Background: No core set of measurement tools exists to collect data within clinical practice. Such data could be useful as reference data to guide treatment decisions and to compare patient characteristics or treatment results within specific treatment settings. Methods: The Dutch Dataset Pain Rehabilitation was developed which included the six domains of the IMMPACT core set and three new domains relevant in the field of rehabilitation (medical consumption, patient-specific goals and activities/participation). Between 2010 and 2013 the core set was implemented in 32 rehabilitation facilities throughout the Netherlands. Results: A total of 8200 adult patients with chronic pain completed the core set at first consultation with the rehabilitation physician. Adult patients (18-90 years) suffering from a long history of pain (38% >5 years) were referred. Patients had high medical consumption and less than half were working. Although patients were referred with diagnosis of low back pain or neck or shoulder pain, a large group (85%) had multisite pain (39% 2-5 painful body regions; 46% >5 painful body regions). Scores on psychosocial questionnaires were high, indicating high case complexity of referred patients. Reference data for subgroups based on gender, pain severity, pain locations and on pain duration are presented. Conclusions: The data from this clinical core set can be used to compare patient characteristics of patients of other treatment setting and/or scientific publications. As treatment success might depend on case complexity, which is high in the referred patients, the advantages of earlier referral to comprehensive multidisciplinary treatment were discussed. Significance: A detailed description of case complexity of patients with chronic pain referred for pain rehabilitation. Insight in case complexity of patients within subgroups on the basis of gender, pain duration, pain severity and pain location. These descriptions can be used as reference data for daily practice in the field of pain ...