Racial disparities in health care utilization at the end-of-life among New Jersey Medicaid beneficiaries with advanced cancer

JCO Oncology Practice Podcast

Dr. Pennell and Dr. Jennifer Tsui discuss the processes that lead to suboptimal EOL care within Medicaid populations and among racial/ethnic minority groups.

Hello, and welcome to the latest JCO Oncology Practice podcast, brought to you by the ASCO Podcast Network, a collection of nine programs covering a range of educational and scientific content and offering enriching insight into the world of cancer care. You can find all recordings, including this one, at podcast.asco.org.

My name is Dr. Nate Pennell, medical oncologist at the Cleveland Clinic and consultant editor for the JCO OP. I have no conflicts of interest related to this podcast, and a complete list of disclosures is available at the end of the podcast.

Aggressive care at the end of life for cancer patients is widely recognized as poor-quality care. And by aggressive care, I don't mean aggressive supportive care or hospice, but rather inappropriate interventions, like chemotherapy or hospital and ICU admissions, near the end of life that rarely improve outcomes and often actually worsen quality of life.

Efforts are being made to educate physicians and cancer patients to try to minimize aggressive treatments near the end of life and to help as many patients as possible benefit from things like hospice benefits and appropriate end-of-life care. However, not all patients receive high-quality end-of-life care, and there may be differences in end-of-life care in various populations. For example, how do race and things like Medicaid status impact aggressive care at the end of life?

With me today to discuss this topic is Dr. Jennifer Tsui, Assistant Professor in the Division of Population Science at Rutgers Cancer Institute of New Jersey. We'll be discussing her paper "Racial Disparities in Health Care Utilization at the End of Life Among New Jersey Medicaid Beneficiaries With Advanced Cancer," currently in press for the JCO OP. Welcome, Dr. Tsui, and thank you for joining me on the podcast.

Thanks so much. Thanks for having me. I have no conflicts of interest with this study whatsoever.

Dr. Tsui, can you please tell me a little bit about what exactly constitutes high or low-quality end-of-life care?

Sure. I mean, I think that, in this study in particular, we wanted to focus on guideline-related end-of-life care. So we wanted to see if it was possible to take a look at patterns at the end of life for breast and colorectal cancer and stage cancer cases and see sort of what the patterns were in relation to adherence to guideline adherence and what they should be receiving at the end of life.

And so that included a set of measures around aggressive care related to hospitalization in the last 30 days, emergency department visits in the last 30 days of life, an ICU admission in the last 30 days of life, and chemotherapy in the last 14 days of life. These are guidelines that have been discussed and published by national organizations. And we also looked at hospice enrollment. So we looked at whether there was any hospice enrollment and whether there was hospice enrollment in the last 30 days of life, I'm sorry.

Are there already data existing for various disparities in end-of-life care among different racial groups or patients of different socioeconomic status?

There are. So there have been a few prior publications before our study that have shown that Medicaid patients frequently-- not just for end-of-life care, but cancer care in general-- that Medicaid patients receive lower quality of care. And there were studies done prior to ours that did show, I think, in New York, for example, that Medicaid patients had lower-quality end-of-life care compared to Medicaid and privately insured patients.

We have seen other studies also mentioning disparities by race in terms of quality of end-of-life care. However, I don't think-- some of these studies have focused on different populations and cancer sites. So some of the studies I mentioned looked at AYA, adolescent and young adult cancer survivors. Other studies have focused on other cancer sites.

And also, these were studies that were conducted in earlier time period. And what we know is that, given all that's happened since the Affordable Care Act and Medicaid expansion in several states, sort of the Medicaid population has changed, but sort of health care delivery has been redesigned in different ways. And so that was really why we wanted to look at this issue sort of in more recent years and during a period that spanned Medicaid expansion within New Jersey.

You know, that's helpful to understand what sort of prompted you to do this. So why don't you tell our listeners basically how you designed your study.

Sure. So this was a large data linkage that we established within the state of New Jersey to better understand cancer care quality for breast and colorectal cancer patients in general, so beyond just the stage IV cancer patients, the patient's diagnosis at stage IV, but breast and colorectal cancer patients in general.

So we worked with our New Jersey State Cancer Registry, identified all our breast and colorectal cancer cases that were diagnosed between 2011 and 2015, and then we linked those with our New Jersey Medicaid enrollment and claims files. So this was a study that was done in partnership with our Rutgers Center for State Health Policy. It was conducted sort of with ongoing input and feedback from both our State Cancer Registry as well as our Medicaid program. They've seen sort of findings related to this and other research questions we've had throughout the process.

And what we really ended up with was a very rich data set that not only gave us all the tumor characteristics that are available in the cancer registry, but also the health care utilization patterns and Medicaid enrollment characteristics that are available on the Medicaid claims and Medicaid program data side.

And how well did this database capture all of these measures of aggressive end-of-life care, so ED visits and chemotherapy and such?

Since we did focus on those who were continuously enrolled in our state Medicaid program, we were able to go through our claims data and be able to, using billing codes, identify things like ED visits, et cetera. This data set only includes cancer cases up to age 64. So we didn't include people who may be dual enrolled in Medicare or have Medicare claims.

We also tried to create some other restrictions so that we can get at just the people where we had a definitive diagnosis month and year and follow them through their death. And so we did use the prior studies that I mentioned earlier to see what kind of codes were used there.

We have quite a multidisciplinary team of collaborators. I worked with quite closely the first author here, Annie Yang. Soon-to-be Dr. Annie Yang is in medical school here at the New Jersey Medical School. And so we did try to be as comprehensive as we could with the codes we used to identify those utilization patterns.

OK, so why don't you tell us some of your findings?

So I would say that the overall finding is that 62% of these stage IV breast and colorectal cancer patients received at least one measure of aggressive end-of-life care. So out of those four individual measures, 2/3 received at least one, which seems quite high. They ranged from 27% having at least one hospitalization to 34% receiving chemotherapy within the last 14 days of life.

What we found is alarming, and it's sort of a call to action for addressing racial and ethnic disparities in health care, is that we found that our non-Hispanic Black patients had a higher odds of receiving any one of those aggressive end-of-life care measures after controlling for other factors. So we did find this racial/ethnic disparity in non-Hispanic Black patients having more aggressive end-of-life care compared to our non-Hispanic white patients, even though these are all the same Medicaid program. This is within one state.

And so it did point to some need to better understand what is going on within the health care system, within the health care encounter between patients and physicians towards the end of life and what needs to be done to address these disproportionate rates in end-of-life care among racial/ethnic minority patients.

When you talk about these numbers, like 2/3 of patients having aggressive end-of-life care and 39% enrolling in hospice, how does that compare to what would be considered a more appropriate level, say, the private insured patients?

So we didn't compare it to privately insured in our state, but we did look at what was published in other states. In New York, for example, I think we had a slightly higher proportion of Medicaid enrollees with stage IV breast and colorectal cancer having more aggressive care. So we did see higher rates. So while we can't compare within our state alone, we do see that the rates are slightly higher than other published studies.

Well, I guess I'm just trying to figure out, is aggressive end-of-life care something that commonly happens to all Americans in general, or is this vastly more in this population than what we would expect?

In the general population?

Yeah, in the general--

You know, I don't-- yeah, I don't have those rates and what our bar should be at baseline, to be honest. I do think that what we see in the end-of-life hospice literature is that there is suboptimal use of hospice care. And what we found was, again, here also sort of suboptimal use of hospice care in this diverse, low-income, Medicaid population.

No, clearly, clearly, definitely so. What are some reasons why Medicaid patients would be more likely to have aggressive end-of-life care, do you think?

I think Medicaid patients sort of broadly often have higher rates of comorbid conditions when they're diagnosed with cancer. There are complex social factors related to the Medicaid population in terms of being low-income, in terms of other social determinants of health and social needs that increase barriers to care or barriers to high-quality care or continuity of care.

In another paper that we published that focuses on this data linkage in the population and this data linkage, there was a lack of primary care utilization prior to cancer diagnosis as well as a lot of the cancer patients in Medicaid enrolling in Medicaid upon diagnosis, right? So I think that there are sort of just several multi-level factors that contribute to why there may be higher rates of aggressive end-of-life care.

I think, from a patient perspective, there is a literature out there on the need for communication tools and sort of interaction and inclusion of caregivers that are a part of the decision-making in end-of-life care. Certainly, we've seen in the literature for a while the issue of providers and providers not only being sort of culturally competent or racial and ethnic concordant, but having the tools that they need, both for the physician or the health care team, to have these end-of-life discussions. So I think it's very multi-level and certainly not just either on the patient or the physician.

Even within the Medicaid population, so independent just of the, say, the financial pressures that would lead one to enroll in Medicaid, what you showed was that non-Hispanic Black patients had a much higher rate of aggressive end-of-life care even than the overall Medicaid population. And why do you think that would be true?

Yeah, so after we finished the analysis for this, we definitely did reach out to other colleagues who have worked in the communities more closely within New Jersey, within the non-Hispanic Black population across New Jersey. We've also talked to our state Medicaid program to, one, think of sort of action items for how to better understand what we're seeing in the quantitative administrative data here, and then two, how to engage communities to understand what to do next about this.

So from the literature and from sort of racial/ethnic disparities literature, we do know that there is often mistrust of the medical system. There's underutilization of preventive screenings, and there are physician biases and structural biases that occur along the way, so again, the multi-level factors that impact why there may be disproportionate aggressive end-of-life care in the non-Hispanic Black population. You know, we can't answer those questions with Medicaid claims and enrollment files. We can identify these patterns.

And I think what our research team has been in communication with our state Medicaid program to do next is then figure out, how do we talk with some of these communities across the state to figure out what the appropriate community-level education tools might be needed to improve an understanding of what end-of-life options are, but also to focus on the hospitals and health care systems that may be disproportionately seeing some of the-- that may be seeing higher proportions of non-Hispanic Black patients, and how can we engage the health care teams within those settings to figure out what can be done at the health care delivery level?

Do you think we have enough information to make interventions now? I mean, are there-- how do you think, aside from further study and trying to understand it better, we might, in the short term, improve rates of quality end-of-life care in these high-risk patients?

We've seen-- and we've talked about this, I think, in this field somewhat-- is, one, there are guidelines. And just having health care teams, providers, hospitals understand where they are, sort of tracking where they are in rates of aggressive end-of-life care and in hospice enrollment, and having that feedback and audit loop to kind of look introspectively within the health care delivery setting, I think is one option that we've seen in terms of trying to increase quality and increase adherence to guidelines for other things.

And then I certainly think that there might be a need for community-level interventions around, what are the barriers to hospice enrollment? Or where is the lack of awareness and knowledge around end-of-life care options? And what does it mean to reduce interventional care, right? And what does it mean for communities? I think that targeting those aspects and having a dialogue that is responsive to the community-level needs are probably ways that we can start.

I think that makes sense. I mean, as an oncologist, a city with a high minority population, I think improving the cultural competence and being able to establish a level of trust with patients who may have an inherent distrust of the medical system, especially when you're talking about something like that's as counterintuitive as trying to recommend not doing more aggressive care, it's certainly a complex conversation and definitely would require a level of trust and communication in order to do that properly. So I completely agree with you that I think, on the community level, the interventions will have to involve that if it's going to be effective.

So Dr. Tsui, thanks so much for joining me on the podcast today.

Thank you for having me.

And until next time, thank you to our listeners for listening to this JCO Oncology Practice podcast. If you enjoyed what you heard today, don't forget to give us a rating or a review on Apple podcasts or wherever you listen. While you're there, be sure to subscribe so you never miss an episode.

JCO OP podcasts are just one of ASCO's many podcasts programs. You can find all the recordings at podcast.asco.org. The full text of this paper will be available online at ascopubs.org/journal/op. This is Dr. Nate Pennell for the JCO Oncology Practice, signing off.

The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions. Guests on this podcast express their own opinions, experience, and conclusions. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.

For more original research, editorials, and review articles, please visit us online at jco.org. This production is copyrighted to the American Society of Clinical Oncology. Thank you for listening.