Abstract:
Three hundred thirty-eight noise exposure samples were collected from 125 construction workers employed by a general contracting company in four construction trades: carpenters, laborers, ironworkers, and operating engineers. The samples were collected on 49 monitoring days over a 22-week period from four sites employing a variety of construction techniques. Each site was sampled at least twelve times on a randomly chosen date. Up to ten volunteer workers were sampled for an entire workshift on each sampling day using datalogging noise dosimeters, which recorded both daily Time-Weighted Averages (TWAs) and 1 minute averages. During the workday, workers completed a short questionnaire detailing the timing, number of tasks performed, and tools used throughout the day.
Regression models were developed to identify work characteristics associated with elevated exposure levels, allowing for identification of factors which can be altered to produce the greatest exposure reduction. Trade was a poor predictor of noise exposure; construction method, stage of construction, and the work tasks and tools used during the day were found to be better predictors of exposure. In addition, comparisons were made between exposures measured using the OSHA exposure metric, the 1996 NIOSH/ISO metric, and a modified NIOSH/ISO metric to examine the effects of differing exchange rates and instrument response times on exposures resulting from highly variable construction noise sources. The mean OSHA TWA for 338 samples was 82.8 dBA (6.8 dBA SD, range 61.6-99.3 dBA), while the mean NIOSH/ISO TWA for 174 samples was 89.7 dBA (6.0 dBA SD, range 76.1-103.9 dBA). Forty percent of OSHA TWAs exceeded 85 dBA, and 13% exceeded 90 dBA, the OSHA PEL. The tasks and tools associated with the highest exposure levels were those involving pneumatically-operated tools and heavy equipment. An internal validation sub-study indicated excellent agreement between worker-reported and researcher-documented tasks and tools.