Parkinson's disease: a test of the multifactorial etiologic hypothesis.

We studied the relative etiologic importance upon the development of Parkinson's disease (PD) of occupational exposure to herbicides and other compounds, ionizing radiation exposure, family history of PD and essential tremor, smoking, and history of various viral and other medical conditions. We identified patients (n = 130) with neurologist-confirmed idiopathic PD through contacts with Calgary general hospitals, long-term care facilities, neurologists, the Movement Disorder Clinic, and the Parkinson's Society of Southern Alberta, and selected two matched (by sex and age +/- 2.5 years) community controls for each case by random digit dialing. We obtained lifetime work, chemical, radiation, medical, and smoking exposure histories and family histories of PD and essential tremor by personal interviews, and analyzed the data using conditional logistic regression for matched sets. After controlling for potential confounding and interaction between the exposure variables, using multivariate statistical methods, having a family history of PD was the strongest predictor of PD risk, followed by head trauma and then occupational herbicide use. Cases and controls did not differ in their previous exposures to smoking or ionizing radiation; family history of essential tremor; work-related contact with aluminum, carbon monoxide, cyanide, manganese, mercury, or mineral oils; or history of arteriosclerosis, chicken pox, encephalitis, hypertension, hypotension, measles, mumps, rubella, or Spanish flu. These results support the hypothesis of a multifactorial etiology for PD, probably involving genetic, environmental, trauma, and possibly other factors.