Studies have shown that industrial hygiene exposure assessments are subject to inaccurate judgment, may not be different from random chance, and may be biased low, resulting in underestimation of exposures. However, these limitations can be mitigated by the use of modeling and statistical tools. This study shows that before sample collection or when few samples have been taken, modeling tools such as the Structured Deterministic Model (SDM) 2.0 and Expostats can augment a practitioner's risk communication to an employer. SDM and Expostats were used to analyze compliance datasets as follows: (1) comparing the performance of the SDM tool to predict airborne lead and silica overexposures as confirmed by sample data; (2) comparing statistical outputs for similar exposure groups (SEGs) between initial samples that did not contain overexposures and respective follow-up samples that did contain overexposures for airborne lead and iron oxide, and noise; and (3) statistical analysis of airborne lead and silica, and noise datasets that did not contain overexposures to determine the prevalence of unacceptable overexposure risk. In evaluation one, SDM correctly predicted overexposures in all SEGs where samples were above the exposure limit. Evaluation two results showed that inferential statistics from the initial results indicated an unacceptable risk of overexposure, and this was confirmed by follow-up samples that contained overexposures. In evaluation three, results suggested an unacceptable risk of overexposure in half of the datasets analyzed. The results of this study demonstrate that compliance-focused practitioners could also voluntarily leverage modeling and statistical outputs to communicate overexposure risk and encourage controls to reduce the risk of future occupational illnesses and compliance activity.
Johnson et al. (Tue,) studied this question.