The aim of this study was to use both quantitative and qualitative data to assess pain in idiopathic Parkinson's disease (IPD) and classify it according to cause. In this descriptive cross-sectional study, 123 patients were interviewed and pain was assessed using patient descriptions, the Brief Pain Inventory, and Visual Analogue Scales. Patients reported 285 pains (median 2 per patient), and 22.8% had 4 or more pains. Pain was reported as a problem in 85% and was IPD-related in 62.6% of patients, unrelated to IPD in 64.2%, indirectly related to IPD in 8.1%, related to multiple causes in 4.1%, and treatment related in 0.8%. Pain unrelated to IPD was more common, more constant, and more severe than IPD-related pain. Overall, analgesic use was low. This study demonstrates the benefits of adopting a systematic and logical approach to the assessment of pain in IPD and the merits of a simple new classification system.