? ? Parents' occupations are one source of exposures that may influence the risk of childhood brain tumors (CBT). Although parental exposure has been studied in relation to CBT, no convincing risk factors have been identified, perhaps because of small sample sizes, heterogeneous case groups, and crude exposure assessment. We propose to study parental occupation and CBT using already collected data from a large case-control study that was limited to one histologic group of tumors and collected detailed job information. We hypothesize that maternal and paternal exposure to three classes of pesticides (insecticides, herbicides, and fungicides) increase the risk of medulloblastoma/ primitive neuroectodermal tumor (MB/PNET). We will also hypothesize a role for job activities, such as welding, painting, and degreasing, suggested by previously reported associations with specific occupations and industries. As secondary aims, we will investigate exposure to animals and animal products. The sample consists of information on 318 case mothers, 318 control mothers, 272 case fathers, and 258 control fathers from a study of MB/PNET in young children conducted by the Children's Oncology Group. For fathers, information on jobs from age 18 until the reference date was collected including dates of employment, substances and tools used, activities and tasks, and for more than 60 jobs, a module tailored to the specific job. For pesticides, two experts will perform the exposure assessment based on the detailed job information provided by the parents, a compilation of an extensive industrial hygiene literature review, and their own experience and expertise. For animal exposure, the two experts will use the detailed job information from the parents and their own expertise to assign exposure. Data analysis for pesticide exposure will include intensity and probability of exposure and the rater's confidence in the assignment and will include sensitivity analyses to determine the likely effect of interrater variability. For the analysis of job activities, we will consider ever versus never performed the activity as well as time spent at it. Demographic and other confounders will be considered for each exposure and activity. Parental occupation is likely to remain a focus of research on CBT and childhood cancer generally. The results of this study will provide the basis for further research in terms of both exposures and methods worthy of further attention. In terms of exposures, we will determine whether pesticide or animal exposure is more likely to explain the reported association with agriculture, and which pesticide class deserves the most attention. The results for job activities will suggest exposures to study further. In terms of methods, the approach to pesticide exposure can be applied to other exposures as well. The assessment of job tasks has been used little or not at all in studies of occupational exposure and childhood cancer. The proposed study will provide information on the usefulness of these approaches. ? ? ?
