Abstract P4-11-05: Opioid prescriptions in the military health system breast cancer population, FY2007 – FY2014

Abstract

Abstract Purpose: Opioid use among individuals with breast cancer is a safety concern because of adverse events including misuse, overdose, and death. To date, there are two published U.S. population-based studies of opioid use among breast cancer patients, both which address persistent opioid use after surgery. Here, we examine rates, amount, and cost of opioids prescribed to breast cancer patients over an 8-year period. Methods: We used Military Health System (MHS) outpatient, inpatient, and pharmacy claims data to select beneficiaries with a primary diagnoses of breast cancer, fiscal years (FY) 2007- 2014. Logistic regression models were used to identify predictors of having received > 1 opioid prescription within a year, and having received > 4 filled opioid prescriptions within a year. Simple linear regression and multiple regression were used to examine trends and predictors of reimbursed opioid cost. Results: Among 25,500 non-elderly breast cancer patients treated in the MHS each year, most were age 55-64, located in the southern U.S., and treated in civilian facilities only. On average per year, 75% had surgery, 90% radiation therapy, 19% chemotherapy, and 8% hospice care. Annual rates of receiving > 1 opiate prescription were stable across 8 years, 48.1% (FY2007) to 50.5% (FY2012 and FY2013). The annual number of opioid prescriptions per patient was also stable, 4.26 (FY2014) to 4.39 (FY2007). Average reimbursed cost of opioids steadily increased over time, $135 (FY2007) to $260 (FY2014) (p < 0.001). Based on regression models, the strongest predictors (p < 0.001) of having received > 1 opioid prescription were (in order of statistical significance): surgery, chemotherapy, civilian care only or mixed use civilian and military care, number of physical comorbid conditions, hospice use, and depression. The strongest predictors (p < 0.001) of > 4 opioid prescriptions per year were: depression, hospice use, number of physical comorbidities, and chemotherapy, followed by anxiety disorder, alcohol use disorder, and drug use disorder. The strongest predictors (p < 0.001) of high reimbursed opioid cost were: drug use disorder, hospice use, depression, civilian care only or mixed use civilian and military care, alcohol use disorder, number of physical comorbid conditions, receipt of chemotherapy, fiscal year, and anxiety disorder. Discussion: Half of individuals with a primary breast cancer diagnosis receive > 1 opioid prescription per year. Four overlapping constructs predict opioid prescriptions and cost: treatment modalities, physical and mental health comorbidities, severity of illness, and system of care. Increased cost over time coincides with current U.S. population reports suggesting rising daily opioid dosage. Despite limitations with claims data, the results suggest that opioid use among breast cancer patients need to be monitored and accompanied by documented clinical management plans. We recommend that oncology providers implement risk reduction strategies including screening for history of substance use and mental health comorbidities, and implementing guidelines specific for cancer patients such as those found in the Veterans Administration/Department of Defense Clinical Practice Guidelines for Opioid Therapy for Chronic Pain, v. 3.0. Citation Format: Jeffery DD, Hopkins L, Spevak CJ, Burke HB. Opioid prescriptions in the military health system breast cancer population, FY2007 – FY2014 [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-11-05.