What can 35 years and over 700,000 measurements tell us about noise exposure in the mining industry?

Abstract

Objective: To analyse over 700,000 cross-sectional measurements from the Mine Safety and Health Administration (MHSA) and develop statistical models to predict noise exposure for a worker. Design: Descriptive statistics were used to summarise the data. Two linear regression models were used to predict noise exposure based on MSHA-permissible exposure limit (PEL) and action level (AL), respectively. Twofold cross validation was used to compare the exposure estimates from the models to actual measurement. The mean difference and t-statistic was calculated for each job title to determine whether the model predictions were significantly different from the actual data. Study sample: Measurements were acquired from MSHA through a Freedom of Information Act request. Results: From 1979 to 2014, noise exposure has decreased. Measurements taken before the implementation of MSHA’s revised noise regulation in 2000 were on average 4.5 dBA higher than after the law was implemented. Both models produced exposure predictions that were less than 1 dBA different than the holdout data. Conclusion: Overall noise levels in mines have been decreasing. However, this decrease has not been uniform across all mining sectors. The exposure predictions from the model will be useful to help predict hearing loss in workers in the mining industry.

Key Words:

• Demographics/epidemiology
• noise
• hearing conservation
• instrumentation

Acknowledgements

We would like to thank Stephanie Sayler, Rachel Long and Taichi Murata for their assistance in collecting data for this project.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

This work was supported by the National Institute for Occupational Safety and Health.

Grant # R21OH0 10482: Development of a US/Canadian Job Exposure Matrix (JEM) for Noise.