Recent investigations by the Occupational Studies section of the National Cancer Institute have found associations between pesticide exposure and cancer. Particularly striking associations were seen between the herbicide 2,4-D and non-Hodgkin""""""""s lymphoma among farmers in Kansas and Nebraska. Other associations have been noted between chronic lymphatic leukemia, soft-tissue sarcoma, and lung cancer and exposure to certain insecticides. Despite these promising leads, epidemiologic studies of cancer and pesticides has been stymied by the difficulty in accurately assessing historical pesticide exposures. Monitoring data or use records are generally not available on study subjects and investigators are forced to rely upon current measurements and information from interviews. The quality and reliability of these data are not known and exposure misclassification that is likely to occur tends to bias estimates of relative risks toward the null. This bias increases the probability of missing important associations. The objective of the methodologic study supported by this contract is to improve exposure assessment of pesticides in epidemiologic studies. Information on pesticide exposure will be obtained from interviews of farmers and their next-of-kin, from interviews with pesticide suppliers, from review of farmer's purchase records, and from air and biochemical monitoring during pesticide mixing and application. The comparability of the information obtained from these sources will be reviewed to develop new and improved algorithms for estimating pesticide exposures in future epidemiologic studies.
