Smoking and Parkinson's disease. Search for a dose-response relationship.

We tested the hypothesis that the inverse association of smoking and Parkinson's disease (PD) results from a direct pharmacologic benefit of smoking on PD. We mailed questionnaires to the 32,000 members of the United Parkinson Foundation and searched for evidence of a dose-response effect between increasing intensity of smoking and decreasing intensity of PD. Of the 6006 respondents, 3693 met our diagnostic criteria. Despite confining the analysis to subgroups where confounding effects would be minimized, we found no significant correlation between any measure of smoking and any measure of PD severity and conclude that smoking is probably not of benefit in preventing, delaying, or ameliorating PD.