Clinical pure-tone audiometry is conducted using stimuli delivered through supra-aural headphones or insert earphones. The stimuli are calibrated in an acoustic (average ear) coupler. Deviations in individual-ear acoustics from the coupler acoustics affect test validity, and variations in probe insertion and headphone placement affect both test validity and test–retest reliability. Using an insert earphone designed for otoacoustic emission testing, which contains a microphone and loudspeaker, an individualized in-the-ear calibration can be calculated from the ear-canal sound pressure measured at the microphone. However, the total sound pressure level (SPL) measured at the microphone may be affected by standing-wave nulls at higher frequencies, producing errors in stimulus level of up to 20 dB. An alternative is to calibrate using the forward pressure level (FPL) component, which is derived from the total SPL using a wideband acoustic immittance measurement, and represents the pressure wave incident on the eardrum. The objective of this study is to establish test–retest reliability for FPL calibration of pure-tone audiometry stimuli, compared with in-the-ear and coupler sound pressure calibrations.
The authors compared standard audiometry using a modern clinical audiometer with TDH-39P supra-aural headphones calibrated in a coupler to a prototype audiometer with an ER10C earphone calibrated three ways: (1) in-the-ear using the total SPL at the microphone, (2) in-the-ear using the FPL at the microphone, and (3) in a coupler (all three are derived from the same measurement). The test procedure was similar to that commonly used in hearing-conservation programs, using pulsed-tone test frequencies at 0.5, 1, 2, 3, 4, 6, and 8 kHz, and an automated modified Hughson-Westlake audiometric procedure. Fifteen adult human participants with normal to mildly-impaired hearing were selected, and one ear from each was tested. Participants completed 10 audiograms on each system, with test-order randomly varied and with headphones and earphones refitted by the tester between tests.
Fourteen of 15 ears had standing-wave nulls present between 4 and 8 kHz. The mean intrasubject SD at 6 and 8 kHz was lowest for the FPL calibration, and was comparable with the low-frequency reliability across calibration methods. This decrease in variability translates to statistically-derived significant threshold shift criteria indicating that 15 dB shifts in hearing can be reliably detected at 6 and 8 kHz using FPL-calibrated ER10C earphones, compared with 20 to 25 dB shifts using standard TDH-39P headphones with a coupler calibration.
These results indicate that reliability is better with insert earphones, especially with in-the-ear FPL calibration, compared with a standard clinical audiometer with supra-aural headphones. However, in-the-ear SPL calibration should not be used due to its sensitivity to standing waves. The improvement in reliability is clinically meaningful, potentially allowing hearing-conservation programs to more confidently determine significant threshold shifts at 6 kHz—a key frequency for the early detection of noise-induced hearing loss.
1Mimosa Acoustics, Champaign, Illinois, USA; 2Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA; and 3Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois, USA.
Acknowledgments: The authors thank Pat Jeng for advice on experimental design; Lynne Marshall for advice on the experimental protocols and for reviewing an earlier version of the manuscript; Bill Ahroon (U.S. Army Aeromedical Research Laboratory) for the loan of an Army-issue audiometer; Laurie Heller for statistical advice; Rob Withnell for sharing data; and to Kurt Yankaskas, Program Officer for Noise Induced Hearing Loss at the Office of Naval Research, for his support.
This article was supported by small business innovation research awards to Mimosa Acoustics from the Office of the Secretary of Defense under the contract number N00014-15-C-0046 and the Defense Health Program under the contract number W81XWH-16-C-0185. Portions of this article were presented at the 43rd Annual AAS Scientific and Technology Conference of the American Auditory Society, Scottsdale, AZ. The content of this report is solely the responsibility of the authors and does not necessarily represent the official views of the Department of Defense or the US Government.
J.L.M. and C.M.R. designed the experiment. C.M.R. and Z.D.P. performed the experiment at the Research Laboratory of Electronics at Massachusetts Institute of Technology. J.L.M. analyzed the data. J.L.M. and S.R.R. wrote the article.
The authors have no conflicts of interest to disclose.
Address for correspondence: Judi A. Lapsley Miller, Mimosa Acoustics, 335 Fremont Street, Champaign, IL 61820, USA. E-mail: firstname.lastname@example.org
Received January 5, 2017; accepted December 21, 2017.