Methods for early identification of noise-induced hearing loss.

An ideal test for identifying shifts in cochlear function would be highly repeatable and sensitive to minor damage. Three types of otoacoustic emission (OAE) test and pure-tone audiometry were evaluated for this purpose. They were compared in terms of test-retest repeatability within subjects and sensitivity to differences between subjects. The OAE measures were transiently evoked either conventionally (TEOAE) or using maximum length sequences (TEOAE-MLS), or continuously evoked as distortion products (DPOAEs). Several stimulus conditions were evaluated for each type. Thirty eight subjects with normal hearing or mild hearing losses were tested on all measures. Test-retest repeatability was rescaled according to the sensitivity of each measure to differences in hearing threshold level, thus allowing a direct comparison across methods. The most repeatable method thus defined was TEOAE-MLS which gave a rescaled standard deviation of 1.8 dB on replication. This was followed by TEOAE and DPOAE which gave rescaled standard deviations of 2.9 and 3.1 dB, respectively. All were more reliable than pure-tone audiometry which had a standard deviation of 4.9 dB. It is concluded that the various OAE measures have the potential to distinguish small changes in cochlear function from measurement uncertainty, and hence show promise for monitoring cochlear function in ears exposed to noise or other hazards.