Racial and ethnic disparities in analgesic prescribing for migraine in U.S. emergency departments

Background: Treatment for migraine exacerbations is often initiated in the emergency department (ED). Racial/ethnic disparities in analgesic prescribing in EDs have been documented for conditions with more objective findings (e.g., long bone fracture), but less is known about conditions such as migraine with few objective findings. Furthermore, disparities in treatment decisions for pain may not be the same for younger and older migraine patients.

Research Objective: To examine racial and ethnic disparities in analgesic prescribing for migraine in U.S. emergency departments, and to explore the effects of patient age on disparities.

Study Design: Data were extracted from the 2007-2010 National Hospital Ambulatory Medical Care Survey (NHAMCS), a nationally representative sample of ED visits. ED visits by adults (18+) with a primary diagnosis of migraine (ICD9:346) and no indication of injury were included. Dependent variables were the prescribing of (1) any analgesic, (2) narcotics, (3) NSAIDs, (4) migraine medications, or (5) other analgesics. Race/ethnicity was classified as non-Hispanic white, non-Hispanic black, Hispanic, or other race. Age was classified as 18-39, 40-65, and 65 years and older. Covariates included sex, insurance, level of immediacy, pulse, pain scale, region of country, and rurality. Survey-weighted logistic regression models adjusting for all covariates were conducted for each dependent variable with standard errors adjusted for clustered sampling.

Results: Our final sample was comprised of 1,480 adults representing approximately 5.6 million (1.4 million annually) U.S. ED patients with migraine. About 79% of these patients received any analgesic, 50% received a narcotic, 4% received a migraine medication, 33% received an NSAID, and 7% received another type of analgesic. Additionally, 22% of patients received pain medications from multiple classes. Compared to whites, African Americans were half as likely to receive any analgesic medication (Odds Ratio [OR] = 0.52, p<0.05) and were less than half as likely to receive a narcotic (OR=0.46, p<0.05). The disparity in migraine medications approached statistical significance (OR = 0.40, p=0.06). The disparity was not observed for NSAIDs or other analgesics. Stratified by age group, the observed racial disparity existed primarily in prescribing to younger (18-39) migraine patients and was not significant for older migraine patients.

Conclusions: Results suggest that disparities exist in the types of analgesics used in emergency department visits for patients presenting with migraine. This disparity is primarily driven by disparities in narcotic use in treating migraines and exists in mostly younger minority patients.