Abstract
In order to establish the testing procedure of the loudness of tinnitus, loudness estimates of tinnitus using balance method in different test conditions were compared each other, and loudness of the external continuous sounds simulating tinnitus were measured by the same procedure as the measurement of tinnitus. In cases having sensorineural hearing loss at the tinnitus frequency, the loudness recruitment is the largest factor that modifies the result. This difficulty can be avoided by using other frequencies as a test tone at which bone conduction hearing is normal. Even in non-recruiting ears, test-retest variability in the successive balance test was minimal (0.5 dB), while this variability becomes larger (3.5 dB) when the second measurements were performed after a certain period for the patients who reported no change of the loudness of their tinnitus. Loudness estimates of the simulating sounds were significantly larger by using delayed balance method than by using simultanuous balance method. This result suggests the possibility that the measurement of tinnitus by balance test gives us lower estimate than real loudness. Subjective assessment of the change of the tinnitus loudness correlated better with the before-to-after ratio of the sensation level of tinnitus in dB than in sone scale which is the countable measure.
