Harvard Medical School and Boston Children's Hospital Informatics program professor Dr. Kenneth Mandl, left, and Harvard Medical School and Children's Hospital assistant professor Dr. Andrew Fine, stand in an examination room at Children's Hospital with an iPad used by patients to gather symptom scores while they wait in the emergency room for treatment in Boston, Monday, Nov. 4, 2013. (AP Photo/Stephan Savoia)
Tuesday, November 05, 2013 7:41 am
Strep scorecard might help tell if you need a doc
By LAURAN NEERGAARDAP Medical Writer
More than 12 million people make doctors' visits for a sore throat every year. Usually the culprit is a virus that they just have to wait out with a little TLC.
In fact, the risk of strep throat is low enough for adults that doctors may skip testing them, deciding not to bother after running down a list of symptoms. That can leave patients wondering why they spent hours in the waiting room and had to pay the doctor's bill.
"If you could know that your risk was low enough that you wouldn't even be tested, you might actually save yourself a visit," said Dr. Andrew Fine, an emergency physician at Boston Children's Hospital.
The trick: Combine some of the symptoms that doctors look for with a bit of computer data to tell if strep throat is circulating in your geographical region. If the bug's in your neighborhood, that increases the chances that you've caught it, said Dr. Kenneth Mandl, a Harvard professor and informatics specialist with Boston Children's.
As a first step, Fine and Mandl turned to the records of more than 70,000 sore-throat patients who got strep tests and had their symptoms recorded at CVS MinuteClinics in six states between 2006 and 2008. They determined those people's risk of strep using the experimental scorecard approach and checked the computer model's accuracy against the strep test results.
Nationally, identifying those with less than a 10 percent chance of strep throat could save 230,000 doctor visits a year, the team reported Monday in the journal Annals of Internal Medicine.
The method wasn't perfect: It meant 8,500 strep cases would have been missed, or the diagnosis delayed, concluded the government-funded study.
But Mandl said it's unlikely that would lead to lasting harm as most of those infections would clear up on their own, or persisting pain eventually would send patients to the doctor. And he noted that the rapid strep tests that doctors use in their offices can miss cases, too.
Much more research is needed to prove if the method would work in everyday life and if a mobile app or a phone call to the doctor would be the best approach. The Boston team has begun the next step: Parents of kids who come to the hospital's emergency room for a strep test are handed a digital tablet and asked to fill out the scorecard first. Researchers will see how the combination of symptoms and local infection trends compare with actual strep test results.
Sore throats are a challenge. Strep throat, caused by bacteria named Group A streptococcus, is to blame for only about 10 percent of cases in adults, and 30 percent in children.
It's hard to tell who needs a strep test based on symptoms alone, cautioned Dr. Chris Van Beneden of the Centers for Disease Control and Prevention, which helped fund the new research. But what is clear: Doctors should be sure it's strep before prescribing antibiotics because those bacteria-fighting drugs have no effect on viruses.
Yet research published last month in the journal JAMA Internal Medicine found 60 percent of adults who sought care for a sore throat received antibiotics. Unneeded antibiotic use can spur development of drug-resistant germs.
The Boston team looked at the flip side of the issue: Who could safely skip a strep check? Because strep is most common in children ages 5 to 15, doctors usually test youngsters with a sore throat for the bacteria.
For anyone 15 or older, Mandl said doctors may skip a test depending on symptoms. While a cough and runny nose are more typical of a cold virus, strep symptoms might include a fever, enlarged lymph nodes, tonsils with swelling or pus and lack of a cough.
So Fine and Mandl focused first on the over-15 crowd. Because feeling lymph nodes and peeking at tonsils could be difficult for the average layman, their scorecard posed easy questions: Is there a fever? Is there a cough?
Then came the key: The scorecard automatically merged those symptoms with local trends in strep diagnosis.
It's a practice called biosurveillance. Already, hundreds of hospitals, clinics and health departments automatically report certain symptoms and diagnoses to the government. That lets officials track the spread of flu every year, for example - and some web sites now show flu activity by zip code so people can check if influenza has reached their community.
Likewise, results of strep tests are available digitally from testing laboratories, clinics, even large doctors' offices, Mandl said. They just have to be collected and used, which isn't routine.
In an accompanying editorial, Dr. Robert Centor of the University of Alabama, Birmingham, said too many clinics and emergency rooms still give a strep test to every sore throat patient. But he questioned if the home scorecard approach would make a difference, wondering if biosurveillance would be too costly or if average patients even would use it before seeking care.