Unveiling Disparities of Patients With Lymphoma by Race, Ethnicity, and Socioeconomic Status

Study findings revealed higher Social Deprivation Index (SDI) scores among African American and Hispanic patients, highlighting the need for prospective data collection to address these disparities, says Alvaro Alencar, MD, associate professor of clinical medicine, chief medical officer, University of Miami Sylvester Comprehensive Cancer Center.

Transcript

Can you provide an overview of the study?

So, this is a very interesting study, because we know—or we have a good sense—that there is an impact of disparities on the outcome of patients treated with cancer in general. In lymphoma, which is my research specialty, we also sense the same thing, but we do this based on suboptimal data. This is data that is retrospective, so looking back and trying to see what was captured and try to see whether we can identify disparities and see how they impact the studies. Whenever you look back, when you're not actively collecting the data to evaluate it, there will be a lot of gaps.

Our study was divided into separate steps. The first step is looking back on patients that were enrolled on lymphoma studies in the cooperative groups, Alliance [Alliance for Clinical Trials in Oncology], SWOG [Southwest Oncology Group], and ECOG [Eastern Cooperative Oncology Group], over the last 15 years. So mainly focusing on lymphoma patients in a cooperative group. The reason for a cooperative group is because a cooperative group is a very unique set of patients, these are centers that both include academic centers and small community practices, but usually patients are treated by doctors who are specialized in this and treated in a clinical trial. So, they are very standardized, they have eligibility criteria, [and] they get treatment under a study. So, it allows you to get data, because they're in the study. They collect a lot of data and they're treated uniformly, so then you can try to really see the impact of other things beyond the treatment itself on the outcome of this patient.

The cooperative group is great for this, but we haven't really had a lot of focus on collecting this data on social determinants of health. So, our initial idea was to look back 15 years, see what was collected, and try to learn from that. See what was collected, how it was collected, where the gaps are, and if we have enough data to try to evaluate something, great. The main idea is to look back and learn, and then really develop in a second step, a robust way to prospectively collect data. So, then you develop a questionnaire to actively collect the data, and then really have reliable data to try to make recommendations or strategies and policies, and so on. So, the study is really divided in 2 phases, looking back first, learning, and then developing something to move forward with.

What we're presenting at ASH [Saturday, December 7] is the first batch of data. It’s really just demographic data on these patients, just the learning from these 15 years what we saw, and this is purely descriptive, just describing the population that we have seen.

Can you discuss the potential association between minorities and higher deprivation scores?

The data that we have is very provocative, so it starts raising questions. There indeed seems to be an association between minorities, so both race and ethnicity, and disparity. So, patients that were African American or Hispanic seem to have higher SDI, higher deprivation. But again, this is based on retrospective data, and retrospective data have a lot of gaps. So, as you will see when I present, we have a lot of unknowns. So, how can you rely on data that have 10% unknowns, right? It brings a lot of questions, and I think that what it does is that it reinforces the importance of doing what we're doing in the second step of the study, which is collecting prospective data that can really give us robust points that we can confirm the issue with disparities, and then collect data that allows you to see what are the low-hanging fruits. What are the things that can really help you identify the problem?

I think that what we do know is there is a strong suspicion, and that kind of goes along with what we feel already. And so, this data just reinforces that. What it really tells us is that you need to look into this prospectively, really, with a focus on collecting data actively, and then learn from that.