This prospective, open-label, before-after trial was designed to compare the efficacy of oral opioids with that of transdermal fentanyl in severe AIDS-related chronic pain, as well as assess barriers, patient satisfaction, and side effects. Thirty-five sequentially selected male and female outpatients with AIDS who were at least 18 years old were enrolled. All had chronic pain requiring continuous treatment with > or = 45 mg/day oral morphine or an equivalent. Eighteen of the patients had a history of chemical dependency. Baseline data were collected while patients received their previously prescribed opioid; assessments were made again after a stable transdermal fentanyl dose (25--300 microg/h) had been maintained for 15 days. Patients completed the International Association for the Study of Pain Classification of Chronic Pain Syndromes questionnaire, the Brief Pain Inventory, and a Satisfaction With Pain Medication questionnaire for assessing pain intensity, relief, and interference with normal functioning. With transdermal fentanyl, pain severity scores decreased significantly, mean pain relief scores increased, and daily functioning measures improved significantly. Most adverse events were mild and unrelated to fentanyl use. Transdermal fentanyl was effective for chronic pain in both chemically dependent and non-chemically dependent patients with AIDS.