Chronic joint pain affects physical well being and can lead to severe psychological and social problems, therefore successful long-term management is highly sought-after. No current behavioural measures of pain used in pre-clinical models mimic the clinical dolorimeter, which provides an objective measure of joint hypersensitivity. In this study we aim to use a novel behavioural readout alongside an established measure to mimic the multifactorial measurements taken in the clinic. Using the pressure application measurement (PAM) device a gradually increasing squeeze was applied across the knee joint of rats until the animal gave an indication of pain or discomfort. PAM and the incapacitance tester were used to detect joint hypersensitivity in a well-established rodent model of adjuvant-induced arthritis. Subsequently, the analgesic effects of prednisolone (1, 3 or 10 mg kg(-1)), morphine (3 mg kg(-1)) and celecoxib (15 mg kg(-1)) were assessed. Both PAM and the incapacitance tester detected a reversal of hypersensitivity 1h post-drug administration. Furthermore, the two readouts were highly correlated, and power analysis indicated that PAM was highly reproducible. In conclusion, PAM provides a novel, accurate behavioural tool for detecting a primary mechanical hypersensitivity in a rat model of chronic inflammatory joint pain.