Validity of automated threshold audiometry in school aged children

Abstract

BackgroundAutomated hearing tests have the potential to reduce the burden of disease amongst learners by introducing such services within the school context. MethodsThe aim of the study was to conduct a validation study on normal and hearing impaired learners, comparing air and bone conduction automated test results to conventional test results in 50 school aged learners (n = 100 ears) within a noise controlled school environment using a cross sectional comparative study design. The KuduWave 5000 (Emoyo.net) was used in this study. Data analysisThe spearman's correlation coefficient was calculated to determine test-retest reliability. The mean and standard deviation (SD) was measured for each frequency. The absolute mean difference (AMD) and SD was calculated for both air and bone conduction testing at each frequency for automated testing. A paired sampled t-test and a one way ANOVA was used to identify any significant differences. Alpha was set as 0.05. ResultsThere was significant correlation between thresholds obtained for automated test one and test two for normal hearing and hearing impaired group. The spearmans correlation coefficient was high (close to +1) for majority of the results for both groups across the frequency range. Both air and bone conduction testing across the frequency range of 250 Hz–2000 Hz and at 8000 Hz were not statistically signfiicant (p < .005) for both groups, however at 4000 Hz for bone conduction testing in the hearing impaired group, there was a statisticially signficant difference (p = .003). This was attirubted to the variaibilty in bone conduction test results often due to force and placement of the bone vibrator. ConclusionThe findings indicate that automated audiometry can yield relaible results that are comparable to conventional test results. Key clinical considerations include extending the response time, regular rest periods, improving instructions and comfort levels.