WASHINGTON – A rapid viral testing panel decreased the odds of inappropriate antibiotic use by 67% in children hospitalized with an acute respiratory illness, allowing about 6% of the children taking them to discontinue them.
Use of such a test could be part of a facility’s overall antibiotic control program, Dr. Russell McCulloh said at the annual meeting of the Pediatric Academic Societies.
"Now we need to talk about when a test like this can be used appropriately," said Dr. McCulloh, an infectious disease specialist at Rhode Island Hospital, Providence. "There is always a strong possibility for a bacterial coinfection in these patients, and that can be very hard to determine because we don’t have a gold standard test for that. Adjunctive diagnostic testing will be necessary if you’re really going to close the loop and implement this as part of an antibiotic stewardship program."
Dr. McCulloh conducted a retrospective chart review of 1,731 children admitted to the hospital for an acute respiratory illness from 2009 to 2011. He then compared the rates of antimicrobial use in children who had the rapid viral panel (809) and those who did not (922). About half of the group (860) had received antibiotics before having the test run, while 255 had received oseltamivir.
Most of the children who had a positive viral panel were prescribed oseltamivir (87%); the drug was used in about 18% of those who had a negative test. A multivariate analysis found that a positive rapid viral panel was the single biggest predictor of getting the antiviral drug, increasing the odds by more than 27 times, Dr. McCulloh said.
Younger age also had a significant association, but the odds ratio was only 1.7. Having a significant past medical history was predictive in an unadjusted analysis, but not in the final model. Abnormal chest x-ray, intensive care unit admission, and abnormal white cell counts had no significant predictive value.
The test also influenced the use of antibiotics, Dr. McCulloh said. Antibiotics were given to 11% of children who had the viral testing, but to 100% of those who did not have it. Antibiotics were used in 51% of those with a positive test, and 67% of those with a negative test. Antibiotics also were discontinued in 6% of children who had a positive viral test.
A multivariate analysis found several factors significantly associated with antibiotic use. A significant past medical history increased the odds by 7%. Stronger predictors included a codiagnosis of acute otitis media (OR 6.0), an abnormal chest x-ray (OR 2.57), and a positive blood culture (OR 3). However, a negative viral panel was associated with a 69% decreased chance of receiving an antibiotic.
The test was a big shaper of physician behavior, Dr. McCulloh said, allowing a more judicious and targeted use of antibiotics while increasing the use of antiviral medication. "Rapid viral panel testing seems to enhance physician decision-making in these patients."
Dr. McCulloh had no financial disclosures.