Meta-analysis of early nonmotor features and risk factors for Parkinson disease.

Objective: To evaluate the association between diagnosis of Parkinson disease (PD) and risk factors or early symptoms amenable to population-based screening.

Methods: A systematic review and meta-analysis of risk factors for PD.

Results: The strongest associations with later diagnosis of PD were found for having a first-degree or any relative with PD (odds ratio [OR], 3.23; 95% confidence interval [CI], 2.65-3.93 and OR, 4.45; 95% CI, 3.39-5.83) or any relative with tremor (OR, 2.74; 95% CI, 2.10-3.57), constipation (relative risk [RR], 2.34; 95% CI, 1.55-3.53), or lack of smoking history (current vs never: RR, 0.44; 95% CI, 0.39-0.50), each at least doubling the risk of PD. Further positive significant associations were found for history of anxiety or depression, pesticide exposure, head injury, rural living, beta-blockers, farming occupation, and well-water drinking, and negative significant associations were found for coffee drinking, hypertension, nonsteroidal anti-inflammatory drugs, calcium channel blockers, and alcohol, but not for diabetes mellitus, cancer, oral contraceptive pill use, surgical menopause, hormone replacement therapy, statins, acetaminophen/paracetamol, aspirin, tea drinking, history of general anesthesia, or gastric ulcers. In the systematic review, additional associations included negative associations with raised serum urate, and single studies or studies with conflicting results.

Interpretation: The strongest risk factors associated with later PD diagnosis are having a family history of PD or tremor, a history of constipation, and lack of smoking history. Further factors also but less strongly contribute to risk of PD diagnosis or, as some premotor symptoms, require further standardized studies to demonstrate the magnitude of risk associated with them.