Page 5
The results were then submitted to statistical analysis for comparison between the control and the tinnitus groups. Because of the reduced size of the tinnitus group, and considering that the purpose of the study was to verify the influence of tinnitus on speech recognition in noise rather than to identify differences between the ears, it was opted to group the results from the right and left ears together for data analysis. It is worth mentioning that comparative tests were performed between the ears and no statistical relevance was found. The results from both ears were grouped together. The SRTN test data from both groups were compared by using Student’s t-test. In all statistical analyzes, a significance level of 5% was adopted for rejecting the null-hypothesis.
RESULTS
Thirteen individuals from the control group and seven individuals from the tinnitus group were evaluated, totaling 26 ears from the control group and 14 from the tinnitus group. The distribution of the study population according to gender and age is well known.
With respect to the psychoacoustic testing of the tinnitus group, i.e. the mean frequency and intensity of the tinnitus and the THI questionnaire results, it was found that, on average, tinnitus could be qualified as negligible (0 to 16%) and, at most, could be considered as having a mild impact (18% to 36%), by one participant .
In regard to the results from the LSP (Lists of Sentences in Portuguese) test, performed with the noise recorded on CD and speech-noise noise, the values refer to the signal/ noise ratio obtained, i.e. to the lowest ratio the individuals needed for recognizing 50% of sentences presented in the presence of noise.
It can be observed for both noises, i.e. both the noise on the CD and speech noise, the tinnitus group needed a higher signal/noise ratio in order to recognize 50% of presented sentences, with the difference being statistically significant between the groups when speech-noise noise was being investigated.