CEO SUMMARY: In reporting on a suspected outbreak of whooping cough (Bordetella pertussis), a national newspaper suggested administrators at Dartmouth Hitchcock Medical Center had placed too much faith in molecular testing. In this exclusive interview, Dartmouth’s Director of Molecular Pathology comments on how the media misunderstood the story, and why molecular testing remains the preferred method.
MARK TWAIN ONCE SAID, “All in all, the newspapers do a wonderful job, except when I know something about the subject.” The great American writer’s quote is particularly telling given the circumstances surrounding the reporting of an outbreak of Bordetella pertussis at the Dartmouth Hitchcock Medical Center in Lebanon, New Hampshire, last year.
The common name for pertussis is whooping cough and it is difficult to diagnose in children and even more difficult to diagnose in adults. So, when healthcare workers at the center began coughing last year, administrators were concerned there was an epidemic of whooping cough.
Working with epidemiologists from the federal Centers for Disease Control and Prevention (CDC), Dartmouth’s pathology department ran molecular tests. The results were typical: Some patients tested positive for pertussis but the results on many more patients were equivocal.
After The New York Times published an article questioning the validity of molecular testing, the pathology department found itself in the middle of a controversy that was needlessly started. The issue shows how little healthcare experts know about pertussis and how to consistently identify it clinically. (See TDR, Feb. 19, 2007.) As a result of the questions raised in the New York Times article, Dartmouth’s lab is now facing inquiries from several federal agencies.
Sensitivity And Specificity
“There was a lot of misinformation out there,” explained Gregory J. Tsongalis, Ph.D., Director, Molecular Pathology for the Department of Pathology at the Dartmouth Medical School and Dartmouth Hitchcock Medical Center. “What didn’t come across was that only about 40 patients tested positive using the molecular test (with polymerase chain reaction or PCR), meaning less than half were true positives. This was in April 2006. The other 60% were in the equivocal zone of the assay and were called positive as a result of the clinical symptoms they had.
“During the outbreak, we thought it best to treat these people rather than wait for the results of a second PCR test or culture to come back to confirm the positive results a few days later,” he noted. “They were given an antibiotic and put on furlough from work. That was the decision and it was the correct one.
“At the time, some people considered it to be a major outbreak rather than just 40 people out of 1,300 who work here and were tested,” Tsongalis said. “That’s a small outbreak because there were 1,200 people who were true negatives and came back to work the next day. But then, in January of this year, The New York Times did an article that made it seem that the PCR-based test was unreliable. As a result, not only did our lab got a bad rap, but molecular testing in general was questioned as well.” The article, “Faith in Quick Test Leads to Epidemic That Wasn’t,” was published by The New York Times on January 22, 2007.
“During the outbreak, federal officials advised us to confirm the positive results by culture, but culture is sensitive in only 30% to 40% of cases,” he said. “If it is present and it grows, you can identify it. But it only grows 30% or 40% of the time.
“In addition, we were criticized because both a federal and an academic lab did serology testing using an assay that was still in development,” added Tsongalis. “Only one lab in the country has a clinically validated serology assay for pertussis. All other labs are still developing their serologic assays, and serologic assays may be even more difficult to interpret than a molecular PCR test. Using a non-developed research-based assay to confirm an outbreak is a big mistake.
“Typically, if we weren’t in an outbreak situation, we would repeat those tests the next day and report results at that time,” he added. “If they came up positive again we would call them positive. If they came up negative, we would call them negative. But we didn’t do that because an institutional decision was made to get people back to work quickly or send them home quickly. That’s the correct way an organization must respond when there is an outbreak like this. It minimizes the risk of spreading the infection.
“The big question involves all the people who were in the equivocal range and were treated,” he continued. “Did they have low level copies of pertussis that were accounting for their respiratory symptoms or did they not have it at all? That’s the million-dollar question because, in adults, pertussis looks similar to a regular cold.
“Most lab professionals around the country are comfortable using molecular tests and believe in the need, as appropriate, to do a second test with a different marker or a different target. But I am not aware of any laboratory moving quickly to change its standard of care in this regard. What I don’t want to see happen is to have anyone recommend that we discontinue using this molecular test.
“Pathologists understand the issues involved with molecular testing, but not everyone in healthcare fully appreciates what it means when results are equivocal,” Tsongalis explained. “There are nuances with molecular testing, as there are with any diagnostic test. And, there is no gold standard for identifying pertussis. So any attempt to compare one test method against another is like comparing apples and bananas. Each one has its own nuances and its own level of sensitivity and specificity.
Standard Of Practice
“Right after The New York Times article came out, I surveyed 30 labs across the country and found that each was doing the same targeted molecular test that we perform here,” Tsongalis said. “And they are not confirming by culture and are not doing serology. In other words, molecular testing has become the standard of practice. What we now see in testing for pertussis are some of the variances in using the test as a screening tool in an adult population for which we would not test routinely.
“The New York Times article made it sound as if molecular testing was not worth doing,” he added. “That is certainly not the case. We do so much molecular testing now that it would be like taking two giant steps backward to go back to some earlier testing methods that we formerly used in these clinical situations.
“The lesson in all of this is that we don’t fully understand pertussis infections in adults,” Tsongalis said. “With children it’s different. They almost always develop the classic symptom, the ‘whooping’ sound with the cough.”
When the Media Fails to Understand How Diagnostic Technology Works in the Real World
IN ARTICLE IN THE NEW YORK TIMES on January 22 questioned the validity of molecular testing. The article, “Faith in Quick Test Leads to Epidemic That Wasn’t,” said almost 1,000 workers at the Dartmouth Hitchcock Medical Center were given a preliminary molecular test. It also said 142 people at the center were told they appeared to have the disease.
Later, the article said, “Not a single case of whooping cough was confirmed with the definitive test, growing the bacterium, Bordetella pertussis, in the laboratory. Instead, it appears the health care workers probably were afflicted with ordinary respiratory diseases like the common cold.”
Still later, the article questioned the value of molecular tests. It said, “Many of the new molecular tests are quick but technically demanding, and each laboratory may do them in its own way. These tests, called ‘home brews,’ are not commercially available, and there are no good estimates of their error rates. But their very sensitivity makes false positives likely, and when hundreds or thousands of people are tested, as occurred at Dartmouth Hitchcock Medical Center, false positives can make it seem like there is an epidemic.”
Compounding the error, ECRI, a nonprofit health services research agency in Plymouth Meeting, Pennsylvania (at www.ecri.org), issued a Health Devices Alerts Special Report on February 2. The alert was based on the Times article. ECRI said 142 individuals were misdiagnosed with whooping cough based on false-positive molecular assay test results. On February 23, ECRI retracted its earlier statements when it issued another alert.
In the second alert, ECRI said, “According to Dartmouth-Hitchcock, only 41 individuals tested positive by molecular methods that have been widely used for over a decade for diagnosing pertussis. The remaining individuals were considered suspect pertussis cases based on equivocal molecular test results, pertussis symptoms, and concern for patient and employee health. Those with equivocal results were treated as if they had positive results, in accordance with standard protocols.”