Accuracy of an internet-based speech-in-noise hearing screening test for high-frequency hearing loss: incorporating automatic conditional rescreening

Purpose: To validate the accuracy of an internet-based speech-in-noise hearing screening test for high-frequency hearing loss (HFHL) ‘Occupational Earcheck (OEC)’ incorporating an automatic conditional rescreening, in an occupationally noise-exposed population. Secondary objectives were to assess the effects of age on test accuracy measures, and to assess the test accuracy for different degrees of HFHL. Methods: A study was conducted on cross-sectional data of occupational audiometric examinations, including the index test OEC and reference standard pure-tone air conduction audiometry, of 80 noise-exposed workers. Sensitivity, specificity, and likelihood ratios were calculated for the OEC, after automatic conditional rescreening, for a younger and an older age group, and for two degrees of HFHL (HFHL25: PTA3,4,6 ≥ 25 dB HL, and HFHL35: PTA3,4,6 ≥ 35 dB HL, both for at least one ear). Results: Test specificity for HFHL25 after a single test was 63%, and improved to 93% after the automatic conditional rescreen. Test sensitivity for HFHL25 decreased from 65% to 59%. Test sensitivity and specificity including automatic conditional rescreening for HFHL35 was 94% and 90%, respectively. The positive likelihood ratio for HFHL25 was 8.4, and for HFHL35 9.4. The negative likelihood ratio for HFHL35 was below 0.1. Conclusions: The OEC is an appropriate screening test, especially for HFHL35. Normal-hearing workers who obtained a positive test result for the first test for one or two ears, benefit from having an automatic rescreen, resulting in an improvement of the test specificity, and hence prevent unnecessary referral.