Table 4

Cumulative exposure to specific subclasses of pesticides and Parkinson disease risk: exposure algorithm

Cases
n (%)
Controls
n (%)
Crude
OR (95% CI)
Model 1*
OR (95% CI)
Model 2†
OR (95% CI)
Insecticides‡
 Never386 (86.9)768 (87.7)111
 >0–2018 (4.1)36 (4.1)0.99 (0.55 to 1.78)0.94 (0.51 to 1.75)0.99 (0.51 to 1.92)
 >21–19417 (3.8)37 (4.2)0.92 (0.51 to 1.68)1.03 (0.54 to 1.96)1.12 (0.57 to 2.22)
 >195–970223 (5.2)35 (4.0)1.31 (0.76 to 2.26)1.29 (0.71 to 2.34)1.46 (0.76 to 2.81)
Herbicides‡
 Never386 (86.9)770 (87.9)111
 >0–1117 (3.8)35 (4.0)0.96 (0.53 to 1.74)0.92 (0.50 to 1.71)0.98 (0.51 to 1.92)
 >11–14718 (4.1)37 (4.2)0.98 (0.55 to 1.76)1.12 (0.60 to 2.09)1.21 (0.62 to 2.33)
 >147–970223 (5.2)34 (3.9)1.36 (0.78 to 2.37)1.33 (0.73 to 2.43)1.52 (0.78 to 2.97)
Fungicides‡
 Never412 (92.8)819 (93.5)111
 >0–739 (2.0)19 (2.2)0.96 (0.43 to 2.15)1.08 (0.47 to 2.51)1.17 (0.49 to 2.81)
 >73–3149 (2.0)19 (2.2)0.97 (0.44 to 2.14)1.52 (0.64 to 3.56)1.66 (0.69 to 4.00)
 >314–668414 (3.2)19 (2.2)1.44 (0.72 to 2.88)1.23 (0.59 to 2.60)1.38 (0.64 to 2.99)
  • *The first adjusted model includes cigarette smoking (5 categories), coffee consumption (4 categories) and occupational skill and status (4 categories).

  • †The second adjusted model additionally includes endotoxin exposure (4 categories). Since information on coffee consumption was missing for one control, this participant was excluded from adjusted analyses.

  • ‡Conditional logistic regression analyses of pesticide exposure as assessed by the exposure algorithm. Cumulative exposures were estimated by multiplying the exposure algorithm, assigned exposure from pesticide application and re-entry work with all years worked in a job, summed across all jobs of a participant's occupational history from 1955. The exposed were divided based on the tertiles of the exposure distribution among the controls.