ATLANTA - A reported boom in U.S. whooping cough cases is now being questioned after health officials discovered a regularly used lab test misdiagnosed cases in suspected outbreaks in New Hampshire, Massachusetts and Tennessee.
The false test results led thousands of people to take antibiotics unnecessarily and even caused a New Hampshire hospital to limit the number of patients admitted since hospital workers were thought to be infected.
The testing errors were reported Thursday by researchers with the Centers for Disease Control and Prevention.
Pertussis, or whooping cough, is a potentially fatal bacterial respiratory infection. Its name comes from the sound victims make as they try to recover their breath after a coughing fit.
Government health officials say cases have tripled in the United States since 2001, with nearly 26,000 cases reported in 2005. Nearly half of those cases were diagnosed with the testing method now called into question, and that has raised doubts about the true number of cases.
"Are we in fact seeing an increase?" asked Dr. Tom Clark, summarizing what some are wondering. Clark, a CDC epidemiologist, is co-author of the report on the misdiagnoses.
The most accurate diagnostic testing for whooping cough requires a week or more to grow the pertussis bacteria from a sample from a patient's nose or throat. Sometimes that's too long for health authorities to take action to prevent the disease from spreading.
Increasingly, doctors have depended on a faster, but less accurate test. Different labs do the tests differently, leading to uneven results, experts say.
Last October, the less accurate method was used to diagnose a 20-month-old child with whooping cough at Children's Hospital Boston. Three dozen specimens from hospital workers also tested positive as well. But those results were wrong, according to the more reliable bacteria culture test results, the CDC reported.
The same situation occurred in March 2006, when a lab worker at Dartmouth-Hitchcock Medical Center in Lebanon, N.H., was diagnosed with the illness. Nearly 1,000 hospital workers were tested, treated and furloughed to prevent infecting patients. Thousands were given antibiotics and vaccinations. The precautions affected staffing levels, and the hospital had to close off some beds. About 100 employees were diagnosed with pertussis using the speedy test, results later found to be wrong.
In April 2004, a 5-week-old infant in one Tennessee community, which CDC wouldn't identify, was diagnosed with whooping cough. Health officials began looking for the illness in other residents. Ultimately nearly 1,500 people were checked or offered antibiotics; 43 tested positive at first. But the more reliable test turned up negative results for all except the baby, the CDC said.
"It's been a roller coaster. Whoa, looks at this big outbreak! Whoa, it wasn't really pertussis!" said Dr. William Schaffner, chairman of Vanderbilt University's department of preventive medicine.
It's not clear why so many errors were detected in each incident, Clark said. Contamination of samples does not appear to be the explanation, he added.
The CDC is planning a study to improve and standardize the method of faster testing. In the meantime, people should still get recommended pertussis vaccinations, Clark said.
Clark said he believes there is a real increase in cases, and that many are going undiagnosed. Two states, Minnesota and Massachusetts, have beefed up their case-finding and testing and have both seen whooping cough increases, he said.
