We evaluated the possible association between smoking, coffee drinking, and alcohol consumption and Parkinson's disease (PD). The FRAGAMP study is a large Italian multicenter case–control study carried out to evaluate the possible role of environmental and genetic factors in PD. Adjusted ORs were estimated using unconditional logistic regression. Smoking, coffee, and alcohol consumption were also considered as surrogate markers of lifestyle and analysis was carried out considering the presence of at least one, two, or three factors. This latter analysis was separately performed considering Tremor-Dominant (TD) and Akinetic-Rigid (AR) patients. Four hundred ninety-two PD patients (292 men and 200 women) and 459 controls (160 men and 299 women) were enrolled in the study. Multivariate analysis showed a significant negative association between PD and cigarette smoking (OR 0.51; 95%CI 0.36–0.72), coffee drinking (OR 0.61; 95%CI 0.43–0.87) and wine consumption (OR 0.62; 95%CI 0.44–0.86); a significant trend dose-effect (P < 0.05) has been found for all the factors studied. We have also found a trend dose-effect for the presence of at least one, two or three factors with a greater risk reduction (83%) for the presence of three factors. However, a different strength of association between TD and AR was found with a greater risk reduction for the AR patients. We found a significant inverse association between PD smoking, coffee, and alcohol consumption. When analysis was carried out considering the association of these factors as possible surrogate markers of a peculiar lifestyle the association was stronger for the AR phenotype. © 2010 Movement Disorder Society.