Test-Retest Reliability of Pure-Tone Thresholds from 0.5 to 16 kHz using Sennheiser HDA 200 and Etymotic Research ER-2 Earphones

In Brief Objective The purposes of the study were: (1) To evaluate the intrasession test-retest reliability of pure-tone thresholds measured in the 0.5–16 kHz frequency range for a group of otologically healthy subjects using Sennheiser HDA 200 circumaural and Etymotic Research ER-2 insert earphones and (2) to compare the data with existing criteria of significant threshold shifts related to ototoxicity and noise-induced hearing loss. Design Auditory thresholds in the frequency range from 0.5 to 6 kHz and in the extended high-frequency range from 8 to 16 kHz were measured in one ear of 138 otologically healthy subjects (77 women, 61 men; mean age, 24.4 yr; range, 12–51 yr) using HDA 200 and ER-2 earphones. For each subject, measurements of thresholds were obtained twice for both transducers during the same test session. For analysis, the extended high-frequency range from 8 to 16 kHz was subdivided into 8 to 12.5 and 14 to 16 kHz ranges. Data for each frequency and frequency range were analyzed separately. Results There were no significant differences in repeatability for the two transducer types for all frequency ranges. The intrasession variability increased slightly, but significantly, as frequency increased with the greatest amount of variability in the 14 to 16 kHz range. Analyzing each individual frequency, variability was increased particularly at 16 kHz. At each individual frequency and for both transducer types, intrasession test-retest repeatability from 0.5 to 6 kHz and 8 to 16 kHz was within 10 dB for >99% and >94% of measurements, respectively. The results indicated a false-positive rate of <3% in reference to the American Speech-Language Hearing Association (1994) criteria for cochleotoxicity for both transducer types. In reference to the Occupational Safety and Health Administration Standard Threshold Shift criteria for noise-induced hazards, the results showed a minor false-positive rate of <1% for the HDA 200. Conclusion Repeatability was similar for both transducer types. Intrasession test-retest repeatability from 0.5 to 12.5 kHz at each individual frequency including the frequency range susceptible to noise-induced hearing loss was excellent for both transducers. Repeatability was slightly, but significantly poorer in the frequency range from 14 to 16 kHz compared with the frequency ranges from 0.5 to 6 or 8 to 12.5 kHz. Measurements in the extended high-frequency range from 8 to 14 kHz, but not up to 16 kHz, may be recommended for monitoring purposes. Little is known about the repeatability of conventional and extended high-frequency audiometry using modern transducers, even though monitoring is critically dependent upon their usage. This study assessed intra-session test-retest reliability of pure-tone thresholds in the 0.5-16 kHz frequency range for a group of otologically healthy subjects using Sennheiser HDA 200 circumaural and Etymotic Research ER-2 insert earphones. Although the repeatability of auditory thresholds within the 0.5 to 12.5 kHz range was excellent for both transducers, it was significantly reduced in the frequency range from 14 to 16 kHz. For both transducers, the results indicated a false-positive rate of <3% in reference to the ASHA (1994) criteria for cochleotoxicity and an even smaller false-positive rate of <1% in reference to the OSHA Standard Threshold Shift criteria for noise-induced hazards. These data suggest that both transducer types perform equally well and can be used for conventional and extended high frequency audiometry.