Associations of direct pesticide application defined broadly and specifically, well-water consumption, and farming residences and occupations with PD were evaluated in 319 cases from 308 families and 296 relative and other controls. Of the controls, 252 were relative controls from the 308 families with at least one case, and these relative controls included 237 siblings, 10 parents or children, and 5 cousin or avuncular relatives (uncles/aunts/nephews/nieces). The remaining 44 controls were ascertained as spouse or other unrelated controls or as relative controls in families with no environmental risk factor data available on the case(s). Overall, males represented 67.7% of cases and 43.2% of controls. AAE of cases ranged from 29 to 94 years with a mean of 65.6 ± 10.1 years. AAE of controls ranged from 24 to 95 years with a mean of 63.8 ± 12.0 years. Cases were thus more likely to be male (p < 0.0001) and older at examination (p = 0.0013) when compared to controls. Given these significant differences, sex and AAE were included as confounders in all models. Also, cases reported onset ages ranging from 16 to 85 years with a mean of 57.6 ± 11.5 years and a mean disease duration of 8.0 ± 6.0 years. Only 39 of the 319 cases (12%) reported having symptoms of PD for two years or less.
Direct pesticide application
Results from models examining associations of childhood and adulthood direct pesticide application with PD, while adjusting for sex, AAE, smoking, and caffeine in the overall data are presented in Table . Individuals with PD were 1.61 times as likely to report ever being exposed to pesticides through direct application as their unaffected relatives. Use of protective gear during application did not alter the significant association between history of direct pesticide application and PD (data not shown). The highest exposure categories of frequency, duration, and cumulative exposure and the lowest duration category were also positively associated with PD, and increasing frequency, duration, and cumulative exposure were associated with PD in a dose-dependent manner. These patterns were unchanged when examining adulthood applications only (data not shown).
Associations of childhood and adulthood direct pesticide application with PD.
Associations of childhood and adulthood direct pesticide application were also examined in males and females, separately. Significant associations between direct pesticide application and PD were present in both males and females, despite a smaller stratum size and lower application levels for females. A significant positive association between the highest frequency category and PD appeared in both sexes (OR = 2.15; 95% CI: 1.06, 4.35 for males and OR = 2.43; 95% CI: 1.18, 5.01 for females), and a dose-response trend existed in females (p = 0.0058). In males, PD was significantly associated with the highest duration (OR = 2.70; 95% CI: 1.35, 5.40) and cumulative exposure (OR = 2.34; 95% CI: 1.14, 4.79) categories, and significant dose-response trends were detected (p = 0.021 for duration and p = 0.036 for cumulative exposure). In females, PD was significantly associated with the lowest duration category (OR = 2.47; 95% CI: 1.12, 5.44).
Data were then stratified by family history to determine if the associations of childhood and adulthood direct pesticide application differed in individuals with and without a family history of PD. These results are shown in Tables and . In the negative family history stratum, all patterns of association resemble those in the overall data (Table ) with all measures of association being stronger. Ever being exposed, being exposed at the highest exposure categories of frequency, duration, and cumulative exposure, and being exposed at the lowest duration category were all significantly associated with PD. In fact, individuals with PD and no family history were over three times as likely to report being exposed at the highest cumulative exposure level as their unaffected relatives. Increasing frequency, duration, and cumulative exposure were all associated with PD in a dose-response pattern. In contrast, there were no significant associations with PD or significant trends in OR estimates in the positive family history stratum. This stratum, however, represented a smaller portion of the overall data with lower power to detect these size effects (OR < 3) when compared to the negative family history stratum.
Associations of childhood and adulthood direct pesticide application with PD in negative history families.
Associations of childhood and adulthood direct pesticide application with PD in positive history families.
Direct pesticide application of specific functional types and chemical classes
The chemical or trade names of pesticides reported by individuals in our sample were categorized into chemical classes used primarily as herbicides (2,6-dinitroaniline, chlorophenoxy acid/ester, phosphonoglycine, pyridinecarboxylic acid, and triazine), insecticides (alkyl phthalate, botanical, halogenated organic, N-methyl carbamate, microbial, organochlorine, and organophosphorus), and fungicides (azole, guanidine, and thiophthalimide) along with the inorganic/metal class with multiple uses. There were too few individuals with reported applications of the 2,6-dinitroaniline, pyridinecarboxylic acid, alkyl phthalate, halogenated organic, microbial, azole, guanidine, and thiophthalimide classes and too few reports of fungicides to examine these associations with PD. The pesticide names provided by 57 of 200 cases and 54 of 147 controls who reported applying pesticides could not be classified due to either an inadequate report, an unknown trade name, or multiple active chemicals of different classes contained in the reported pesticide.
When examining associations of direct herbicide application with PD, there were significant positive associations between direct pesticide application and PD for both individuals who reported ever applying herbicides (OR = 1.59; 95% CI: 1.00, 2.54) and for individuals who reported ever applying pesticides other than herbicides (OR = 1.61; 95% CI: 1.09, 2.38) as compared to individuals who never applied pesticides. When examining associations between direct insecticide application and PD, only individuals who reported ever applying insecticides were significantly more likely to develop PD (OR = 1.83; 95% CI: 1.20, 2.81). Individuals who reported ever applying pesticides other than insecticides were not significantly associated with an increased risk for PD.
Associations between childhood and adulthood direct pesticide application and PD were then examined for each chemical class by comparing individuals who reported ever applying pesticides from the relevant chemical class and those who reported ever applying other pesticides to the referent group of those who never applied pesticides. These results are summarized in Table . For each chemical class, there was a significant positive association of reporting a direct application of other pesticides. However, application of only the organochlorine and organophosphorus chemical classes were found to also be significantly associated with PD. In our sample, chlordane and dichloro-diphenyl-trichloroethane (DDT) were the most common of the 10 organochlorine chemicals, while chlorpyrifos, diazinon, and malathion were the most common of the eight organophosphorus chemicals. Two less common classes, the botanical class (including rotenone) and the chlorophenoxy acid/ester class [including 2,4-dichlorophenoxyacetic acid (2,4-D) and Agent Orange], showed strong OR estimates possibly indicative of a positive association with PD, but these associations were not significant.
Associations of childhood and adulthood direct pesticide application with PD as divided into chemical classes.
Well-water consumption at childhood and adulthood residences was not significantly associated with PD in our data. Neither a history of well-water consumption nor duration at any exposure level was significantly associated with PD (Table ). There was also no significant trend in OR estimates for duration of well-water consumption (Table ). Examination of well-water associations at adulthood residences only did not alter the significance of these results, although the OR estimates were higher than those observed with all residences considered for history (OR = 1.22; 95% CI: 0.87, 1.72) and duration (OR = 1.51; 95% CI: 0.90, 2.56 for >21 years, OR = 1.20; 95% CI: 0.71, 2.03 for 6–21 years, and OR = 1.27; 95% CI: 0.80, 2.02 for <6 years).
Associations of well-water consumption at childhood and adulthood residences with PD.
Farming residences and occupations
Living or working on a farm was not significantly associated with an increased risk of PD in our data as shown in Table . Individuals with PD were neither more nor less likely than their unaffected relatives to report ever living or working on a farm or to report any duration level. There was also no significant dose-response relationship between farming residences/occupations and PD. These patterns were unchanged when examining only adulthood farming residences and occupations, and estimated ORs were only slightly higher than those observed with all farming residences and occupations considered (data not shown). There were also no significant associations with PD when examining farming residences and occupations separately (data not shown).
Associations of childhood and adulthood farming residences and occupations with PD.