Parkinson’s disease (PD) is one of the most common neurodegenerative disorders, with an estimated 6.2 million individuals affected worldwide (Group GBDNDC, 2017). Several lifestyle factors have been shown to influence predisposition to PD (Bellou, Belbasis et al. 2016). Over the last few decades, several case-control and cohort studies have highlighted the possible role of pesticides as a risk factor for PD among individuals working as farmers in villages (Yan, Zhang et al. 2018). However, a substantial heterogeneity has been observed among the effect estimates reported by various studies, which could be attributed to the exploration of a wide variety of pesticides such as fungicides, herbicides and insecticides reported in the literature, often with a shorter duration of exposure. The present proposal aims to explore the relationship between pesticide exposure and PD predisposition using a case-control study design.

The objectives of this study are multifold.

1. As a primary objective, the study aims to explore the association of pesticide exposure and PD in rural Pennsylvania counties.
2. As a secondary objective, the study aims to explore the association of duration of pesticide exposure (dose-effect relationship) and the type of pesticides with PD in rural Pennsylvania counties.

The study findings may help inform local policy makers and health authorities on potential harms of specific classes of pesticides, thereby reducing the incidence of PD.

Background

PD is one of the most common neurodegenerative diseases of old age resulting from progressive degeneration of dopaminergic neurons in the brain (Dorsey, Constantinescu et al. 2007). Due to rapid medical advancements in recent decades, several world populations have shown exponential growth in the proportion of ageing individuals. It is further estimated that this prevalence of late-onset PD will show a similar trend in the coming decades, resulting in an increased public health burden.

Occupational exposure to various classes of pesticides through farming have long been known to pose a risk of PD. It is believed that chemical agents present in the pesticides lead to loss of dopaminergic neurons through accelerated oxidative stress and neuronal apoptosis, as shown in various animal model studies (Nguyen, Wong et al. 2019). Several human population studies based primarily on case-control epidemiological studies have shown an increased risk to PD due to pesticide exposure. A comprehensive meta-analysis comprising of 24 various studies showed that exposure to PD increased the risk of PD susceptibility by 66% (RR:1.66, 95% CI: 1.42,1.94) (Gunnarsson and Bodin 2019). It is, however, believed that a particular group of pesticides pose a greater risk compared to others. For instance, studies have consistently that herbicide named paraquat, containing a chemical 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), is highly neurotoxic (OR:1.64, 95%: CI = 1.27-2.13) (Tangamornsuksan, Lohitnavy et al. 2019).

A recent case-control study in rural California using 360 PD cases and 827 controls concluded that occupational use of carbamates, organophosphorus and organochlorine doubled the risk of PD (Narayan, Liew et al. 2017). The risk was specifically higher among individuals exposed to pesticides for more than ten years compared to those who were never exposed to pesticides. In addition to California, Pennsylvania is also known to have one of the highest prevalence of PD among states in the US (California: 1,520/100,000, Pennysylvania: 1,549/100,000) (Mantri, Fullard et al. 2019). However, studies exploring the possible role of pesticide exposure in rural Pennysylvania counties are lacking.

Methods

Study design

The present proposal aims to conduct an age and gender-matched case-control study to explore the relationship between pesticide exposure and PD susceptibility across rural Pennsylvania counties in the U.S. A convenience sampling method will be used a set of pre-determined five counties in the south of Pennsylvania will be selected. The PD patients will be identified through the local insurance database, using the information on anti-Parkinsonian drugs during the last ten years. Only idiopathic cases of PD will be included. Additionally, those who have a family history of PD will be excluded. Five age and gender matched controls will be randomly recruited from the same county for each PD patient. For both cases and controls, individuals aged 80 years or those with significant cognitive issues will be excluded.

Data collection

A self-administered questionnaire will be given to study participants by a team of epidemiologists and a neurologist who will visit their households after obtaining consent to participate in the study using a telephone call. The primary outcome of the study will be a diagnosis of PD.  Specifically, information on past jobs, whether living or working in farms, the duration of the job, the brand of pesticide used will be collected. Besides, information on potential confounding factors such as age, gender, education levels, ethnicity, and smoking will be collected. The Pennsylvania Department of Pesticide Regulation (PDPR) product label database will be used to identify the pesticide ingredient, and the chemical class of pesticide will be identified using Pesticide Action Network (PAN) pesticide database.

Data analysis

All the data handling and statistical analyses will be performed using IBM Statistical Package for the Social Science (SPSS) for Windows, Version 21.0. (SPSS Inc, Chicago, Illinois). The categorical variables such as gender, education level, and ethnicity will be expressed in proportions. The normality distribution of continuous variables, including age, will be checked using the Shapiro Wilk test. The distribution of categorical variables will be compared between different groups (PD cases and controls) using Chi-square tests. The distribution of parametric variables will be compared between different groups using a t-test or analysis of variance test (ANOVA), as appropriate. The distribution of non-parametric variables will be compared between different groups using the Mann-Whitney U test or the Kruskal-Wallis test. Confidence intervals (CI) will be computed at the 95% significance level. A p-value below 0.05 will be considered statistically significant. A multivariable logistic regression model will be performed for variables that showed borderline significance with PD susceptibility (P<0.01) in the univariable analyses. Odds ratios (OR) will be computed for expressing effect estimates in the case of a binary exposure variable, respectively. The strata specific ORs will be reported after stratifying on the basis of duration of pesticide exposure and the class of pesticide. Specifically, all the PD cases will be sub-divided into two groups on the period of exposure to pesticides (0-10 years and above 10 years).

Strengths and Limitations

Despite being one of the most comprehensive planned study investigating the role of varying duration of different classes of pesticides in a US population, the possibility of various epidemiological biases influencing the study results cannot be ruled out. For instance, due to the nature of the study participants, dependence on recall of cases and controls to collect data on PD exposure could lead to misclassification bias. We, however, plan to adopt a stringent study design by excluding individuals who show considerable cognitive disability. Another potential bias could be that study participants reporting the use of pesticides may have used different brands of pesticides during the period of exposure. Nevertheless, it is expected that only a small proportion of such individuals will be present in the collected data and will be adjusted using regression-based approaches. It may be possible that we may have excluded individuals who do not have access to medical insurance. Such individuals may be relying on the poor quality of potentially neurotoxic pesticides due to poverty, thereby leading to selection bias. And lastly, despite attempts made to exclude genetic cases of PD based predominantly on the family history of participants, the existence of a small number of familial cases in the present study cannot be completely ruled out,

References

Group GBDNDC (2017). “Global, regional, and national burden of neurological disorders during 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015.” Lancet Neurol 16(11): 877-897.

Bellou, V., L. Belbasis, I. Tzoulaki, E. Evangelou and J. P. Ioannidis (2016). “Environmental risk factors and Parkinson’s disease: An umbrella review of meta-analyses.” Parkinsonism Relat Disord 23: 1-9.

Dorsey, E. R., R. Constantinescu, J. P. Thompson, K. M. Biglan, R. G. Holloway, K. Kieburtz, F. J. Marshall, B. M. Ravina, G. Schifitto, A. Siderowf and C. M. Tanner (2007). “Projected number of people with Parkinson disease in the most populous nations, 2005 through 2030.” Neurology 68(5): 384-386.

Gunnarsson, L. G. and L. Bodin (2019). “Occupational Exposures and Neurodegenerative Diseases-A Systematic Literature Review and Meta-Analyses.” Int J Environ Res Public Health 16(3).

Mantri, S., M. E. Fullard, J. Beck and A. W. Willis (2019). “State-level prevalence, health service use, and spending vary widely among Medicare beneficiaries with Parkinson disease.” NPJ Parkinsons Dis 5: 1.

Narayan, S., Z. Liew, J. M. Bronstein and B. Ritz (2017). “Occupational pesticide use and Parkinson’s disease in the Parkinson Environment Gene (PEG) study.” Environment international 107: 266-273.

Nguyen, M., Y. C. Wong, D. Ysselstein, A. Severino and D. Krainc (2019). “Synaptic, Mitochondrial, and Lysosomal Dysfunction in Parkinson’s Disease.” Trends Neurosci 42(2): 140-149.

Tangamornsuksan, W., O. Lohitnavy, R. Sruamsiri, N. Chaiyakunapruk, C. Norman Scholfield, B. Reisfeld and M. Lohitnavy (2019). “Paraquat exposure and Parkinson’s disease: A systematic review and meta-analysis.” Arch Environ Occup Health 74(5): 225-238.

Yan, D., Y. Zhang, L. Liu, N. Shi and H. Yan (2018). “Pesticide exposure and risk of Parkinson’s disease: Dose-response meta-analysis of observational studies.” Regul Toxicol Pharmacol 96: 57-63.