Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Application of statistical modeling to occupational exposure assessment

Thesis/Dissertation ·
OSTI ID:5018460
This dissertation applies statistical modeling to two problems: (1) describing a single worker's exposure distribution and estimating its associated arithemetic mean; and (2) describing the distribution of inhalation exposure levels among a population of respirator wearers while accounting for variability in ambient exposure and respirator penetration values within and between wearers. A task-based statistical construct for a single worker's exposure levels for a single agent is developed; the model accounts for variability in short-term time weighted average (TWA) exposure values within a task, and for variability in arithmetic mean exposure levels between tasks. Five sample survey designs for estimating a worker's arithmetic mean exposure level are examined. Stratified random sampling designs, in which short-term TWAs are measured for time periods selected on a task basis, can provide a more precise estimate of the arithmetic mean exposure level than the traditional survey design for the same fixed cost. For describing inhalation exposure levels (C{sub i}) among a population of air-purifying respirator wearers, a synthesis of lognormal one-way analysis of variance models for ambient exposure levels (C.) and respirator penetration (P) values provides the most tractable construct. The model is applied to assessing the risk of toxicant overexposure for a respirator wearer population. Overexposure to a chronic toxicant is equated with an arithmetic mean exposure level above the permissible exposure limit (PEL) value, while overexposure to an acute toxicant is equated with a 95th percentile exposure level above the PEL value.
Research Organization:
California Univ., Berkeley, CA (United States)
OSTI ID:
5018460
Country of Publication:
United States
Language:
English