Parkinsonism and pesticide exposure among rural residents of Washington State
Engel, Lawrence Stuart
MetadataShow full item record
Objective. To determine the prevalence of parkinsonism among a cohort of rural men, primarily orchardists, and to examine associations between that prevalence and smoking, occupational pesticide use, and MAO-B and ND1/4216 genetic polymorphisms. Also to assess the accuracy of pesticide use recall. Methods. All 323 subjects had participated during the 1970s in a cohort study of men occupationally exposed to pesticides. Subjects were given a neurologic examination and completed a questionnaire concerning lifetime occupational pesticide use. Parkinsonism was defined by the presence of two or more of resting tremor, rigidity, bradykinesia, and postural reflex impairment. Prevalence ratios were estimated for parkinsonism. in relation to smoking, farming, pesticide use, genotype, and gene-environment interactions. Recall sensitivity and specificity were estimated by comparing pesticide use reported in the current study to that reported in the original study. Results. Mean age of subjects was 69.4 years (range: 49--96). Prevalence of slight or greater parkinsonism was 20.7% and mild or greater parkinsonism was 1.5%. A prevalence ratio of 2.0 (95% CI: 1.0, 4.2) was observed for subjects with the longest duration of general pesticide exposure; the prevalence ratio for intermediate duration was elevated but non-significant. However, we observed no increased risk associated with specific pesticides or pesticide classes, nor with a history of farming. We found no association between smoking or genotype and parkinsonism, nor any modification of effect of smoking, pesticide exposure, or farming by either genotype. Sensitivity of pesticide recall was good to excellent (0.6--0.9) for broad categories such as insecticides, herbicides, and fungicides, and for widely used pesticides and pesticide classes, but was lower (0.1--0.6) for other pesticide categories. Conclusion. There was an increased risk of parkinsonism associated with long-term occupational exposure to pesticides, but we could not detect any associations between parkinsonism and specific pesticides. Parkinsonism was not associated with smoking or with polymorphisms of MAO-B intron 13 or NDI/4216, nor were there interactions between genotype and any exposure in relation to parkinsonism. The level of recall accuracy observed here is probably adequate for epidemiologic analyses of broad categories of pesticides, but is a limitation for detecting more specific associations.
- Epidemiology