Estimation and validation of biomarker-based exposures for historical ammonium perfluorooctanoate.

Ammonium perfluorooctanoate (APFO) exposures were estimated for use in an occupational mortality study using detailed work histories of cohort members and an exposure reconstruction model developed from occupational information and serum PFO(-) data collected in 2004 as part of a cross-sectional health survey. Measured serum PFO(-) levels of the health survey participants were linked with the job title held by the individuals at the time of sampling. The median, range, and distribution of serum levels were calculated to determine the typical exposure intensity for each job title. High variability was observed in the serum levels of workers within the same job titles. In addition, working in many "APFO-use" jobs did not result in higher exposure than working in "no APFO-use" jobs. Each job title was then assigned to one of three relative APFO job exposure categories (low, medium, or high). Participants' length of time in their job was examined in relation to their serum PFO(-) level and found unlikely to contribute to misclassification of job titles within exposure categories. The mean of the serum PFO(-) measurements for each job exposure category served as the mean intensity factor. Subsequently, the job exposure categories were applied to all historical job titles of the mortality cohort based on their correspondence with job titles represented in the health survey. The resulting job exposure matrix was validated with additional historical blood data collected between 1979 through 2002 from voluntary participants in a separate biomonitoring program. The validation analyses showed general agreement between estimated and measured exposure, reflecting the within-job-title variability observed in measured serum levels used to classify job exposure.