There are not enough documented cases of permanent hearing loss in which both hearing and evoked otoacoustic emissions have been followed in individuals over time to allow generalizations of how noise exposure affects emissions relative to hearing thresholds. Early results from longitudinal studies show some interesting case studies in which the emissions confirm and extend audiometric results. To collect valid, reliable, long-term data on noise-exposed people, a number of factors must be considered. These factors include (1) choice of stimulus;(2) measurement of the emission, including analyzing bandwidth, absolute or relative emission level, unmeasurable emissions, and noise floor issues;(3) probe fit and ensuring a good stimulus waveform and spectrum; and (4) middle-ear pressure. Emissions can be used in a clinical setting if the test-retest reliability is established for the protocol, equipment, testers, and clinical population of interest. This allows identification of abnormal emission shifts, but interpretation and implications at this stage are limited.