Exposure-response analysis of cancer mortality in a cohort of workers exposed to ethylene oxide.

The authors previously reported results from the largest cohort mortality study of ethylene oxide-exposed workers that has been conducted to date. Here they extend their previous work by quantitatively examining the relation between cancer mortality and ethylene oxide exposure. This study included workers from 13 of the 14 geographically distinct facilities that were included in the previous investigation. These facilities began regularly using ethylene oxide to sterilize medical supplies or spices sometime between 1938 and 1969. Workers were followed from first exposure through December 31, 1987. Historical exposures to ethylene oxide were estimated using a regression model. Standard life-table analysis was used to examine cancer mortality in three categories of cumulative exposure to ethylene oxide. The Cox proportional hazards model was also used to examine cumulative and other measures of ethylene oxide exposure as predictors of cancer mortality. In both the life-table analysis and the Cox model, a positive trend was observed in all lymphatic and hematopoietic cancer mortality for cumulative ethylene oxide exposure. This trend was strengthened when ethylene oxide exposures 10 years prior to death were discounted (lagged) and when the analysis was restricted to neoplasms of lymphoid cell origin. Despite limitations discussed in this paper, the authors believe that these findings provide some support for the hypothesis that exposure to ethylene oxide increases the risk of mortality from lymphatic and hematopoietic neoplasms. The authors intend to continue follow-up of this relatively young cohort, which may allow more definitive conclusions to be drawn in the future.