Calculating annoyance: an option to proof efficacy in ENT treatment of snoring?

To develop and validate an objective method for calculating the annoyance caused by snoring sounds. 53 subjects assessed 50 different snoring and breath sounds on a visual analog scale for level of annoyance. A linear regression analysis was used to correlate these subjective assessments with objectively calculated psychoacoustic parameters (loudness, roughness, sharpness, and fluctuation strength, calculating the maximum, mean, and 5th percentile in each case). The quality of the resulting formula was checked, and additional validation was performed using subjective assessments of 60 new snoring and breath sounds by 52 new subjects. Receiver operating characteristic (ROC) analysis was used to scale the annoyance ranges obtained with the formula. A score consisting of the 5th percentile of loudness and the mean of roughness was developed. The formula displays high goodness of fit (R ² = 0.91) and quality. In the validation phase, a highly significant correlation (r _s = 0.95; p < 0.01) was obtained between the scores calculated with the formula and the subjective assessments. ROC analysis was able to define the annoyance ranges with a discriminatory power between 52 and 73 % (optimum sensitivity/specificity). The subjective assessments made by the participants were distinctly scaled and were reflected in a psychophysical algorithm. In the setting of polygraphy and polysomnography, this means that a reliable annoyance score that is not dependent on the bed partner can be obtained to establish the indication for anti-snoring treatments and to review their utility.