Respiratory secretions for viral diagnosis are often collected with nasopharyngeal (NP) swabs, although many laboratories recommend NP aspirates or washings. We compared results using NP washings and NP swabs in three diagnostic RSV tests, a rapid RSV EIA antigen test (Abbott Laboratories), an indirect fluorescent antibody test (FAT) with rabbit antiserum, and virus culture (HEp-2 cells). Paired samples were collected from 121 children with suspected RSV bronchiolitis or pneumonia. A minitip swap was passed into the nasopharynx for 10 sec, rotated and withdrawn. The opposite nares was irrigated with approximately 1 ml of saline and aspirated using a syringe and plastic feeding tube. Fifty-one children (42%) grew RSV in culture, 49 from NP washings versus 27 from NP swabs (p less than 0.001). Fifty-three (44%) were positive by FAT, 52 from NP washings versus 12 from NP swabs (p less than 0.001). Fifty-eight children (48%) had positive RSV EIA tests, 57 from NP washings versus 35 from NP swabs (p less than 0.001). Detection by EIA was more sensitive than culture regardless of the method of specimen collection. We conclude that NP washings are superior to NP swabs for RSV culture and rapid diagnosis by EIA or FAT.