Addressing Disparities at Every Step of the Stroke Care Pathway: Luke Messac, MD, PhD

Disparities in access to endovascular thrombectomy (EVT) persist across every step of the stroke care pathway, even within systems designed to promote equity through regionalization and interhospital transfer. In this interview from the International Stroke Conference 2026, Luke Messac, MD, PhD, an attending emergency physician at Brigham and Women’s Hospital, discusses findings from a large, statewide analysis of more than 300,000 ischemic stroke encounters in California between 2015 and 2021.

The study examines how race, ethnicity, and insurance status intersect to influence where patients initially present for care, whether they are transferred to EVT-capable centers, and ultimately whether they receive EVT. Although Medicare eligibility at age 65 offers some protective effect, data reveal that it does not eliminate disparities for racially minoritized patients, and for those under 65, patterns of cumulative disadvantage are stark. Notably, interhospital transfer, often viewed as a key lever for leveling access, failed to mitigate gaps for the most disadvantaged groups, raising urgent questions about how emergency and transfer systems are functioning in practice and where targeted interventions could make the greatest impact.

Transcript

Your data suggest that transfers help level access in later stages of care. What does that say about where systems are succeeding and where they may still be falling short?

We initially thought that the transfer system might be working as intended, but then we revised our abstract to show that due to Medicare eligibility at age 65, things have changed. One of the most concerning things is that there are significant disparities in both presentation to an EVT-capable center, transfer to an EVT-capable center, and receipt of EVT. In patients under 65 [years], these tend to follow patterns of cumulative disadvantage, so patients who are racially minoritized or have undesirable insurance payer status tend to have the worst outcomes. In patients over 65 [years], those who are eligible for Medicare get some protected benefit from having Medicare, but not as much as you would hope, particularly if they are racially minoritized, so there are still patterns of disadvantage, even in insured patients.

Interhospital transfer was a key outcome. What did you learn about how transfer patterns function across different hospital types or regions in California?

One would hope that transfer would help alleviate disparities so that patients who show up to centers without EVT capability would be able to transfer to those with EVT capability. Unfortunately, what we found is that for certain groups of patients, racially minoritized and poorly insured patients, that transfer system doesn't work nearly as well as you would hope. So, particularly for those patients who show up to fewer resource centers, which also follow patterns of disadvantage, they're not getting access to the best quality care that they need.

What role do historical patterns of hospital investment and resource distribution play in the disparities your team observed?

Well, it has a lot to do with it. One thing you do see in the data that we have is that patients present to EVT-capable centers in widely disparate ways. So if you are White or well insured, you have private insurance or Medicare, you are much more likely to show up to an EVT-capable center. If you are not, then you are less likely to show up to an EVT-capable center, and that often has to do with patterns of investment in racially minoritized areas, where patients who come to their closest center with stroke symptoms because of histories of disadvantage have fewer resources and therefore rely on systems of transfer that aren't serving them as well.

What do these findings suggest about where disparities are most likely to emerge along the stroke care pathway?

Well, they seem to be emerging along every stage of the continuum. So in presentation to an EVT-capable center, transfer, and EVT receipt, we see cumulative patterns of disadvantage. The transfer one is particularly disconcerting because that's the one that we have a lot of control over. In the emergency room. I work in the emergency room. A lot of our work is deciding who needs transfer, how quickly they can be transferred, and where they're going to be transferred to, and so seeing that pattern of disadvantage right there is particularly concerning. And that's one that we can intervene on.

How might patient trust in the health care system influence access to timely stroke care across racial and ethnic groups?

One thing we did see is that the people who presented for stroke, regardless of where they presented, tended to have more severe strokes if they were from racially minoritized groups or had poor insurance status. That might be people waiting at home longer or being less certain about coming into the hospital due to concerns about cost or concerns about trust. We don't know that entirely from our data, but that is something that we need to look into further.

How will future innovations in stroke care need to be designed differently to avoid widening existing disparities?

There's been a ton of work done in recent years on stroke transfer patterns, in regionalizing systems of care, and in ensuring that patients who show up to less-resourced transfer centers have access to more resource centers if they need them. But what we're showing is that that is not yet working as intended. That is not working for the most disadvantaged patients, and it follows intersectional patterns of disadvantage, where racially minoritized and low-income patients still face the greatest barriers. And so what we need to do is figure out why those barriers still persist and what we can do to make sure that the system works for everybody.

Reference

Messac L, Reeves MJ, Schwamm LH, et al. Insurance, race, and access to endovascular thrombectomy in California, 2015-2021. Presented at: International Stroke Conference 2026. February 3-6, 2025; New Orleans, LA. Abstract A141.