The use of the distraction test as a screen for hearing loss in infancy was reviewed in the Bristol and District Health Authority. Inaccuracies in data recording were found which challenged the screening and referral rates produced by the National Child Health Computer. Representative samples of children failing the distraction test and children needing hearing aids were followed up through the screening process. Of 130 children failing the test, only 46 (35%) had moderate conductive hearing loss and none had a severe loss. Of 34 children with sensorineural hearing loss (SNHL) requiring hearing aids, seven (21%) were identified by the test and five (15%) presented late after having been screened. The distraction test as now used in Bristol identifies large numbers of children with minor hearing loss. The referral of these children to the secondary services leads to delays for those children with more significant loss. The test is no longer the main method for identifying children with SNHL.