Exposure models for the prior distribution in bayesian decision analysis for occupational hygiene decision making

Area of Science:

• Occupational Health
• Industrial Hygiene
• Risk Assessment

Background:

• Estimating worker exposure to hazardous substances is crucial for risk management.
• Traditional methods often rely heavily on subjective expert judgment.
• Improving the accuracy and objectivity of prior probability estimation is essential for Bayesian decision analysis tools.

Purpose of the Study:

• To introduce and evaluate two semi-quantitative methods: Structured Subjective Assessment (SSA) and Control of Substances Hazardous to Health (COSHH) Essentials.
• To compare these methods with expert judgment for determining prior probabilities in Bayesian analysis.
• To assess the practical applicability of these methods in real workplace settings.

Main Methods:

• Utilized Structured Subjective Assessment (SSA) and COSHH Essentials for semi-quantitative exposure assessment.
• Employed two-dimensional Monte Carlo simulations to determine prior probabilities.
• Included expert judgment as a comparative baseline for prior distribution estimation.
• Applied methods to personal exposure data for isoamyl acetate and isopropanol in manufacturing and printing facilities.

Main Results:

• Demonstrated practical application of SSA and COSHH Essentials in occupational settings.
• Showcased methodological improvements in estimating prior distributions for Bayesian decision analysis.
• Highlighted the influence of expert judgment, though reduced, in the proposed methods.
• Provided insights into the advantages and disadvantages of each method for workplace applicability.

Conclusions:

• SSA and COSHH Essentials offer a structured approach to estimating prior probabilities for worker exposure.
• These methods provide a logical framework for decision-making by considering key exposure determinants.
• The study indicates a methodological advancement in Bayesian decision analysis for occupational risk assessment.
• Further validation in diverse industrial environments is recommended to fully establish applicability.