The physical performance of chronic pain patients is of major concern both for their assessment and for treatment evaluation. However, there are few widely used physical tests, a shortage of reliability and validity data on published tests, and an over-reliance on self-report or on clinical measures of dubious generalisability. A set of tests was designed to cover speed and endurance in walking, stair climbing, standing up from a chair, sit-ups, arm endurance, grip strength, and peak flow. Standard instructions and testing conditions were used by a trained tester on a population of chronic pain patients before and after a cognitive-behavioural chronic pain management programme. Reliability, validity, and acceptability of each test was examined, and recommendations made for their relative utility.