The results of 940 patients treated with diclofenac for 3 to 24 months in comparative and non-comparative trials are presented. Maximal improvement tended to occur in the first 3 to 6 months of treatment and was generally maintained. Diclofenac was at least as effective as equivalent doses of indomethacin and naproxen and, when treatment lasts more than 3 months, may be more effective. The majority of patients reporting unwanted effects or discontinuing treatment did so in the first 6 months. Unwanted effects (similar to those in short-term trials) were mainly gastrointestinal. Central nervous system, cardiovascular and dermatological side-effects were reported in 1% or less of patients. The long-term laboratory tolerability of diclofenac was good, with no changes in the nature, frequency or severity of abnormal tests with increasing duration of treatment. During the development of diclofenac sodium (Voltarol) various types of long-term investigation were conducted (Table I). This paper presents the results, covering a total of 940 patients treated for 3-24 months, and discusses their significance.