Table 2. Interaction, main, and joint effect estimates between HLA rs3129882 and pyrethroid exposure in PEG study population of European ancestry, using both an additive genetic model and AA versus GG; n=962 (patients =465, controls =497).

Ambient pyrethroids a	None				1+ Pesticide
	Cases, n (%)	Controls, n (%)	Adjusted OR b (95% CI)	P value	Cases, n (%)	Controls, n (%)	Adjusted OR b (95% CI)	P value
Additive genetic model
AA	95 (0.30)	117 (0.33)	1.00 (ref)		47 (0.32)	58 (0.42)	0.83 (0.53, 1.28)	0.42
AG	172 (0.54)	161 (0.45)	0.91 (0.73, 1.13)	0.38	71 (0.48)	66 (0.48)	1.25 (0.88, 1.78)	0.22
GG	50 (0.16)	81 (0.23)	0.82 (0.53, 1.27)		30 (0.20)	14 (0.10)	1.87 (1.08, 3.35)	0.03
P value for interaction								0.02
AA versus GG
AA	95 (0.66)	117 (0.59)	1.00 (ref)		47 (0.61)	58 (0.81)	1.04 (0.65, 1.67)	0.87
GG	50 (0.34)	81 (0.41)	0.73 (0.47, 1.14)	0.17	30 (0.39)	14 (0.19)	2.48 (1.24, 4.97)	0.01
P value for interaction								0.007

Abbreviations: CI, confidence interval; OR, odds ratio; PD, Parkinson’s disease; PEG, Parkinson’s Environment and Gene.

Using both an additive genetic model and comparing only the homozygous groups, we assessed the association between pyrethroid exposure and the risk rs3129882 genotype in the risk for PD. The table indicates the adjusted OR and P values for the main and joint effects and the P value for interaction.

^aAmbient pesticide exposure to any pyrethroids (at or above the median level seen in exposed controls) at both occupation and residence, from 1974 (year of California State mandated pesticide use reports implementation) to 10 years before diagnosis or interview. Pyrethroid group includes fenvalerate, permethrin, phenothrin, resmethrin, flucythrinate, cypermethrin, (S)-cypermethrin, tau-fluvalinate, fenpropathrin, lamda-cyhalothrin, bifenthrin, esfenvalerate, and tralomethrin; Cyfluthrin had no exposure in study population.

^bAdjusted for age (continuous), sex, and smoking history.