Historical perspective on approaches to estimation of inhalation risk by air sampling.

Epidemiological investigation involving historical cohorts requires reconstruction of individual cohort member exposures during the period of study. However, past approaches to exposure assessment by industrial hygienists were characterized by different professional expectations for air sampling, and different orientations for evaluation of air sampling results. The activity designated as "exposure assessment" today was not seriously undertaken prior to the late 1970s. It is important to understand the methodologies utilized in earlier years to assess the quality of workplace air because assumptions of exposure during those years are critically dependent on the utilization of shortterm air sampling results obtained by methods which differ from those employed today. This paper reviews the approaches and methodologies utilized in the United States during the 1930-1980 period to measure airborne concentrations of suspended particulate matter and gaseous contaminants. Specific examples are used to illustrate the obstacles encountered when translating these usually limited data into cumulative exposure indices utilized in testing the hypothesis of a dose-response relationship in epidemiological investigations. The misuse of these data results in exposure misclassification and obscuration of any dose-response relationship(s). Selected options for validation of earlier exposures and for increasing or decreasing confidence in recognized incomplete earlier exposure information are discussed. The problems identified and explored in this presentation are inevitable in dealing with available historical occupational exposure data. They will persist for years and demand attention if the information associated with these incomplete data sets is to be defensibly utilized.