Graduation Year


Document Type




Degree Granting Department

Epidemiology and Biostatistics

Major Professor

Amy R. Borenstein, Ph.D.

Committee Member

H. James Brownlee, Jr., M.D.

Committee Member

James A. Mortimer, Ph.D.

Committee Member

Wei Wang, Ph.D.

Committee Member

Theresa Zesiewicz, M.D.


Novelty-Seeking, Routinization, Risk-Taking, Occupation, Premorbid, Risk Factors, Epidemiology


Introduction: Previous epidemiologic studies suggest that the personality of Parkinson's disease (PD) patients differs from that of controls, and laboratory evidence supports a potential common pathophysiology of personality traits and PD. One nested case-control study found that PD cases were significantly more anxious than controls before the clinical onset of the disease, and additional data suggest that certain occupations may be risk factors for the disease. Additionally, the latent period that precedes the onset of motor symptoms of PD is unknown.

Objectives: The objectives of this study were to evaluate the association of PD with objective indicators of current and pre-morbid personality, to determine the correlation of early-adult life personality indicators with current personality characteristics and to evaluate the role of personality as indicated by occupational choice and employment patterns in the risk for PD using the Dictionary of Occupational Titles job classification system.

Methods: Eighty-nine cases and 99 controls completed in-person structured interviews. Assessments included measures of current personality characteristics and indicators of early-adult (ages 20-35 years) personality, such as activities and lifestyle patterns. Associations between these latent personality variables and current personality characteristics were studied using correlation, partialling out the effects of age, sex and education. Multiple logistic regression was used to evaluate the associations of early-adult personality and occupational characteristics and the risk for Parkinson's disease.

Results: Cases with Parkinson's disease reported higher levels of neuroticism (OR=1.05 (95% CI 1.00-1.11)) and harm-avoidance (OR = 1.07 (95% CI 1.00-1.15)) compared with controls on measures of current personality. A stable association among many traits, particularly traits such as novelty-seeking, which are driven by dopaminergic function, was present not only among controls with presumably normal dopaminergic function throughout their lives, but also among cases. Early-adult life routinization was correlated with current levels of neuroticism (cases: r=0.33, p=0.01; controls: r=0.26, p=0.04), extraversion (cases: r=-0.33, p=0.01; controls: r=-0.33, p=0.04), novelty-seeking (cases: r=-0.33, p=0.015; controls: r=-0.34, p=0.007) and harm-avoidance (cases: r=0.47, p=0.0003; controls: r=0.45, p=0.0002) and for the association of early-adult life activity risks with harm-avoidance (cases: r=-0.47, p=0.0004; controls: r=-0.42, p=0.0006). Taking or wanting to take "activity risks," such as riding on roller coasters as a young adult was found to reduce the odds of Parkinson's disease (OR = 0.78 (95% CI 0.63-0.97)) in the entire sample, while higher levels of early-adult routinization were associated with a greater risk for Parkinson's disease among women (OR=1.63 (95% CI 1.05-2.53)). Parkinson's disease was inversely associated with the total number of jobs held (OR=0.87 (95% CI 0.75-0.99)) but not with the number of job categories or duration of the primary occupation. Increased complexity of work with people was associated with PD among women (OR=0.69 (95% CI 0.53-0.89), as was less complex work with things (OR=1.45 (95% CI 1.11-1.88). The complexity of work with data, people or things was not associated with the risk for PD among men or in the sample as a whole. Men with PD whose work involved greater complexity with data took fewer activity risks (r=0.32, p=0.02) and reported greater routinization (r=-0.34, p=0.01) as a young adult.

Conclusions: This evaluation of early-adult risk factors advances current knowledge about the premorbid period of PD and supports the hypothesis that a long period of subclinical disease precedes the onset of motor symptoms. These findings validate the association of these early-adult personality traits and PD and support the idea that behaviors associated with PD personality exist many years before the presentation of motor symptoms. Dopaminergic aspects of personality were related to occupational choices and future consideration of this hypothesis is warranted. Since PD is a degenerative disorder, determining the age of onset of this illness is important in the search for modifiable risk factors and neuroprotective strategies.