The test characteristics of rapid tests for respiratory syncytial virus (RSV) in infants may differ from older children secondary to a lower likelihood of previous illness with RSV. Our main goal was to establish the test characteristics of the RSV Abbott Testpack (TP) enzyme-linked immunoabsorbent assay (EIA) in febrile infants < or = 60 days of age. Our secondary goal was to determine the likelihood of RSV given a particular clinical syndrome and a negative or positive EIA. A prospective sample of infants with a temperature > or = 38.0 degrees C was evaluated during 2 successive RSV seasons. Conventional tissue and shell vial viral cultures were utilized as the reference standard. The RSV Abbott Testpack EIA had a sensitivity of 75% (95% CI 60-90%), a specificity of 98% (95% CI 96-100%), a positive predictive value of 89% (95% CI 77-100%), a negative predictive value of 95% (95% CI 91-98%), a likelihood ratio for a positive test of 35.5 (95% CI 11.4-110.7), and a likelihood ratio for a negative test of 0.26 (95% CI 0.14-0.47). Even with a negative EIA, patients with lower and upper respiratory tract illness still had a 22.3% and 5.5% chance of harboring RSV, respectively. The RSV Abbott Testpack is a useful diagnostic tool in the detection of RSV in febrile infants but has limitations. During months typically associated with RSV disease, a positive RSV TP indicates a high likelihood of illness, but clinicians should be wary of false negatives.