Pesticide use in agriculture and Parkinson's disease in the AGRICAN cohort study.
Level 4 - case-series / case-control
Cross-sectional analysis of baseline enrolment data within a cohort study
PubMed 29136149 · doi:10.1093/ije/dyx225
What was done
Participants in the French agricultural cohort AGRICAN (181,842 individuals enrolled between 2005 and 2007) reported their history of lifetime exposure to 13 crops, 5 types of animals, and pesticide use. Parkinson's disease (PD) status was self-reported at enrollment. Exposure to specific active ingredients and lifetime duration of use were estimated using the PESTIMAT crop-exposure matrix. Associations between pesticide use and PD were evaluated using logistic regression adjusted for sex, age, education, smoking status, and alcohol consumption.
What was found
PD was reported by 1,732 participants (1.2%) at enrollment. Lifelong pesticide use was associated with an increased risk of PD across all evaluated farming activities (OR = 1.31 for cattle to 1.79 for peas, P < 0.05). Specific active ingredients were associated with increased risk: rotenone, diquat, paraquat, and several dithiocarbamates (OR = 1.31 for cuprobam to 1.57 for rotenone), particularly among farmers with the longest exposure durations.
Why it matters
This study provides large-scale epidemiological evidence identifying specific pesticide classes—including rotenone, diquat, paraquat, and dithiocarbamates—and longer exposure durations as associated with Parkinson's disease in agricultural workers.
Limits
PD status was based entirely on self-report without clinical or neurological confirmation. Exposure assessment relied on recall and an indirect crop-exposure matrix rather than direct individual exposure measurements. The cross-sectional baseline design limits causal and temporal inference, and exact confidence intervals were not reported in the abstract.
Cited by
- supports Farmers who work with the herbicide paraquat have a higher incidence of Parkinson's disease.