Changes over time in audiometric thresholds in a group of automobile stamping and assembly workers with a hearing conservation program.

Noise induced hearing loss (NIHL) is among the 10 leading occupational diseases, afflicting between 7.4 and 10.2 million people who work in noise above 85 dBA. Although mandatory hearing conservation programs (HCPs) have been in effect since 1972, this problem persists, as hearing protectors are not consistently used by workers, or may not attenuate to manufacturer's estimates in real world conditions. In this study, information from noise and hearing protection use measurements taken at an automobile assembly plant were used to construct average lifetime noise exposure and hearing protection compliance estimates for use in modeling to predict both total hearing loss and onset of two accepted definitions of hearing loss. There were 301 males and females in this cohort; their mean age was 42.6 (7.2) years, and mean tenure was 14.3 (3.5) years. Average length of follow-up was 14.0 years. There were 16 members of this cohort who had hearing loss at the speech frequencies (defined as an average hearing level > or = 25 dB at 500, 1000, and 2000 Hz). In cross-sectional multivariate analyses, years of employment, male gender, and proportion of time wearing hearing protection were the factors most associated with hearing loss at the average of 2000, 3000, and 4000 Hz (p < 0.0001) controlling for age, transfer status (as a surrogate for previous noise exposure), race, and lifetime average noise exposure. The most consistent predictor of hearing loss in both univariate and multivariate analyses was percentage of time having used hearing protection during the workers' tenure.