The Agricultural Health Study (AHS) is a long-term prospective study of potential health effects associated with exposure to pesticides and other agricultural exposures. The study is a collaboration of the National Cancer Institute, the National Institute of Environmental Health Sciences, and the US Environmental Protection Agency. Farmers and their spouses and children are exposed to numerous potential hazards - many of which, such as pesticides, are relevant to the population at large. Farmers have been shown to be at increased risk for some cancers. Experimental and human studies of acute or high-dose exposures suggest that farmers may also be at increased risk of other adverse health effects. We are examining cancer incidence and other health endpoints in a prospective study of licensed pesticide applicators, spouses and children from North Carolina (NC) and Iowa (IA). Over a 3-year period, we enrolled more than 57,000 licensed applicators, representing 82% of eligible private pesticide applicators (largely farmers) in IA and NC and 43% of commercial applicators from Iowa. About 40% of the private applicators also completed a more detailed take-home questionnaire covering farming practices and health. Nearly 32,000 spouses of farmer applicators enrolled and 2/3 of these also provided data on reproductive health, including information on all children under age 21. Cancer incidence among the applicators, spouses, and children is determined through linkage with the population-based cancer registries in IA and NC. Follow-up computer-assisted telephone interviews are used to update exposure information and to collect information on changes in health status since enrollment in the study. Buccal-cell samples are being collected from participants in order to extract DNA for future analyses of gene and environment interactions. In addition, a dietary assessment (food frequency questionnaire) is self-completed and returned by mail. In-depth assessment of agricultural exposures is being made for a small sample of farmers and their families in order to validate the questionnaires and facilitate exposure classification. Both environmental and biological samples are being collected. Non-cancer outcomes of particular interest in this follow-up period include Parkinson's Disease, autoimmune diseases, and premature ovarian failure in addition to cancer. An add-on study funded by the Office of Research on Minority Health involves African American farmers and farm workers recruited through churches in eastern North Carolina and includes health questionnaires, collection of a blood sample to measure DDE, lipids, and steroid hormones, and measures of adiposity and hair loss. Recent results from the Agricultural Health Study include assessment of cancer incidence after approximately 5 years of follow-up and continued evaluation of potential links between farm exposures and health endpoints. We have nearly completed 5-year follow-up interviews with the cohort members and are planning for a second round of follow-up interviews to begin in 2005. Selected results: The overall cancer incidence among private pesticide applicators and their spouses is lower than expected based on cancer rates in the two participating states. The standardized incidence ratio for all cancers was 0.8 (95 % confidence interval (CI) 0.76-0.84) for private applicators and 0.83 (95% CI 0.77-0.89) for spouses. This reduced cancer risk is largely due to the low levels of smoking related cancers (and associated low levels of smoking in this cohort. Cancer incidence in commercial applicators was similar to that expected (SIR = 1.01, 95% CI 0.81-1.24). Rates for some individual cancer were increased, notably prostate cancer which was increased for both commercial and private applicators, and melanoma, which was significantly increased among spouses. The prostate cancer excess was explored in greater detail using follow-up data through 1999. Use of chlorinated pesticides among applicators over age 50 and the use of methyl bromide were significantly associated with prostate cancer risk. A number of pesticides were associated with prostate cancer among applicators with a family history of prostate cancer whereas the same chemicals were not associated with prostate cancer among those without a family prostate cancer history. Applicators that used alachlor appeared to have increased risk for leukemia and multiple myeloma as well as bladder and prostate cancer, with suggestive dose-response trends observed for leukemia and multiple myeloma. Overall cancer incidence did not appear to be associated with alachlor use. We also found that parent's use of pesticides was associated with risk of cancer in their children. The overall risk of cancer before age 20 was slightly increased for children of Iowa farm applicators (SIR 1.4 (95% CI 1.0-1.8) compared to state rates, as was risk of Hodgkin's lymphoma (SIR 2.6, 95% CI 1.1-6.1). Risk could not be linked to specific pesticides although there was suggestive evidence of increased risk associated with father's use of aldrin prior to conception and with poor pesticide hygiene among fathers. We explored risk factors for wheeze, a cardinal symptom of asthma. Wheeze in the past 12 months was associated with raising animals requiring direct contact. Wheeze was associated with the number of poultry and the number of livestock on the farm, as well as with performing tasks that increased direct animal contact. Risks were moderated by smoking and history of atopy. We also found that wheeze was associated with a wide variety of other farm activities, notably those that result in exposure to diesel exhaust and solvents. We are also exploring several reproductive health outcomes potentially associated with agricultural exposures. For example, in a preliminary analysis, parental use of solvents appears to be associated with self-reported infertility. While use of pesticides in general did not appear to be associated with age at menopause, some specific chemicals were associated with early age at menopause. We found links between non-agricultural exposures and specific menstrual cycle characteristics which could be indicative of hormone disruption. A preliminary analysis suggests that endocrine disrupting pestsicides affect the cycle characteristics.
Showing the most recent 10 out of 139 publications
