BACKGROUND: Extended high frequency audiometry is able to detect early cochlea damage. So it is useful in monitoring ototoxic, senile and noise induced hearing loss.OBJECTIVE: To study with extended high frequency audiometry the auditory function of people in highland of qinghai province without noise pollution and evaluate the sensitivity of this method in diagnosing early noise induced hearing loss. To provide the epidemiological data for early control of noise induced hearing loss in this area.DESIGN:It is a transverse comparison among normal people in two areas.SETTING: This study was cooperated by the Otolaryngeal Department of Gulou Hospital affiliated to Nanjing Medical College, the Otolaryngeal Department of Xijing Hospital of the Fourth Medical College and Auditory Research Center of Giessen University, Germany.PARTICIPANTS: Totally 300 people lived in highland of qinghai province received hearing test including extended high frequency audiometry. At the same time another 300 people were chosen in Nanjing city as control group.METHODS: The subjects in the two groups received the whole set of hearing tests including extended high frequency audiometry and answered a questionnaire about their personal history. The mean hearing thresholds from 0. 25-16 kHz between the two groups were compared with t-test.MAIN OUTCOME MEASURES: The difference of mean hearing thresholds from 0. 25-16 kHz between the two groups.RESULTS: There was no significant difference between hearing threshold of low frequencies and speech frequencies. The hearing threshold of high frequencies rose with the stepping-up of testing frequencies which made the possibility of measuring the hearing threshold reduced. But the hearing threshold above 6 kHz of the people in area without noise pollution was significantly higher than that of the control( P ＜ 0.05) . No difference between children and adolescence was found in the 2 groups. The epidemiological data revealed that the people without noise pollution had unperceptible hearing damage compared with the control group.CONCLUSION: The damage of impulsive noise to hearing began at 20 kHz and developed from 20 kHz to 10 kHz. If it was not protected, it would be extended to the speech frequencies and the hearing loss would get worse. So the extended high frequency audiometry was thought helpful to screen noise induced hearing loss early.