“…We selected potential covariates based on prior studies on air pollutants and PD. 21 , 25 , 29 , 42 Covariates were assessed from the study enrollment questionnaire, including demographics [i.e., age, self-reported race and ethnicity (non-Hispanic white, non-Hispanic black, and others, including American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and all those of Hispanic ethnicity regardless of race), 43 , 44 and educational attainment (high school or below, some college or bachelor’s degree, and graduate degree)] 44 – 46 ; lifestyle factors [i.e., body mass index (BMI) ( ), 47 , 48 smoking status (never, former, and current smoker), 49 – 51 alcohol consumption status (current vs. noncurrent drinker), 51 , 52 daily caffeine intake (quartiles), 53 and physical activity (metabolic equivalent-hours/week for all activities)]; geographic covariates [i.e., US national percentile of block group area deprivation index (2000) defined by percentiles (ADI, , , , and percentile), 54 , 55 US census region (Northeast, Midwest, South, and West), 56 , 57 and residential area type (rural, small town, suburban, and urban)] 57 , 58 ; and finally self-reported health status as a surrogate for general health and comorbidities (excellent, very good or good, and fair or poor). Age at baseline was used in all analyses as a continuous variable.…”