboratory staff let all eight false positives go out along with the other positive Chlamydia test results reported that day,” continued this expert. “Most labs wouldn’t let that number of positives go out, because the history of sexually transmitted diseases is fairly steady. So if on one day, the lab had a larger than expected number of positives, it might be best to look for a reason before releasing all the positive test results.”
Questions about Proficiency Testing
COULD BETTER PROFICIENCY TESTING help medical laboratories prevent false positive test results? “Yes! Almost certainly!” asserts pathologist Michael A. Noble, M.D., who is active in laboratory proficiency testing programs at the University of British Columbia, in Vancouver, Canada.
“In North America, proficiency testing is required of clinical laboratories,” stated Noble. “But what prevents proficiency testing from making a greater contribution to improved quality in laboratory testing is its punitive nature.
“There is a huge consequence if the lab gets something wrong and fails the proficiency test,” he explained. “This gives the laboratory a strong incentive not to do proficiency testing in a straightforward manner, but to game the process as much as possible to get the correct result.
“If a lab can get the correct result by repeating the test, or by getting the correct result from someone else, then the lab will do so,” continued Noble. “The consequences of failing the proficiency test are so high that no lab can afford to fail, so we miss the opportunity to identify problems and correct them—then share those findings with other laboratories so they can eliminate similar problems in their laboratories.”