Hernán and colleagues1 used meta-analytic methods to show that there is strong epidemiological evidence that smokers and coffee drinkers have a lower risk of Parkinson’s disease (PD) than controls. Marder and Logroscino2 commented on this study, writing “Reverse causation (individuals prone to the development of PD are either less likely to initiate the behavior or more likely to quit) is hard to prove. Individuals who are destined to develop PD might have a genetic predisposition or a metabolic factor associated with reduced novelty seeking.” I suggest that this is correct and that the factor is estrogen (as well, possibly, as testosterone). Zuckermann3 noted that people who score highly on psychological scales of sensation seeking (which is associated positively with both smoking and coffee drinking) have higher levels of the gonadal hormones estrogen and testosterone than controls. Thus, gonadal hormones may be expected to act as confounders between sensation-seeking behaviors (treated as risk factors) and disease conditions that are caused by low (or high) levels of these hormones. And there can be little doubt that estrogens provide neuroprotection of central nervous system cells implicated in PD. 4 I have given additional reasons to support the hypothesis that estrogen acts as a confounder between sensation-seeking behavior and PD. 5 It is not clear whether this form of confounding is the sole cause of the associations between PD and various behaviors, but it is highly likely that it is a cause.