Clinical Question: How accurate are rapid influenza tests and what is their utility in managing febrile children?

Setting: Outpatient (any)

Study Design: Randomized controlled trial (nonblinded)

Synopsis: Children younger than five years who had fever or acute respiratory symptoms and presented to an emergency department or an acute care clinic were randomly assigned to usual care (n = 263) or to a point-of-care rapid assay for influenza (n = 205). All children had nasal and throat swabs for viral culture and polymerase chain reaction (PCR) assays for influenza. The PCR assays were performed by personnel blinded to the results of the rapid test.

After the study, the medical record was reviewed to evaluate additional test ordering and medication use. The medical record review was mentioned without description, so it is assumed that the review was done implicitly (which is not a good approach to medical record review). The rapid assay was 95 percent accurate, detecting 82 percent of patients with influenza and having 99 percent specificity. This translates to a positive likelihood ratio of 127 and a negative likelihood ratio of 0.18.

In the emergency department, but not in the acute care clinic, rapid testing was associated with fewer diagnostic tests (39 versus 51 percent). Rapid testing was not associated with reduced antibiotic use (approximately 30 percent) in either setting. Among children with influenza, the rate of antibiotic prescribing was approximately 16 percent in each group.

Bottom Line: Rapid influenza testing is highly accurate, but the results do not appear to influence care in a meaningful way other than to decrease testing in children treated in the emergency department. (Level of evidence: 1b)