The extensive use of sedative and analgesic medication at the end of life is often controversial due to the perception that death may be hastened as a result of progressive drug escalation. Physician attitudes toward prescribing medication in this setting vary, resulting in widely differing prescribing patterns for patients with advanced cancer. This investigation attempted to identify overall prescribing patterns and variation in the use of sedation and analgesia in an inpatient hospice setting at the end of life. A retrospective case review was undertaken of 102 consecutive patients who died in a palliative care hospice. A detailed review of medication prescription, with particular attention to sedation and analgesia in the last week of life, was performed. The review revealed that regular sedation was prescribed in 68 percent of the patients. Almost two-thirds of the patients began regular sedation on admission or within seven days of admission. Although survival was higher in patients who received regular sedation (mean, 36.5 days) versus those that did not (mean, 17 days), the difference was not significant (p = 0.1). Overall, regular sedation with moderate dose increases was observed. In patients prescribed morphine from the time of admission, morphine oral equivalents increased from a mean of 111 mg on admission to a mean of 346 mg at time of death for a mean escalation of 311 percent. The mean duration of admission was 26 days with an opioid-escalation index of 12 percent per day. Survival is a multifactorial phenomenon and was unrelated to the level of analgesia in this cohort. Findings showed that sedation dose increased modestly toward the end of life, and that the increase was not associated with a significant reduction in survival. Further there was no significant impact on survival related to an individual physician's prescribing pattern at the end of life. These results suggest that, in the institution where the review was conducted, neither sedation nor individual variation in physician prescribing habits in terminally ill patients was associated with hastening of death. Overall, the amount of sedative drugs required for adequate symptom control during terminal care was moderate.