Objective: We sought to develop an algorithm and estimate unmeasured exposures to particulate matter (PM) and metals in an epidemiologic study of boilermakers.
Methods: The algorithm was based on limited measurements and workers' task and time activity patterns. Half of the measurements were used to develop exposure estimates for unmeasured person days. The other half was used for method validation.
Results: The validation demonstrated good approximations of actual exposures with differences less than 5% for PM and vanadium (V). Average estimated exposures to PM (mg/m3) and V (microg/m3) were significantly higher for workers doing boiler repair than utility work (0.36 vs. 0.09 for PM and 5.99 vs. 0.38 for V).
Conclusions: This algorithm provided reasonably accurate exposure indices for our epidemiologic study in this population. It also is likely applicable to similar exposure scenarios in other studies.