Osteoarthritis is the most frequently encountered form of arthritis worldwide. Pharmacological therapies consisting of nonsteroidal anti-inflammatory drugs (NSAIDs) and analgesics such as paracetamol (acetaminophen), codeine and dextropropoxyphene are the principal means of symptom control. Osteoarthritis appears to have little impact on the pharmacokinetics of these drugs per se. However, other clinical features commonly associated with patients with osteoarthritis, such as advanced age, obesity or concurrent diseases and medications, may be important. A limited number of concentration-response studies have demonstrated the existence of a pharmacokinetic/pharmacodynamic relationship for these drugs. However, the information derived from patients with osteoarthritis is very limited. Simple analgesics have been shown to be as effective as NSAIDs in some studies. Furthermore, the response does not appear to be able to be predicted on the basis of clinical features of inflammation. It appears that a defined concentration-toxicity relationship exists and should be considered when dosage regimens for patients are formulated. Symptoms occur with a circadian rhythm, and similarly, there is circadian variation in the pharmacokinetic/pharmacodynamic response of a number of NSAIDs in patients with osteoarthritis. Selection of the optimal drug formulation will also be influenced by the need for continuous or intermittent therapy in these patients.