Opioids as the earliest acute treatment for emergency department headache and short‐term all‐cause emergency department return: A two‐center comparative cohort study

OBJECTIVE: To compare first-line emergency department (ED) treatment classes for acute headache on short-term all-cause ED return. BACKGROUND: In-ED pain relief anchors acute headache trials but is recorded incompletely in routine care. All-cause ED return has not been used to compare first-line treatments; guidelines favor dopamine-receptor antagonists over opioids for acute migraine. METHODS: Retrospective two-center cohort of adults treated for headache in the ED (MIMIC-IV-ED, 2011-2019; MC-MED, 2020-2022); first-line treatment was the earliest qualifying acute agent. The primary contrast was opioids versus guideline-preferred dopamine-receptor antagonists. The primary outcome was 72-h all-cause ED return among discharged patients; 7-day return was secondary. Estimates, adjusted for prespecified confounders available in both databases, are reported as risk ratios (RRs) with 95% confidence intervals (CIs), pooled and site-specific. RESULTS: Among 13,285 treated adults (10,799 MIMIC-IV-ED; 2486 MC-MED), MIMIC-IV-ED opioid recipients were older than dopamine-antagonist recipients (median 48 vs. 39 years) and more often high-acuity at triage (980/1895 [51.7%] vs. 900/3056 [29.5%]). First-line opioids were associated with higher 72-h all-cause ED return (adjusted risks 6.8% vs. 3.8%; RR 1.78; 95% CI 1.31-2.32) and 7-day return (10.7% vs. 6.6%; RR 1.62; 95% CI 1.28-1.99). The MC-MED estimate was directionally concordant but anchors direction only (85 opioid-treated discharged patients). Restricting to a headache-coded or headache-triaged return attenuated the association, which was no longer statistically significant (72-h RR 1.31; 95% CI 0.91-1.88). CONCLUSION: Patients selected for first-line opioids were older and higher acuity and returned more often within 72 h and 7 days at both health systems. These risk-adjusted associations, conditional on discharge, reflect all-cause ED utilization rather than confirmed headache recurrence and do not establish that changing first-line class would change return rates. The attenuated headache-specific result is as consistent with administrative undercoding as with no association and is not a justification for prescribing opioids.