Two tests for assessing under- and over-amplification with hearing aids

Under- or over-amplification of sound is a common problem in hearing aid (HA) fitting. This submission describes the implementation of two variations of the Hearing In Noise Test (HINT) for assessing under- and over-amplification with HAs. Results are presented of experiments carried out with HA users to determine the respective test-retest reliabilities. In this study, thirty adult HA users (18 males, 12 females, age 49-84 years, mean age 70.97 years and SD of 9.15 years) were tested with a test battery consisting of the HINT, a variation targeting the lower threshold of audibility (HINT-L) and a variation targeting the limit of loudness discomfort. The participants were tested twice in order to perform the reliability analysis. The results showed that there was a significant difference between test and retest for all three variations of HINT. However, a calculation of Cohen’s d showed that the effect size for the differences could be considered as a very small effect at most. A calculation of Pearson correlation coefficients showed that the three variants of HINT had a significant and strong correlation between test and retest, with r(29) = 0.95, p < 0.01 for the HINT, r(29) = 0.98, p < 0.01 for the HINT-L, and r(29) = 0.94, p < 0.01 for the HINT-U. The within-subject standard deviation were determined to be 0.77 dB for
HINT, 0.91 dB for HINT-L, and 2.15 dB for HINT-U. The overall conclusion of the experiment is that the variations of the HINT for assessing under- and over-amplification with HAs has a high test-retest reliability, and that the test-retest reliabilities are comparable to the traditional version of HINT.