The aim of this study was to determine the association between the Oxford knee score (OKS) and direct assessment of outcome, and to examine how this relationship varied at different time-points following total knee replacement (TKR). Prospective data consisting of the OKS, numerical rating scales for 'worst pain' and 'perceived mean daily pain', timed functional assessments (chair rising, stairs and walking ability), goniometry and lower limb power were recorded for 183 patients pre-operatively and at six, 26 and 52 weeks post-operatively. The OKS was influenced primarily by the patient's level of pain rather than objective functional assessments. The relationship between report of outcome and direct assessment changed over time: R(2) = 35% pre-operatively, 44% at six weeks, 57% at 26 weeks and 62% at 52 weeks. The relationship between assessment of performance and report of performance improved as the patient's report of pain diminished, suggesting that patients' reporting of functional outcome after TKR is influenced more by their pain level than their ability to accomplish tasks.