Implications of different fiber measures for epidemiologic studies of man-made vitreous fibers

Background Several researchers have proposed fiber size ranges representing hypotheses about the biological activity of fibers. Each of these indices has a biologically plausible rationale, yet they propose different fiber dimensions. Methods Four biologically based and one standard fiber analytic method index, the NIOSH 7400 method with B rules, were evaluated in air monitoring data collected in a variety of fiber glass (FG) manufacturing settings to determine their impact on an assessment of fiber exposure for an epidemiologic study of the risk of respiratory cancer among FG production workers. Results All indices varied considerably within and among the range of fiber size distributions sampled in the FG factories. Conclusions The asymptotic relative efficiencies (R2) calculated among the five indices indicate a potentially important loss of power from the use of the standard method index, if one of the biologically based indices is more closely related to the disease outcome in an epidemiologic study. Am. J. Ind. Med. 38:132–139, 2000. © 2000 Wiley‐Liss, Inc.