| Summary: | The objective of this study was to assess the association of self-reported noise exposure and audiograms processed with ten algorithms proposed to quantify noise-induced hearing loss using receiver operating characteristic (ROC) curves. Participants were placed into groups based on self-reported noise exposure. Self-reported noise exposure served as a predictor for noise-induced hearing loss (NIHL). Audiograms were analysed with ten algorithms: The Guidelines, Brewster's Rules, two versions of military Noise-induced Hearing Loss, the Bulge Depth, the age-adjusted 8 kHz threshold and four versions of a new algorithm called the Adjusted Notch Depth (AND). The area under the ROC curves were calculated for each algorithm. Data were collected from three cycles of the National Health and Nutrition Examination Survey. Only one version the AND significantly identified those with self-reported noise exposure with an area under the curve of 0.562. The association between the AND and self-reported noise exposure was marginally better than the previous algorithms in identifying those with self-reported noise exposure. These findings do not support using puretone thresholds for identifying those with NIHL in a cross-sectional research study without stratifying the participants. More research is needed to determine how the AND can be applied to stratified designs.
|
|---|