Quitting smoking: An early non-motor feature of Parkinson's disease?

I am surprised by this even if I agree with some of the conclusions. There are a number of studies that have looked at former versus never smoking prior to PD diagnosis (despite what's stated here) but they generally suggest a weaker negative effect than for current smoking, not a positive association. In conclusion...reverse causality...very likely

http://www.sciencedirect.com/science/article/pii/S1353802014004805

Parkinsonism Relat Disord. 2014 Dec 17. pii: S1353-8020(14)00480-5. doi: 10.1016/j.parkreldis.2014.12.008. [Epub ahead of print]

Moccia M, Erro R, Picillo M, Vassallo E, Vitale C, Longo K, Amboni M, Santangelo G, Palladino R, Nardone A, Triassi M, Barone P, Pellecchia MT.

INTRODUCTION:

Epidemiological studies report a 60-70% reduced risk of Parkinson's disease (PD) in smokers as compared to non-smokers. However, relationships between former smoking and PD have been poorly investigated.

METHODS:

We recruited 116 de novo PD subjects, and investigated current, former and never smoking, and reasons for smoking cessation among former smokers. Two hundred and thirty-two controls were matched by Propensity Score.

RESULTS:

PD subjects and controls were found to be current smokers (7.7 vs. 39.6%), former smokers (43.9 vs. 6.5%) and never smokers (48.2 vs. 53.9%). Logistic regression showed that current smokers were less likely to have PD (p < 0.001; OR: 0.22; 95% CI: 0.10-0.46), while former smokers were more likely to have PD (p < 0.001; OR: 7.6; 95% CI: 4.09-15.75), as compared to never smokers. Fifty-one PD patients reported quitting smoking before PD diagnosis (mean time since cessation 9.4 ± 7.3 years). Most important reasons to quit smoking in PD group were illness different from PD (26 subjects, 51.0%), knowledge of the harmful effects of smoking (24 subjects, 47.0%), and physician's advice (1 subject, 2.0%).

CONCLUSION:

The reduced prevalence of current smokers among PD subjects as compared to healthy controls is consistent with previous findings, suggesting a possible neuroprotective effect of smoking. However, it could be due, at least in part, to the increased prevalence of former smokers among PD patients, that were more prone to quit smoking as compared to healthy controls. We suggest that smoking cessation could be an early preclinical condition occurring in PD.