A DESCRIPTIVE ANALYSIS OF LONG‐ACTING OPIOID USE IN END‐STAGE CANCER PATIENTS: CORRELATION OF AGE, GENDER, AND PRIMARY SITE WITH DOSE

Susannah Hall, ExcelleRx, Inc, Rollin M. Gallagher, MCP Hahnemann School of Medicine, Calvin Knowlton, ExcelleRx, Inc, and Douglas Weschules, ExcelleRx, Inc. Long‐acting opioids are considered to be the cornerstone of cancer pain treatment, especially in the hospice patient. Increasing age has been associated with decreased opioid requirements in cancer and post‐operative patients, although published reports do not agree about the age at which this decreased requirement appears. In terminal cancer patients, there are mixed reports as to whether an association between gender and opioid dose exists. Possibly the presence of relatively more or less painful gender‐specific cancers contributed to those differing results. Some studies suggest that certain primary cancer sites require higher opioid doses than others, but few have controlled for age. The objective of this retrospective analysis is to describe the correlation of age, gender, and primary cancer site with long‐acting opioid doses prescribed for hospice patients at the end of life, defined as the last long‐acting prescription written prior to death. All hospice patients admitted to a North American palliative care specialty pharmacy between 4/1/00 and 3/31/01 with the primary diagnosis of cancer and who were prescribed transdermal fentanyl, sustained release oral morphine, or sustained release oxycodone were included in this analysis. A total of 7201 patients met the inclusion criteria. Age was a highly significant determinate of dose (p ≤ 0.0001) with patients prescribed increasingly lower long‐acting opioid doses with each decade of life. The final morphine equivalent dose of long‐acting opioid prescribed prior to death was ≤ 60 mg/day in 25.7% of patients 40–49 years, 29.1% of patients 50–59 years, increasing to 57.8% of patients 80–89 years, and 69.6% of patients ≥ 90 year. The sexes were compared for lung, colorectal, and pancreatic cancer separately and no significant difference between the sexes was found by Chi‐square test, nor by logistic regression with age controlled. The correlation of the primary site of cancer with the final prescribed long‐acting opioid dose will be presented.