Comparison of Conventional Audiometry with a Game-Based Audiometric Application for Screening the Hearing Thresholds of Children and Adults

Background: This study aimed to compare the accuracy and the feasibility of a low-cost automated game-based hearing screening application (Aud⋅It) with conventional audiometry to screen the hearing thresholds of children and adults. Concurrent validity between the two audiometric tests was also explored. Methods: Aud⋅It was designed to use a three-alternative choice paradigm to test hearing thresholds and its transducer output accuracy was tested and checked extensively to ensure the conformance to the expected levels of insert earphones specified in ANSI/ASA S3.6: 2010 (2018). It is also equipped with a calibration function that can allow user calibration of other types of transducers (e.g., supraural headphones). A total of 131 children and adults aged between 3 and 82 were recruited for this study. Hearing thresholds were measured using Aud⋅It and conventional audiometry with the order of the tests counterbalanced and the differences between the thresholds were compared for each participant at corresponding frequencies and test ear. Tympanometry and otoacoustic emissions were used to check the middle ear and the outer hair cell functions. The Montreal Cognitive Assessment was used to screen the cognitive abilities of the adult participants. Results: Ninety-five percent of the thresholds obtained using Aud⋅It and conventional audiometry agreed within 5 dB, which is an acceptable clinical variability for audiometry testing. Pearson correlation coefficients indicated a strong and significant correlation between the two audiometric measures (r = 0.977), demonstrating excellent concurrent validity. Interclass correlation [ICC3,k; 0.986 (95% CI = 0.980–0.990)] and Cronbach’s alpha (0.988) values indicate the hearing thresholds obtained using Aud⋅It had excellent agreement with those obtained using conventional audiometry.. Most participants aged 3 years and the participants with lower cognitive ability either could not be tested using Aud⋅It or showed large threshold differences between the two audiometric tests, suggesting that they need to be tested using conventional audiometry by trained audiologists. The average test completion time of Aud⋅It was 7 minutes and 48 seconds. Conclusion: Aud⋅It is a viable hearing screening tool for testing children more than four years of age and adults without cognitive impairments.