Chronopharmacologic studies of indomethacin have indicated that time of dosing influences tolerance and effectiveness. A double-blind, crossover chronotherapeutic trial was undertaken in 66 subjects with osteoarthritis of the hip or knee who were treated with an indomethacin sustained-release (ISR) oral preparation once a day. Varying the ISR dosing time resulted in a quadrupling of tolerance and a doubling of analgesic effectiveness. Three dosing times (8 a.m., noon, and 8 p.m.), each tested during 1-wk spans and randomized for sequencing, were compared in each subject. Subjects self-rated pain intensity every other hour before (1 to 2 days) and during each week of treatment. Morning dosing was associated with a 32% incidence of undesirable effects, whereas the comparable rate was 7% for evening dosing. Ninety-five percent of the subjects reported increased drug effectiveness with a change in ISR ingestion time. The time of dosing that resulted in optimal effectiveness differed among subjects. This was explainable by large interindividual differences in the circadian variation of self-rated pain intensity. Evening dosing was most effective in subjects with predominantly nocturnal or morning pain; conversely, morning or noon dosing was most effective in subjects with greater afternoon or evening pain. The differences that resulted from varying the timing of the identical ISR dose in the same subject greatly exceeded those reported for other nonsteroidal anti-inflammatory drugs.