This randomized, double-blind, multi-centre study was undertaken to evaluate the efficacy and safety of treatment for 4 weeks with codeine plus paracetamol versus paracetamol in relieving chronic pain due to osteoarthritis of the hip. A total of 158 outclinic patients entered the study. Eighty-three patients (mean age 66 years) were treated with codeine 60 mg plus paracetamol 1 g 3 times daily, and 75 patients (mean age 67 years) with paracetamol 1 g 3 times daily. Ibuprofen 400 mg was prescribed as rescue medication. Due to an unexpected high rate of adverse drug reactions, the study was closed before the planned 400 patients had entered. Over weeks 1-4, 87%, 64%, 61% and 52% of patients in the codeine plus paracetamol group, and 38%, 31%, 22% and 29% of patients in the paracetamol group had one or more adverse drug reactions. Significantly more patients in the codeine plus paracetamol group had adverse drug reactions in each of the 4 weeks. Nausea, dizziness, vomiting and constipation were predominant adverse reactions in the codeine plus paracetamol group. During the first week of treatment, 30 patients (36%) in the codeine plus paracetamol group and 9 (12%) in the paracetamol group dropped out. As evaluated from patients completing the first week of treatment, the pain intensity during that week compared to their baseline pain was significantly lower in the codeine plus paracetamol group than in the paracetamol group. Moreover, during the first week the paracetamol group received rescue medicine significantly more frequently. In conclusion, when evaluated after 7 days of treatment, the daily addition of codeine 180 mg to paracetamol 3 g significantly reduced the intensity of chronic pain due to osteoarthritis of the hip joint. However, several adverse drug reactions, mainly of the gastrointestinal tract, and the larger number of patients withdrawing from treatment means that the addition of such doses of codeine cannot be recommended for longer-term treatment of chronic pain in elderly patients.