This double-blind study compared the analgesic effects of morphine with those of butorphanol in patients who presented with pain because of sickle cell crisis. Patients were placed at bed rest, administered IV hydration, and randomly assigned on each visit to receive either 2 mg IM butorphanol or 6 mg IM morphine every 30 to 60 minutes as needed to produce a pain intensity of 50 mm or less on the linear analog pain scale until the patient was discharged. Linear analog scale for pain and pain relief, level of alertness, and vital signs were assessed at 60 and 120 minutes after each study drug dose, before additional doses, and at discharge. Eighteen patients (12 men, six women; mean age, 29.3 +/- 7.7 years) were studied. Six received only morphine, six received only butorphanol, and six received each treatment at some time during the study period, resulting in 45 randomizations to treatment. The two therapies did not differ significantly (P greater than .40) with respect to pain or relief of pain scores, level of alertness, or vital signs. The discharge rate was 69.6% and 68.2% with morphine and butorphanol, respectively (P = .92). The incidence of adverse effects was 13% and 23% with morphine and butorphanol, respectively (P = .46). We conclude that morphine and butorphanol are equally effective in the treatment of sickle cell crisis pain.