Patterns, predictors, and impacts of opioid use among migraine patients at the emergency departments

Migraine is a primary headache disorder associated with high health resource utilization and poor quality of life. There is limited evidence on opioid use among migraineurs, specifically at emergency departments (EDs). The purposes of this study were 1) to understand the treatment patterns, health resource utilization (HRU), demographics, and clinical characteristics of patients with migraine presenting to the ED; 2) to examine potential risk factors associated with opioid use at the ED; 3) to assess the impact of opioid use on future ED visits. This was a retrospective study using Electronic Medical Record (EMR) data from Baylor Scott & White Health Plan (BSWHP) from December 2013 to April 2017. The index date was defined as the first migraine-related ED visit after >= 6 months of initial enrollment. Patients aged >=18 with a diagnosis for migraine who had at least 6 months of continuous enrollment before and after the index date were included in the study. Descriptive statistics were used to describe and compare HRU rates and comorbidities between opioid users and non-opioid users. Stepwise logistic regression was used to identify predictors of opioid use at the index ED visits. Cox proportional hazard regression was used to compare the risk of future ED visits. A total of 788 patients were included in the study, with an average age of 44.5 (±14.6) years, 85.9% female, and 76.1% Whites. More than one-third (n=283, 35.9%) of the patients were identified as opioid users at the index ED visit. Pre-index migraine-related opioid use (2-4 RXs OR=1.66; 5-9 RXs OR=2.12; >=10 RXs OR=4.43), pre-index nonmigraine-related opioid use (>=10 RXs OR=1.93), pre-index ED visits (1-3 visits OR=1.84), age (45-64 years OR=1.45), and sleep disorder diagnosis (OR=1.43) were all significant (p