Preceding pain symptoms and Parkinson's disease: a nationwide population-based cohort study.

BACKGROUND AND PURPOSE: Painful sensations are recently reported to be a non-motor feature of Parkinson's disease (PD). The non-steroidal anti-inflammatory drug ibuprofen is a common painkiller and was reported to be associated with a decreased risk of PD. The aim of the present study was to examine the relationship amongst preceding pain symptoms, use of ibuprofen and risk of PD in a nationwide population-based cohort. METHODS: The data of participants who were free of PD at baseline were obtained from two large National Health Interview Surveys (NHIS) in Taiwan, conducted in 2001 and 2005. The information regarding pain status included severity and location of pain. Information regarding pain status, use of ibuprofen, comorbidity of depression and PD-associated risk/protective behaviors was adjusted using proportional hazards models. RESULTS: Amongst 33 388 participants, 32 cases of incident PD were identified after a mean follow-up of 3 years. After adjusting for the use of ibuprofen and other PD risk factors, subjects with preceding pain symptoms had a higher incidence of PD than those without pain at baseline, and the hazard ratio was 1.79 (95% CI: 0.71-4.51, P = 0.21) for mild pain and 2.88 (95% CI: 1.05-7.86, P = 0.04) for moderate or severe pain. The PD risk increased by 34% with each additional increment in pain score [hazard ratio = 1.34 (1.03-1.75), P = 0.03], showing a dose-response relationship. CONCLUSIONS: These findings support the hypothesis that pain is associated with PD in the pre-motor stage of the disease. Further research is needed to clarify the role of sensory system involvement in the pre-motor phase of PD.