National trends in opioid-related hospitalizations among patients with cancer.
199 Background: Cancer patients are routinely prescribed opioids for cancer-related pain. With recent attention to the opioid epidemic, we sought to identify risk factors and to describe the incidence of opioid-related hospitalizations among cancer patients. Methods: Serial cross-sectional study of adult cancer patients with opioid-related hospitalizations using the National Inpatient Sample (NIS) database from January 2006 to December 2014. We identified cancer patients using the International Classification of Diseases, Ninth Revision (ICD-9) codes. We defined opioid-related hospitalizations as ICD-9 codes for heroin poisoning, opioid poisoning, or opioid dependence or abuse in the primary diagnosis field. A logistic regression model identified predictors for opioid-related hospitalizations. We adjusted temporal trends for opioid-related hospitalizations for all-cause hospitalizations among cancer patients. Results: Among 25,443,362 hospitalizations for cancer patients, there were 14,336 opioid-related hospitalizations. Non-heroin opioid poisoning made up 88% of opioid-related hospitalizations. Predictors for opioid-related hospitalizations for cancer patients included drug abuse (OR 9.40, 95% CI 8.28 - 10.66), younger age [age 18 - 29 (OR 4.00, 95% CI 3.10 - 5.17); age 30 - 49 (OR 3.99, 95% CI 3.43 - 4.65)], depression (OR 2.17, 95% CI 1.97 - 2.39), alcohol abuse (OR 1.21, 95% CI 1.03 - 1.41), and year of hospitalization [2009-2011 (OR 1.19; 95% CI 1.07 - 1.32); 2012 - 2014 (OR 1.19; 95% CI, 1.06 - 1.32)]. On average, opioid-related hospitalizations increased by 0.003% per year (p = 0.002). Conclusions: Opioid-related hospitalizations among cancer patients are rare, appear to be increasing over time, and are largely due to non-heroin opioid poisoning. Standardized opioid risk screening based on validated predictors may identify cancer patients with the greatest risk of an opioid-related hospitalization.