Opioids as the First ER Treatment for Headache: What a 13,285-Person Study Found

Why This Study Was Done

Headache is one of the most common reasons people go to the emergency department. Guidelines from the American Headache Society advise against opioids for migraine in the emergency department and favor dopamine-receptor antagonists such as metoclopramide (Reglan) and prochlorperazine (Compazine). Our summary of the emergency department migraine treatment guidelines covers those recommendations in full.

Opioids are still given in many emergency departments. Most studies of emergency headache treatment measure pain relief during the visit, which is often recorded incompletely in routine care. A new study in Headache looked at a different, easier-to-measure outcome: whether people came back to the emergency department within a few days.

How the Study Was Done

This was a retrospective cohort study using two large emergency department databases. The first, MIMIC-IV-ED, covered visits from 2011 to 2019 at Beth Israel Deaconess Medical Center in Boston. The second, MC-MED, covered 2020 to 2022 at a second US health system. Adults treated for headache were grouped by the first acute medication they received.

The main comparison was opioids versus the dopamine-receptor antagonists that guidelines prefer. The main outcome was a return to the emergency department for any reason within 72 hours among people who were sent home. Return within 7 days was a secondary outcome. Results were adjusted for factors measured in both databases.

What They Found

People who were given opioids first were not the same as people given dopamine-receptor antagonists first. In the larger database, opioid recipients were older (median age 48 vs 39) and more often triaged as high acuity (51.7% vs 29.5%).

People given opioids first were about 1.8 times as likely to return within three days. A risk ratio of 1.78 here means roughly 3 extra returns for every 100 people sent home.

The smaller database pointed in the same direction, but it included only 85 discharged patients who received opioids first, so the authors used it to confirm the direction of the result rather than its size.

The headache-specific result

When the researchers counted only return visits coded or triaged as headache, the difference shrank and was no longer statistically significant. The authors note this could mean there is no headache-specific difference, or it could reflect return visits that were about headache but were coded as something else. They state plainly that this result is not a reason to prescribe opioids.

How to Read These Results

This study shows an association, not cause and effect. Doctors chose opioids more often for older and sicker patients, and that difference could explain part of the extra returns even after adjustment. This is the problem of confounding, and the authors say directly that their findings do not prove that changing the first treatment would change return rates.

The results still line up with the wider evidence against opioids for migraine. Our article on a 12,612-person survey of opioid use in migraine found opioid use was linked to more disability and more reported thinking problems. Opioids also carry a high risk of medication overuse headache, which our medication overuse headache checker can help you screen for. And in our article on how migraine is treated after diagnosis in the US, nearly half of newly diagnosed people filled an opioid prescription within three years.

Limitations

• The study was retrospective and based on records, so the reason each patient received a given drug is unknown.
• Opioid recipients were older and higher acuity, and some differences between the groups may not have been fully adjusted for.
• Only people who were sent home were included in the return analysis.
• Returns to a different hospital would not have been captured.
• Return visits were counted for any reason, so not all of them were about headache.
• The second database contributed few opioid-treated patients.

Key Takeaways