Rheumatoid Arthritis among Women in the Agricultural Health Study: Risk Associated with Farming Activities and Exposures ANNECLAIRE J. DE ROOS, MPH, PHD, GLINDA S. COOPER, PHD, MICHAEL C. ALAVANJA, DRPH, AND DALE P. SANDLER, PHD PURPOSE: Farming has been associated with increased risk of rheumatoid arthritis (RA) in some studies, but specific causes have not been identified. We studied risk factors for RA in the Agricultural Health Study, a cohort of over 57,000 licensed pesticide applicators and their spouses. METHODS: We used a nested case-control design, limited to female participants. Physician-confirmed cases (n Z 135) were matched to five controls each (n Z 675) by birth date. We used logistic regression, adjusting for birth date and state to examine associations, as estimated by odds ratios (OR) and 95% con- fidence intervals (CI). RESULTS: Risk of RA was not associated with mixing or applying pesticides overall or with any pesticide class, nor did it vary by number of days or years of use. Certain pesticides were associated with small non- significantly increased risks, including lindane (OR Z 1.8, 95% CI: 0.6–5.0). RA risk was associated with welding (OR Z 2.1, 95% CI: 0.8–5.4), albeit imprecisely, but not with solvents or sunlight. CONCLUSIONS: We did not identify any strong risk factors for RA. Because of the severe disability associated with this relatively common disease, further investigation into causes is warranted both in the Agricultural Health Study and elsewhere. Ann Epidemiol 2005;15:762–770. Ó 2005 Elsevier Inc. All rights reserved. KEY WORDS: Rheumatoid Arthritis, Autoimmune Diseases, Autoimmunity, Pesticides, Farming, Occupation. INTRODUCTION Rheumatoid arthritis (RA) affects approximately 1% of the United States population, and as many as 2%–3% of those over age 60 (1, 2). Women are more likely to be affected than men, for unknown reasons (3). The course of the dis- ease varies widely, but is generally associated with progres- sive disability and early mortality (1). Several epidemiologic studies have reported increased risk of RA among farmers (4–8). Use of pesticides has been associated with slightly increased risk of RA (20%– 30% increases) (4, 6); however, specific pesticides have not been studied epidemiologically. The prevalence of anti- nuclear antibodies (ANAs), a serologic expression of auto- immunity that is not specific to RA, has been associated with residence on a farm among women, as well as with exposure to insecticides including organochlorines, carba- mates, and pyrethroids, and exposure to phenoxyacetic acid herbicides (9). Several organophosphate insecticides have been implicated as having toxicologic properties rele- vant to systemic autoimmunity, including malathion (10) and chlorpyrifos (11). Case ascertainment and exposure assessment are major difficulties in studying farming exposures as potential causes RA. Self-reporting of RA has been shown to be extremely unreliable in other studies, with confirmation as low as 21%–22% (12, 13). With case ascertainment often requir- ing extensive validation through medical records, case groups for epidemiologic studies tend to be small. Farming occupation and farm residence are not common in most study populations, and exposures to specific pesticides are even less frequent. The combination of small case groups and infrequent exposures can hinder the informativeness of such studies. The Agricultural Health Study, a cohort of licensed pes- ticide applicators and their spouses in Iowa and North Car- olina, provides an excellent opportunity to study the health effects of pesticide exposures. Because this population af- fords a large number of persons involved in farming, the effects of pesticides can be investigated in relation to relatively rare diseases. We investigated pesticide use and other farm-related activities and exposures as possible risk From the University of Washington Department of Epidemiology and Fred Hutchinson Cancer Research Center, Seattle, WA; the Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD; and the Epidemiology Program, National Institute for Environmental Health Sciences, Research Triangle Park, NC. Address correspondence to: Anneclaire J. De Roos, M.P.H., Ph.D., Fred Hutchinson Cancer Research Center and University of Washington, De- partment of Epidemiology, 1100 Fairview, Avenue N, M4-B874 Seattle, WA 98109. Phone: (206) 667-7315; Fax: (206) 667-4787. E-mail: deroos@u.washington.edu Received September 23, 2005; accepted April 21, 2005. Ó 2005 Elsevier Inc. All rights reserved. 1047-2797/05/$–see front matter 360 Park Avenue South, New York, NY 10010 doi:10.1016/j.annepidem.2005.08.001