Duration of agricultural pesticide exposure application and Parkinson's disease in California's central valley.

While numerous pesticides have been linked to Parkinson's disease (PD), few studies have evaluated the long-term effects of ambient exposure to the wide range of pesticides used in commercial agriculture across multiple decades and exposure windows. We examined the contribution of exposure duration and timing to PD risk in 829 PD patients and 824 controls enrolled in California's Central Valley. Using a validated geospatial model integrating California pesticide use reporting and land-use data, we quantified proximity-based ambient pesticide applications at residential and workplace addresses for 287 pesticides applied within 500 m of at least 25 participants' locations between 1974 and 10 years prior to the index year (diagnosis for cases, enrollment for controls, 1998-2016). The exposure duration was calculated as the proportion of eligible years with any nearby application. Unconditional logistic regression models (false discovery rate (FDR)-corrected) estimated exposure effects for each pesticide on PD risk, and constrained distributed lag models evaluated associations across time windows before diagnosis. Longer exposure duration to 56 pesticides was associated with increased PD risk (FDR < 0.05), with odds ratios ranging from 1.10 to 1.25 per standard deviation increase in exposure. Among these, 34 overlapped with pesticides previously identified using an intensity-based metric. Endothall showed the strongest association (OR = 1.27, 95% CI: 1.16-1.39). Associations were generally stronger for exposures occurring 11-20 or 21-30 years prior to diagnosis. These findings indicate that prolonged exposure duration to specific pesticides is relevant to PD risk and suggest latency periods of two to three decades for several agents.