Acquisition of Noise‐Induced Hearing Loss by Railway Workers

Objective: The rate of growth of hearing loss was examined in a population of workers exposed to noise. Of special interest was the course of loss in the first 10 yr of work, for which international standards provide little information. Based on the International Standards Organization (ISO) 1999 compilation, it was hypothesized that the proportional growth of hearing loss would be constant in log years. Design: Of 5000 workers compensated for hearing loss by a railroad, 500 were selected at random for analysis. For these workers, 2660 audiograms were available. The audiograms were distributed unsystematically over work careers, and for most workers were sparsely distributed. Thresholds were age corrected using ISO 1999 Annex B and averaged as a function of years on the job for two partitions, 0.5, 1, 2, and 3 kHz and 2 and 4 kHz. The average curves were compared with projections based on the ISO 1999 data as suggested by Dobie (1993a). A subset of workers with substantial hearing loss when hired was examined to evaluate the effect of pre‐existing hearing loss. Results: Hearing was worse at hire for this sample of workers than suggested by ISO 1999 Annex B for an unscreened sample of the general population at comparable ages. As a result, the early hearing loss was underpredicted by the ISO 1999 projection. Except at the earliest years of employment, the rate of growth toward the final hearing loss for the mean of the sample of workers was approximated by the projection from ISO 1999. The ISO 1999 projection is based on growth of the percentage of loss at the final audiogram obtained each year. In subgroups of workers, i.e., for the lower quartile and for the upper quartile of the sample, the projection based on ISO 1999 was similarly effective. Workers hired with prior loss showed growth of hearing loss that was flatter than that of either the mean of all workers or the ISO 1999 projection, but the growth curve for these workers approximated the average curve shifted to later years to bring the initial loss into correspondence with the predicted loss. Conclusions: For a group of workers exposed to a noisy environment, growth of hearing loss can be projected fairly well based on ISO 1999, without knowledge of the details of spectrum or level of noise exposure. This suggests that it would be possible to allocate the percentage of a known total hearing loss acquired in a group of workers on a year‐by‐year basis, given knowledge of work history for a claimant population in the absence of detailed knowledge of the exposure or detailed sequential audiometric data.