Background: Previous research has demonstrated a protective effect of coffee intake on the risk of developing Parkinson’s Disease (PD).
Objectives: To systematically review and critically appraise published evidence examining the effect of coffee consumption on the risk of developing PD.
Methods: We identified eligible English language cohort and case-control studies from Medline and Embase data-bases (1950 and 1980 to June 4, 2009, respectively) and assessed studies for quality using the Newcastle-Ottawa Checklist. Data extraction was carried out independently by two reviewers before results were meta-analysed using Review Manager 5.0.
Results: Seven cohort studies involving 248,511 participants and 18 case-control studies with 14,345 participants met eligibility criteria for inclusion. Only one case-control study and one cohort study met all quality assessment criteria. Results from six cohort and 16 case-control studies were included in the meta-analysis. The pooled Relative Risk from these studies demonstrated a protective effect of coffee drinking (RR=0.68; 95% CI 0.59-0.77). The protective effect appeared stronger in males (RR=0.60; 95% CI 0.51-0.69) when compared with females (RR=0.78; 95% CI 0.65-0.93). Significant heterogeneity existed in results from case-control but not cohort studies. There was evidence of publication bias when comparing studies published before and after 2002 (Chi² = 3.78 (P = 0.05)).
Conclusions: This systematic review confirmed an inverse relationship between coffee intake and PD, suggesting that caffeine may be neuroprotective. However, publication bias and confounding may have influenced results.
Limitations: Unpublished literature was not systematically identified. Reliance on unadjusted estimates may have biased results.