The Role of Social Determinants of Health in Epilepsy-What We Don’t Know Can Hurt Others: Disparities in Treatment of Infantile Spasms

Video Transcription

Thank you so much for that wonderful introduction and for inviting me to speak today. It's an honor to be on this panel of incredible speakers  and to present on such an important topic. As I was reflecting about the title of my talk today and the crux of it referring to the adage, we don't know what we don't know,  I thought, well, the curtain really has been pulled back here. We, over the last several years, have seen time and time again  how ubiquitous healthcare disparities are across our health system. And so this, unfortunately, is not a new story, but I still think it's a really important one to be told,  especially in the context of a state-of-the-art symposium. I think as we continue to try to work to move our field forward, we really have to continue to intentionally create space  to have these discussions around healthcare disparities so that we can make sure that as we advance to new frontiers, as we answer these important clinical questions  and advance diagnostic and therapeutic approaches, we really think about how we can be sure we are moving all of our patients forward with us.  And so with that, the title of my talk is What We Don't Know Can Hurt Others, Disparities in Treatment of Infantile Spasms. I'd like to also acknowledge Dr. Fiona Baumer,  who was the lead on the main project that I will be presenting today. I have no relevant disclosures, but I also want to start by acknowledging the list of people here  that helped with the collaborative efforts to work on this project, as well as others not listed here who have worked through the Pediatric Epilepsy Research Consortium  and the National Infantile Spasms Consortium to really be able to collect this data in a meaningful way so we can look at healthcare disparities in this way  for this patient population. So my learning objectives for this portion of the talk are really two-fold. The first is to analyze how these national multi-center databases  can be used to investigate important questions in healthcare disparities, and in particular the treatment of infantile spasms. And that really is the first problem,  is to recognize and define what the problem is. But the second is to consider ways to address these barriers to standard therapy. So just getting at the why.  Why do these disparities exist, and more importantly, what can we start to do about them? So to set the stage, as we all know, infantile spasms  is associated with a severe developmental and epileptic encephalopathy, and despite being a large part of many of our clinical practices, it's actually quite rare,  2 to 3.5 in 10,000 children. But it does occur in children of all races and ethnicity. And as we've heard, the 2010 expert consensus  and publications since then have really demonstrated that it's important to have rapid initiation of one of the three first-line therapies,  which includes adrenocorticotropic hormone, or ACTH, oral corticosteroids, or vigabatrin. And that after one of these first-line therapies are initiated,  rapid reassessment for the persistence of spasms is important. One to two weeks later, and seeing if spasms is still persistent, then transitioning to  or adding on one of the other three therapies by day 14 is recommended. The concern being that delays in therapy can lead to worse neurodevelopmental outcomes  and really have significant clinical impact on patients and families. So the way to start to really look at  the effectiveness of these therapies and these recommendations is, as Dr. Grinspan said, using these multi-center prospective databases.  So the Pediatric Epilepsy Research Consortium in 2012 developed this National Infantile Spasms Consortium. And what this was was a prospective database  that enrolled infants 2 months to 24 months of age from 2012 to 2018 across 25 different centers across the country.  And you can see them represented on this map here. And the prospective data was entered during this time period and ended up collecting information  on 629 infants with a new diagnosis of infantile spasms. And so just to briefly kind of recap what these studies have showed, as Dr. Grinspan mentioned too,  the first early analysis of this database in 2015 included 230 infants from 22 of these centers  and really demonstrated that the response to treatment was significantly higher in the standard therapy group. So it was one of the three first-line treatments.  In this study, there were 14% of infants who received a non-standard therapy. So that was either a different anti-seizure medication or the ketogenic diet.  And you can see only 9% of infants who received non-standard therapy had a response to treatment. Response to treatment included two-week resolution of spasms  or by the three-month data entry point, resolution of spasms without evidence of recurrence during that time frame. And that 9% response to non-standard therapy  really only represented at this time point three individuals who had response to non-standard treatment. So the NISC database did record variables  for race and ethnicity that were derived from the medical record. And there were seven variables for race that are listed there.  It included an option for other or the race was unknown or not reported in the medical record. And then there were three variables for ethnicity  being Hispanic, Latino, non-Hispanic, or ethnicity unknown. And so to begin to look at race and ethnicity as a factor for treatment choice for infantile spasms,  this analysis looked at children who were identified as black, white, or other, and then also those who were identified as Hispanic. So there could be some overlap.  But what you can see is that regardless of race or ethnicity, the most common first therapy chosen for a new diagnosis of infantile spasms was indeed ACTH,  which we saw in our poll today as well. There was some variability for the other treatment options for the first line therapy, including that other  or non-standard therapy option, which is that cobalt blue color.  But this initial analysis did not show a significant association with race or ethnicity in treatment choice. However, what we can note is that 7 out of 31 black children,  so 22% in this cohort,  did receive a non-standard therapy. And when we think to the implications of the response to treatment with non-standard therapies, that was something that was noted, but again,  there was not a statistically significant difference between these groups at this time point in the analysis. So then a follow-up analysis in 2016 set out to look  at the response to treatment to a second medication if the first medication failed. So at that two-week time point, spasms persisted and a second therapy was needed.  And we can see that those who received standard therapy, which was the hormonal therapy or vigabatrin, had a 36% response rate, and those who received  non-standard therapy had a 20% response rate. So suggesting that the entire treatment course may be important, particularly when you consider that for all of these infants  who did not respond to the first treatment,  greater than a third of them may respond to a second medication. So highlights the need for rapid re-evaluation and analysis and when appropriate, starting that second therapy.  So again, looking at the race and ethnicity variables, this analysis also did not show a difference in treatment response based on race and ethnicity.  So the analysis looked at treatment response  but when we look to see what the treatment choice was, we can see from kind of the raw data that children who did not identify as white did have a higher rate  of receiving non-standard therapy for that second treatment. So then fast forward to 2020, we have our entire prospective database collected.  And to look specifically at these variables of race and ethnicity, we included 550 individuals who had received non-standard therapy for the first two treatments.  And for race and ethnicity, we included 555 infants across 25 centers from this NISC data set. The inclusion criteria were the same,  where the diagnosis of infantile spasms was 2 to 24 months of age. They had received at least one therapy within 12 months of onset  and had at least two weeks of follow-up data recorded. So with this larger database, we really set out specifically to investigate  whether children from racial and ethnic minoritized groups who presented with new onset infantile spasms received similar treatment as white children with infantile spasms.  So again, looking at the race and ethnicity variables that were collected during this time period, we made a couple decisions about how to analyze this.  And so first we looked at all of those who identified as Hispanic and that represented 14% of the database. And then we separately looked at those who identified  as white non-Hispanic, black non-Hispanic, or Asian non-Hispanic. And then the remaining variables had either quite low numbers or identified as other  or the race or ethnicity variables were not known. And so we collapsed that into a separate category that represented 13% of the data set.  So that left us with five race and ethnicity variables as you can see in the information shown here. So now that we've defined our race and ethnicity variables  that we used for this analysis, I want to take some time to look at what we defined as the treatment course. So again, this data set included 555 infants  and a standard treatment course was if you received one of the three standard therapies for treatment one and when needed, you received one of the three  standard therapies for treatment two. It also included patients who did not have persistent spasms and did not need that second treatment.  So this was considered a standard therapy course. A non-standard therapy course was if you received something other than those three standard treatments  for treatment one and that could include other anti-seizure medications or the ketogenic diet. Or you did not receive standard therapy  for treatment two when it was needed for ongoing spasms. So a non-standard treatment course was not receiving a standard therapy for treatment one or treatment two.  So when we look at the first treatment alone, the good news is that the majority of patients did receive standard therapy. However, we can see that black children were still  the least likely to get this for even the first treatment. 20% of black children received non-standard therapy. When we looked at treatment two individually,  we can see a much starker difference between the groups as well. Nearly 50% of black children received a non-standard therapy for that second treatment.  This did not reach statistical significance and one possible explanation could be in part due to a loss of power since we lost about half  our sample size when we looked at treatment two alone. So then we looked at our combined outcome of a standard treatment course.  Did you receive standard therapy for both treatment one and treatment two when needed? And we can see that black children received standard treatment course significantly less often  than the other groups. And again, nearly half of black children received a standard treatment course. So looking at this a slightly different way,  children who identify as black, non-Hispanic had a lower odds of receiving a standard treatment course when compared to white, non-Hispanic patients.  So we have the unadjusted odds ratio on the left, but even after adjusting for a number of variables, including other demographic variables such as insurance type  and a plethora of clinical variables, time to treatment onset, other patient characteristics such as head circumference, history of prematurity,  presence of developmental delays or regression, even after adjusting for a number of factors, the adjusted odds ratio still clearly showed that black children had a lower odds  of receiving a full standard treatment course.  The other thing to note was that we found that children with public insurance were less likely to receive standard therapy for even treatment one.  So in summary, black, non-Hispanic children had a lower odds than white, non-Hispanic children of receiving a standard therapy course for the treatment of their infantile spasms.  There are no other racial or ethnic groups that we identified that had different odds, but we did see that children with public insurance had a significantly lower odds of receiving  standard therapy for treatment one, and this can have implications on clinical outcomes and is one of the reasons that we felt it was so important to look at.  So I would like to end on this call to action. There are admittedly still a lot of unanswered questions about what are the optimal therapies and management for infantile spasms,  as we already heard today and will continue to hear, and luckily there's a lot of incredibly smart people that are actively working on these issues.  So I think the first step is just right here, right now, as we're all sitting in this room and we are gathering this information,  let's use it to incorporate it into updating protocols and operationalizing these protocols that really rely on the best level of available evidence  to help optimize treatment for all of our patients. Really relying on these standardized protocols can help mitigate potential sources of bias in our treatment plans.  Thinking a little bit more broadly outside the walls of this conference hall, we could potentially go back to our own states and advocate within our state's Medicaid programs  to ensure that standard therapies are covered from the onset,  and then narrowing our scope just a little bit, we can, as we all kind of go back to our home institutions and pick up our clinical practices,  we can try to identify barriers to standard of care that may exist within our own hospital settings and clinics.  And then the last and perhaps most immediately impactful is that all of us can take a moment to pause and look inward at our own clinical practice  and examine how our own treatment recommendations may be prone to assumptions and biases about our patient's social status, however well-meaning.  And so I think that really moving forward, we can all take this information to inform our own clinical practice.  As I mentioned, this is not a new story. This is something that just reveals to us that we are not immune to the problem, that we, including pediatric neurologists  and epileptologists, are working within the same social systems that have created and perpetuated these disparities for a long time now.  And the fact that perhaps we could be part of this problem is uncomfortable, but also allows us to all be part of the solution. And so I think opportunities like this  to think about how we can move our field forward while also making sure that we are taking all of our patients with us is the next step. So thank you very much for your time today.

Video Summary

The talk highlights the pervasive issue of healthcare disparities, particularly in treating infantile spasms, underscoring the importance of addressing these disparities in clinical practice. Despite guidelines advocating rapid initiation of standard treatments like ACTH, oral corticosteroids, or vigabatrin, disparities persist, particularly affecting Black children and those with public insurance. Data from the National Infantile Spasms Consortium show Black children received non-standard treatment more frequently, impacting their clinical outcomes negatively. The speaker advocates for incorporating evidence-based protocols, advocating for policy changes to ensure coverage of standard therapies, and examining personal biases in treatment decisions to improve equity in healthcare. The talk calls for collective efforts to mitigate these disparities and ensure all patients can benefit from advances in infantile spasms treatment.

Asset Subtitle

Presenter:

Kerri Neville, MD, is a Clinical Assistant Professor of Pediatrics in the division of Pediatric Neurology at Mott Children's Hospital in Ann Arbor, MI. She completed her child neurology residency and epilepsy fellowship at Michigan Medicine before staying on as faculty. She works primarily as a clinical pediatric epileptologist including pediatric epilepsy surgery evaluations. In addition to her clinical work, she is the Medical Director of the Pediatric Clinical Neurophysiology Lab. She also serves as the DEI lead for the division of pediatric neurology and has an interest in health equity research.

Keywords

healthcare disparities

infantile spasms

treatment equity

evidence-based protocols

policy changes