OBJECTIVES: To compare EQ-5D and SF-6D utilities in patients with chronic pain due to osteoarthritis (OA) of the knee or low back pain (LBP) derived from phase III trials with tapentadol. METHODS: Three phase III trials with identical design included EQ-5D and SF-36 questionnaires to measure quality of life of patients with pain due to OA or LBP treated with tapentadol prolonged release (PR), oxycodone controlled release (CR) or placebo. EQ-5D and SF-6D indices were obtained using the UK weights. The ability of the two utility measurements to discriminate among different health states was tested. RESULTS: Both SF-6D and EQ-5D utility values increased from baseline to endpoint (15 weeks). The increase (mean of all patients with active treatment) was substantially higher when measured with EQ-5D (0.16 vs. 0.06). The EQ-5D better distinguished among health states (different severity of adverse events, pain relief, withdrawal rates). While utilities were very similar in a group of patients who tolerated the treatment (0.695 and 0.694 for EQ-5D and SF-6D, respectively), EQ-5D utilities were considerably lower in patients who withdrew due to adverse events (0.503 and 0.597 for EQ-5D and SF-6D, respectively). A similar pattern was seen in patients with various levels of pain relief. In patients with >30% pain relief mean EQ-5D and SF-6D utility was 0.716 and 0.708, respectively. The EQ-5D utility in patients who withdrew due to lack of efficacy was 0.405, when analyzing the SF-6D utility this resulted in 0.580. CONCLUSIONS: Both generic instruments to measure quality of life, EQ-5D and SF-6D, showed that avoidance of severe treatment-related adverse events and sufficient pain relief has a large beneficial impact on patient’s wellbeing. In the clinical trials analyzed the discriminative power of the EQ-5D was stronger showing that this instrument is a useful tool also in pain studies to analyze patient’s QoL.