Need for Guidelines on Migraine in Emergency Department Settings

Anne Kelly, MD

This is a 2-part interview. For part 1, click here.

On top of the daily array of assignments, physicians in the emergency department (ED) are also often faced with patients who present with nontraumatic headache. According to a recently published international, multicenter, observational study, diagnosing and managing these patients is a challenge, with more than 30 different diagnoses recorded in a cohort of 4536 individuals. The results also showed that hospitals across different countries were inconsistent with how they treated patients and their use of neuroimaging tools such as CT scans.

These data could serve as a reminder for institutions to self-evaluate themselves in how they investigate and treat nontraumatic headache, according to lead author Anne-Maree Kelly, MD. Kelly, who is the director of the Joseph Epstein Center for Emergency Department Research, and her colleagues concluded that the findings highlight the need for additional research, as well as consensus guidelines on the best practices.

In an interview with NeurologyLive®, Kelly discussed the necessary steps to create unform change across EDs for the betterment of patient care. She also stressed the need for guidelines to be implemented on both a national and international level, with the sole focus that some countries do not have the same capabilities as others, and thus their recommendations should be slightly altered.

NeurologyLive®: Where do the gaps in diagnosing and managing headache in the ED setting lie? Are these things that can be fixed with education?

Anne-Maree Kelly, MD: We didn’t look at education as part of this research, but we also know that there’s wide variation in experience and training of doctors working in emergency departments around the world. In an US emergency department, you have mostly specialists. In other parts of the world, there might be a specialist within 10 miles and there might not be a specialist in the department. And a doctor might only be working in an emergency department for a few months, so although education is important, it’s more important to have pathways and guidelines. That helps everybody be consistent.

It’s also hard to know how much education is enough because in the emergency department, we’re dealing with everything from trauma to pregnancy, and everything else. If we knew enough about everything, I think my brain would explode. The point of education is a good one, but education can only focus on the top notes, but pathways can help consistency with respect to investigation and drug treatments. One, avoid opiates and give better migraine medications, and better identify the high-risk patients.

Who needs to come to the table to progress consensus guidelines?

There are obviously national bodies. In the US, the American College of Emergency Physicians, American Headache Society, and Neurology Society are obvious players, and to a lesser extent, the neurosurgeons, because a small proportion of these are neurosurgical. Getting people in a room, however, is not easy. In other parts of the world, the groups are not so easy to identify. But I do think that in some countries they have what’s called “excellence commissions,” which are nonspecialty-based bodies, like the National Institute for Clinical Excellence in the United Kingdom.

There’s an institute here in New South Wales, in Australia, that specializes in trying to bring these people together to come up with simple guidelines that can be accessed from any emergency center or general practice. They give guidance about who should be imaged, who shouldn’t, who should be going to the hospital, who shouldn’t, and what the basic treatments are. I think it’s a bit context specific, but it would be good if all the leaders in all of the disciplines made themselves available to those bodies. Additionally, have a commitment to coming up with consensus guidelines rather than being fixed in positions about what they think is the best. A bit of compromise and acceptance that the contexts are different, and that not everybody can have a CT scan within 10 minutes. Especially for headache rather than stroke. That’s what’s needed, and there will be an element of national, but I think that national approaches can be really effective as well.

What types of studies need to be conducted to fully capture the variance in data?

First, demonstrating variation in practice is a first step in improving practice. If you don’t know where you are, you don’t know where you can go. The logical follow-up question is, “why is there variation? Does it matter?” To do the why part of it, we would need to have data on the reasoning of the clinician, ideally collected prospectively and throughout the clinical encounter, so that there’s blinding to the outcome. Why did my gut feeling, let’s just say as a multi-year emergency physician, respond that way about a patient? Why did I go with a CT scan, even though they didn’t have any obvious red flags? What was my reasoning? And then try to work out what works and what doesn’t.

The "does it matter” question is much more complicated. First, does improved diagnostic accuracy improve more of the patient outcome or the patient experience? What about the downstream impacts like cost or service delivery? For those things, we would again, need some prospective studies, with a follow-up period. What we couldn’t collect was, the things we missed. How many misdiagnoses did we make? How could we have avoided that? How could we improve the sensitivity of our process?

We also have to do some work on whether the red flags are validated in the emergency department. We’re hoping our data set, which is now about 5000 patients strong, should be able to do that. For example, we’ve shown that severity of headache is not helpful, it doesn’t predict a serious cause. We need to be able to go, “yes, that works,” and “No, that doesn’t. We’ve done an analysis which showed that age has a big impact and are three times more likely to have a serious cause. We’re hoping to build on the data we have and then encourage spin off studies by our own group and by anybody else who’s interested. We can only do what we can do. We’re hoping that this will engage with the neurology community, and we can partner in some studies, not just looking at ED or neurology, but the whole patient journey.

Could seniority and years of experience play a factor into the inconsistencies seen within the data?

I think that it’s possible. We purposely didn’t collect that in this dataset. We wanted to make it easy to have just patient data. We didn’t want there to be any possibility of any group being singled out as problematic or different. We wanted to collect the basic data to raise the questions. Our study is about raising those questions and saying, “we accept we can do better. How can we go forward?” I think that’s an issue and goes again to the need of pathways or processes to support people who are only there for a short period of time.

I don’t know about the system in America, but a lot of places in the world, including my country, where there are health care workers that are non-medical, non-nursing people in remote places. Having a system that empowers them with information to know who is potentially seriously ill, who they need to get out of there for further investigation, etc., is important in some countries with lower densities of doctors and nurses than what is present in the US, for example.

The variation in healthcare access around the world is something we haven’t accepted well. Same for how we can maximize outcomes for everybody, not just countries with high access to medical care. Our study included countries like Turkey and Romania, which have quite different health systems and access to care.

Transcript edited for clarity.

REFERENCE

Kelly AM, Kuan WS, Chu KH, et al. Epidemiology, investigation, management, and outcome of headache in emergency departments (HEAD study)–a multinational observational study. Headache. 2021;61(10):1539-1552. doi:10.1111/head.14230