Characteristics of Worker Populations: Exposure Considerations in the Selection of Study Populations and Their Analysis

Abstract

For many chemical exposures, the most appropriate study populations are among the diverse end-users of the materials rather than the large producers or processors. Important exposures often occur in complex process settings that are not easily characterized but which could be epidemiologically evaluated. Industrial hygiene approaches are needed for characterizing process emissions when the relevant constituents are unspecified. Aggregating study populations from plants where there is a mature labor-management dynamic allows access to the collective memory of the plant populations, a major resource in exposure reconstruction. In population-based studies that apply exposure matrices to interview data, the use of interviewers trained in industrial hygiene would help minimize misclassification. Restricting population-based studies to a subset of employers would also facilitate exposure classification. Analyses using internal exposure comparisons are essential to address noncomparability of study and reference populations. Improved statistical efficiency can be achieved through the latency weighting of exposures and the modeling of exposure-mortality associations using external reference rates. Paradoxical dose-response can arise from imprecise exposure classifications if a worker's duration in the exposure is dependent on exposure level. When exposure data are limited, statistically significant excesses showing paradoxical dose-responses should be interpreted as evidence in support of an exposure effect.