This study will examine variation in the levels of biological measures associated with occupational exposures. Although intra- and inter-individual differences have been characterized for a large number of airborne exposures across occupational groups, similar variation has not been investigated extensively for biological measures. A primary objective of this research will be to compile a database of repeated biological measures so that the within- and between-person sources of variability can be partitioned for as wide a range of bio-markers as possible. Given the psychological and kinetic parameters that influence body burdens of contaminants, statistical methods that incorporate the serial dependence among measures will be applied. The proposed study will also investigate whether systematic changes in the underlying body burden distribution have taken place and, if so, account for such changes in the analyses. If the magnitude of the within-person component of variance is large, relative to the between-person component, then large numbers of repeated measurements would be necessary to sort out differences among workers. Under these circumstances when small numbers of repeated measurements are collected, epidemiological studies of occupational cohorts relying on these data would be subject to misclassification and a significant loss of power. Thus, the impact of intra- and inter-individual variation on the design of health-effects studies will be investigated. For data sources comprised of repeated air and biological measurements, information about the variance components will be used to compare the sampling demands of airborne and biological measures of exposure to a particular contaminant. This comparison should enhance the ability of investigators to select the most appropriate exposure measure when epidemiologic studies are carried out. In addition to evaluating airborne and biological measures as independent s u r rogates of exposure, further investigation is needed to evaluate exposure/bio-marker relationships. A secondary objective of the proposed study will be to investigate exposure/bio-marker relationships taking into account both intra- and inter-individual variation and serial correlation. In particular, mixed-effects analysis of variance models will be employed and will provide information as to whether workers within the same occupational group have similar background levels and share uniform underlying kinetic processes (rates of uptake, metabolism, and elimination). By addressing questions about inter-individual differences and taking full advantage of all of the data that have been collected, this application should provide investigators with a useful and novel method to investigate exposure/bio-marker relationships in the future.
