Objectives: To conduct a systematic review of specific opioid (morphine, oxycodone, hydromorphone) pain trials in adult cancer patients in order to 1. evaluate the general methodological quality of randomized, controlled trials of opioids in cancer pain 2. identify factors related to poor methodological quality 3. suggest standardized clinical models which may be used to study pain treatment in palliative care.
Methods: Randomized trials, described as double-blind and having either placebo or active controls will be included. A comprehensive search of electronic databases including MEDLINE, EMBASE and the Cochrane Library will be performed. Included studies will be assessed for quality using the Oxford Quality Scale and for validity using the Oxford Pain Validity Scale.
Results: A preliminary search of PubMed gave 36 eligible trials (pure opioid trials). Six were excluded because they were not described as double-blind. Of the 30 randomized, double-blind trials, 23 had crossover, and 7 parallel group design. Ten studies had a group size of 20 or less. Nineteen studies had a duration of more than one week. Twenty four studies had a withdrawal/dropout/non-evaluable rate over 10%, with 7 studies exceeding 30%. Preliminary conclusions Conducting scientifically sound trials in a population of terminally ill cancer patients is a considerable challenge. There are a number of aspects to be considered when designing opioid studies, including study sensitivity, duration of study and adverse events due to drugs or disease.