Coffee consumption, caffeine and self-reported parkinson's disease in the adventist health study-2

Epidemiologic studies consistently find inverse associations between Parkinson's disease (PD), coffee drinking and cigarette smoking. However, an independent association between coffee consumption and PD risk has been difficult to assess given the correlation between coffee drinking and smoking. The Adventist Health Study-2 provides an opportunity to study PD while minimizing confounding by smoking, as the cohort consists predominantly of never smokers. A third follow-up questionnaire administered on average 5 years (range 1 – 9) after baseline identified 260 newly reported PD cases (8.2 per 10,000 average annual incidence). Mean age at diagnosis was 72.6 years; 186 (71.5%) reported having been treated for PD within the last 12 months. Of the incident cases, 107 were female (41.2%), 229 (88.1% were non-Hispanic white); 95 (37.1%) had a college education or higher. Four controls were matched to each case based on age (5-year categories), sex and race. 80.9% of cases and 76.7% of controls were never smokers; 4.2% of cases and 15.9% of controls reported consuming some coffee at varying frequencies. In adjusted logistic regression models, any coffee consumption (times per month, week or day) was not associated with decreased PD risk (OR = 1.07, 95% CI = 0.75, 1.52). Frequency of consumption weighted for serving size of coffee or caffeine from all sources (caffeinated tea, chocolate) was also not associated with PD. This study is not supportive of a protective effect of coffee or caffeine in PD while minimizing confounding by smoking. Additional analyses of PD in AHS-2 will merit from the non-smoking makeup of the cohort.