Parkinson's disease and occupational exposures: a systematic literature review and meta-analyses

Objectives We conducted a systematic literature review to identify studies fulfilling good scientific epidemiological standards for use in meta-analyses of relevant risk factors for Parkinson's disease. Methods Our search identified 103 original publications on associations between work and Parkinson's disease. GRADE guidelines were used to ensure high scientific quality, and reliable guidelines were applied to classify the papers. Of the 103 articles, 47 fulfilled good scientific standards while 56 were methodologically deficient and thus excluded from our meta-analyses. Results A total of 23 publications concerned work exposure to pesticides. The weighted relative risk estimate was 1.67 (95% confidence interval 1.42-1.97). A funnel plot and bias test indicated that some publication bias concerning smaller studies might have been present. The risk estimate was not influenced by study design (case-control, cohort, or cross-sectional study) or gender. Higher estimates were found when there was a hereditary taint or onset below age 60. Studies on exposure to metals or electromagnetic fields did not show increased risk. Conclusions Using an elaborated quality protocol, there is now strong evidence that exposure to any pesticide involves a ≥50% increased risk for developing Parkinson's disease.