Developing a Framework for Industrial Noise Risk Management Based on Noise Kurtosis and Its Adjustment.

Objectives: Noise risk control or management based on noise level has been documented, but noise risk management based on a combination of noise level and noise's temporal structure is rarely reported. This study aimed to develop a framework for industrial noise risk management based on noise kurtosis (reflecting noise's temporal structure) and its adjustment for the noise level.

Design: A total of 2805 Chinese manufacturing workers were investigated using a cross-sectional survey. The noise exposure data of each subject included L EX,8h , cumulative noise exposure (CNE), kurtosis, and kurtosis-adjusted L EX,8h (L EX,8h -K). Noise-induced permanent threshold shifts were estimated at 3, 4, and 6 kHz frequencies (NIPTS 346 ) and 1, 2, 3, and 4 kHz frequencies (NIPTS 1234 ). The prevalence of high-frequency noise-induced hearing loss prevalence (HFNIHL%) and noise-induced hearing impairment (NIHI%) were determined. Risk 346 or Risk 1234 was predicted using the ISO 1999 or NIOSH 1998 model. A noise risk management framework based on kurtosis and its adjustment was developed.

Results: Kurtosis could identify the noise type; Kurtosis combining noise levels could identify the homogeneous noise exposure group (HNEG) among workers. Noise kurtosis was a risk factor of HFNIHL or NIHI with an adjusted odds ratio of 1.57 or 1.52 ( p < 0.01). At a similar CNE level, the NIPTS 346 , HFNIHL%, NIPTS 1234 , or NIHI% increased with increasing kurtosis. A nonlinear regression equation (expressed by logistic function) could rebuild a reliable dose-effect relationship between L EX,8h -K and NIPTS 346 at the 70 to 95 dB(A) noise level range. After the kurtosis adjustment, the median L EX,8h was increased by 5.45 dB(A); the predicted Risk 346 and Risk 1234 were increased by 11.2 and 9.5%, respectively; NIPTS 346 -K of complex noise at exposure level <80, 80 to 85, and 85 to 90 dB(A), determined from the nonlinear regression equation, was almost the same as the Gaussian noise. Risk management measures could be recommended based on the exposure risk rating or the kurtosis-adjusted action levels (e.g., the lower and upper action levels were 80 and 85 dB(A), respectively).

Conclusions: The kurtosis and its adjustment for noise levels can be used to develop an occupational health risk management framework for industrial noise. More human studies are needed to verify the risk management framework.