Simultaneous assessment of occupational exposures from multiple worker groups.

Area of Science:

• Occupational hygiene and environmental statistics
• Exposure assessment and risk management
• Statistical modeling for public health

Background:

• Existing methods assess workplace exposures on a group-by-group basis.
• Simultaneous assessment of multiple worker groups within an industry is needed.
• Log-normal distribution and mixed-effects regression models are commonly used for exposure data.

Purpose of the Study:

• To extend existing methods for simultaneous assessment of data from multiple worker groups.
• To develop a procedure for testing worker overexposure, conditional on variance component homogeneity.
• To provide recommendations for study design and intervention strategies.

Main Methods:

• Extended Rappaport et al. and Lyles et al. methods for simultaneous multi-group analysis.
• Utilized mixed-effects regression models assuming log-normally distributed exposures.
• Employed a likelihood ratio test to assess homogeneity of within-worker variance components and a conditional procedure for overexposure testing.

Main Results:

• The extended methods successfully fit models to data from multiple groups, even with limited data per group.
• The overexposure testing procedure yielded consistent conclusions under both homogeneous and heterogeneous within-worker variance assumptions.
• Simulation studies indicated that the testing procedure maintains adequate significance levels when using a preliminary likelihood ratio test with a significance level of at least 0.01.

Conclusions:

• The developed methods enable robust simultaneous assessment of workplace exposures across multiple groups.
• The procedure effectively identifies overexposed worker groups, supporting targeted risk management.
• Recommendations are provided for study design and determining the most beneficial intervention types (e.g., engineering controls vs. individual-level interventions).