The emergency physician treats many patients with mild to moderate pain due to musculoskeletal injury. The physician must consider the extent of injury, the patient's medication history, and the potential for abuse when prescribing an oral analgesic. A study was designed to compare the efficacy of two oral analgesics, one containing a narcotic and one nonnarcotic, in relieving mild to moderate pain associated with grade 2 ankle sprain. Forty patients were enrolled--all with moderate pain--and were randomly allocated to treatment with either diflunisal or acetaminophen with codeine. Both analgesic agents were equally effective in relieving the pain. Side effects were experienced by six patients, all of whom were receiving acetaminophen with codeine; none of the patients given diflunisal noted side effects. Global assessments of the efficacy and tolerability of the study drugs showed that 89% of 19 patients given diflunisal and 43% of 21 patients given acetaminophen with codeine considered their respective analgesics excellent or very good.