VALIDATION OF PARAMETERS FOR ASSESSING OBSCURE AUDITORY DYSFUNCTION -ROBUSTNESS OF DETERMINANTS OF OAD STATUS ACROSS SAMPLES AND TEST METHODS

About 5% of those presenting at an ENT clinic with ear or hearing comp laints are found to have normal pure-tone thresholds. Many of this gro up complain specifically of difficulties understanding speech in backg round noise. We have termed this complaint 'Obscure Auditory Dysfuncti on' (OAD). Previous research on OAD (Saunders and Haggard, 1992) on a sample of 50 patients and 50 matched controls showed that 83% of the d eviance (binary variance) in membership of group (OAD/control) could b e explained by four variables, with three underlying factors. These fo ur variables are performance threshold in noise for speech; discrepanc y between this and a self-adjusted signal-to-noise ratio value (i.e. a measure of auditory confidence); dichotic listening performance; and performance on a psychoacoustic test, the threshold for detecting a to ne in a spectrally notched noise. This paper reports a study on a furt her 59 new OAD patients and 64 unmatched controls, using the tests in the form present in the test package, plus certain others. For routine practice, slightly differing implementations of the auditory test tec hniques were used. Compared to the earlier sample, the new sample had poorer measures of speech reception threshold in noise and they consid ered themselves more handicapped by their OAD state; however there was no significant difference between the two samples on the other perfor mance variables. Overall, the results demonstrate the robustness of th e conclusions reached from the previous study, as the test-variable de terminants of OAD status (i.e. case-control differences) were similar. The items on the OAD interview questionnaire which had shown signific ant differences between OADs and controls were also similar for the tw o OAD samples, giving a consistent picture of patient profile. These w ere: difficulties learning to read or write as a child (which relates to dichotic listening scores); adult ear problems; and tinnitus (which relate to the psychoacoustic scores). This necessary replication conf irms the validity of the three-factor model of OAD status and the util ity of the clinical test package based upon it.