Health Inequity in Cancer Research: [Slides] Examining and Uncovering the Disparities by Hala Borno MD

Hala Borno, MD discusses Health Inequity in Cancer Research: Examining and Uncovering the Disparities


Health Inequity in Cancer Research: Examining and Uncovering the Disparities by Hala Borno MD

By Hala Borno, MD

What also energizes me is that while I was walking down here, I was smiling because I actually got married in this hotel about 10 years ago in the ballroom where the NAFFLE Conference is going on right now.

It is just very interesting nostalgia that I was experiencing, but I am delighted to be here and very grateful for the invitation to speak with you. So today I'll be talking about a topic that's very near and dear to my heart, which is Approaches to Advanced Health Equity in Clinical Trials. These are my disclosures. So the outline of my talk will first be really thinking about definitions. What are health inequities? What is recruitment science, and how can we understand factors that are driving inequities? We'll also be thinking about recruitment practices from an institutional perspective, and I'll share what I learned through my UCSF experiences, and then study sponsor considerations to advance equity.

So now definitions, health inequities are systematic differences in health outcomes. So they're differences. But what I want you to walk away from this conversation understanding is that there are differences in health status or in the dis distribution of health resources between different populations that arise from the social conditions in which people are born.

Grow, live, work, and age. So there's a fundamental relationship between these differences and these other social determinants. . So when we think about differences, let's consider how we do in clinical trials. These are data published by ASCO, where in one column you see here the percent of US populations.

So these are general demographic data. And then on the other side you see the percent of participants in oncology clinical trials. And what you see here is an overrepresentation of non-Hispanic white populations in our clinical trials. Relative to the US population, right? What you hope to achieve is at a minimum representation so that these are symmetrical, right?

But we don't achieve that. We indeed see an underrepresentation of racial ethnic minorities and clinical trials. Now this concept of underrepresentation is actually not new. We've known about it for quite some time, and there's been legislation about it for three decades. In fact, in 1993, there was the NIH Revitalization Act, which mandated the inclusion of racial ethnic minorities and women in publicly funded clinical trials.

Yet this mandate did not really have a lever to ensure accountability, and yet, three decades later, we persist to see the same inequities in whom clinical trials are serving. What we do know is that the NIH Revitalization Act spurred a new field of inquiry, and this field is called recruitment science, which is an empirical body of studies scientifically evaluating the efficacy of various social, cultural, psychological, technological and economic means of convincing people, especially members of hard to recruit populations that they want to become and remain human subjects. This term, also known as recruitment, was coined by a social scientist, Steven Epstein, and is the area of research that I've dedicated most of my career. And I'm going to walk you through some of the key learning.

So let's think through the literature in recruitment science. We know there are a variety factors. This is a multifaceted problem, and these factors can be on a patient level, a societal level, they can be provider factors, institutional factors. Or really fundamentally related to trial design. So when we think about patient factors, there may be socioeconomic or insurance status issues. There may be beliefs, attitudes towards clinical research. On a societal level, there may be this historical context that we need to consider. There may be this concept of medical mistrust, provider level there could be bias and who we are offering clinical trials to, and bias and diagnostics testings as we recognize that increasingly our clinical trials are biomarkers selected. On an institutional level there could be differences in clinical resources, translation resources, research support and trial design, DEI prioritization, eligibility criteria, study materials, and trial site selection. This is a tour de force of these concepts, but we're gonna dig deeper into all of these now . So let's think about patient and societal factors. When you think about this, you have to have an understanding of this core concept of social determinants of health. I'm sure many of you have heard this term already, which are the conditions in the environment where people are born, live, learn, work, grow, play that affect a wide range of health outcomes. And what you can understand is that these social determinants of health are related to our understanding of health inequities, right? So differences that you see driven by social determinants of health are what health inequities are. And there are different groups, that we think about in social determinants of health, and these domains include economic stability, education, healthcare, access, neighborhood, and built environments in social and community context. So when we are doing research on health inequity, we always relate them to these different factors. So in the context of clinical trials and social determinants of health, my team published a piece in JCO Oncology Practice really describing this concept of a long commute where when you're thinking about what an experience looks like for a patient to participate in a clinical trial, there's a variety of burdens that you're asking that patient to undertake.

Perhaps it's waking up early, getting gas. Perhaps arranging some sort of childcare or dependent care. Then they arrive to the therapeutic clinical trial. They receive their treatment, they receive advanced diagnostics. They go on to experience side effects of the treatment. They have to find a pharmacy to get the medication to manage their side effects, and eventually they return home. This is a windy road that has bumps in it, and we're asking patients to undertake this for a very long time, and what you can imagine is there's selection bias and who's willing to actually take this trip. 

So there have been different interventions trying to characterize factors related to social determinants health and address them, and one experience is really this cancer care equity program that was developed first at a Dana-Farber, where they characterize, how much financial burdens are we asking patients to experience? And are we seeing that patients that are on clinical trials are very concerned about the financial risk they may experience on study? And what they deployed was an out-of-pocket reimbursement program to address, costs associated with travel. So really addressing some aspects of that long commute. And what they observed in an interrupted time series analysis, which you see here at the bottom, is that there was a significant increase in recruitment overall, across all phases of study. For their center after they addressed out-of-pocket costs associated with travel to a clinical trial, so this one intervention to address social determinants of health, increased recruitment to study, and then when they looked at the populations that were being served and asked them about how worried they are about costs, and other factors about their daily living and stress. They found that they were preferentially providing resources to patients that had high levels of what we call financial toxicity or high concern about cost for receiving care. We deployed a similar intervention at UCSF and USC where we addressed the out-of-pocket costs associated with travel in a program called the Impact Program, and we also, in our interrupted time series analysis, observed a significant increase in recruitment overall across all phases of study with a larger effect actually in late phase clinical trials. So this again, is a validation of these key learnings. You can recruit more patients if you help them out with that long commute. When you look at the different types of costs associated with travel and you characterize them by study site, you can see that overall most of the costs are land travel. That's something we would expect, but then when you break it down on a study site level, there'll be different costs. Breakdowns based on the topography and the context in which patients have to travel. If it's a urban metropolitan with a lot of tolls and lots of traffic, you, you will see a different expense associated with gas, parking, and tolls, then you would for another environment that might be easier to get around. So these are considerations to think about in the context of what trial site a patient has to travel to. 

Now let's shift gears a little bit and think about the patient experience as it relates to our history in society and we can all acknowledge that there have been pretty egregious examples of exploitation in our history as it relates to human subjects. And here I'm sure you recognize this photograph of Henrietta Lax, whose immortalized cell lines were exploited and used for scientific discovery without her or her family's consent. And this history brings up this notion of hereditary mistrust, fears of experimentation, as well as therapeutic misconception. But I bring this up to actually make a different point. I want you to know that amplifying the role, Trust is actually problematic because it reinforces a model of research, a deficit model of research. Really our team published this in JAMA Oncology where we stated that am amplifying the influence of mistrust implicitly places blame on patients without acknowledgement of the larger determinants. Such as racism that creates circumstances of disadvantage for certain populations. Receiving care in the US health system, this further reinforces a model of a deficit model of research and engagement. I brought it up for completeness sake, but wanting us to all know that is not a justifiable reason for health inequities that we observe, and in fact, propagating it is quite problematic.

So when we think about structural determinants of health, let's think about the context in which patients are receiving care, the provider and the clinical environment. So providers, including myself, have biases fundamentally, when we're interacting with a patient our own personal experiences, culturals and values influence the way that we perceive the information we're receiving. Influence, the clinical decisions we're making influence our treatment recommendations, this has an impact in clinical trial access. So ethnographic research has been done. Ethnographers, pre predominantly. Sociologists actually sat in and observed oncologists in a top comprehensive cancer center and characterized the types of patients they were looking for to offer a clinical trial opportunity and observe that they tended to offer a clinical trial if the patient appeared to be more proactive, meticulous, perceived to have good communication, adherent embedded in a social network. These are implicit biases that a provider has that is influencing when they're asking about a clinical trial for a patient and giving them that chance. So that's the provider perspective. And Nadine Burke from Duke and other investigators have launched the "Just Ask" initiative, which is really this promotion of just ask the patient. And recognize that asking you might be surprised in what patients are interested in learning more about.  Now thinking about the context, because context matters, providers in a lot of clinical environments might be too stretch thin to actually wanna bring up clinical trials. They're just trying to deliver standard of care at a high quality and they're moving between room to room and they don't have the capacity to also have a clinical trial discussion. Recruitment ties up clinic exam rooms and that's a real barrier for providers who may have a high volume. For other contextual factors, especially thinking about the patient experience. Not all clinic environments are easy to navigate for a patient. So if you think you're asking a patient to come back and go get phlebotomy here, go get the scan here and then come see your provider for a physical exam, and they're like, I don't know where any of these places are, asking me to do this over and over again every three weeks. That could be quite a really challenging thing to ask for patients. Patients may also be experiencing long wait times in certain clinics. So think about a patient who has economic opportunity loss for every hour of wait time, and then they have to do that every three weeks where they may have to wait for four hours, that's an incredible expense and it's hard to measure that economic opportunity. . Yet we know that influences their decisions to engage. Multilingual capacity is variable across different clinics trial information accessibility is variable and the role of navigate navigation also to serve providers needs and patients needs is limited in most clinical environments. So these are contextual factors that fundamentally add to this picture of inequity. 

Now if you want to address inequity, you have to make sure you're measuring it and following it, and so at UCSF I co-lead the development of this dashboard called the "Minority Equity and Recruitment into Trials Dashboard." And what we did was provide real-time analytics about how we were doing in terms of our recruitment equity goals. And we pulled three distinct data sources. The first data source we pulled were the incident cancer cases in our catchment area, which is the area that we are meant to serve. So pulling these data from the population cancer registry. We also captured incident cancer cases at our cancer center. So these are patients that are being seen by us, who are new cancer cases, and then looking at the cases of patients enrolling in our trials, we also broke these data down by race, ethnicity, age, and gender, and then we were able to filter based on disease stay research program time, even on a PI level, trying to characterize how are we doing from an equity perspective, recognizing that any given clinic might be doing fantastic in recruiting to their breast cancer studies, but really terribly recruiting for the thoracic studies. So we built this out and of course, recognize that we did have underrepresentation of racial, ethnic minorities and other identities, but now we were able to actually monitor this in a more robust fashion and start capturing when we were making an intervention that was making an impact. 


I personally am very interested in the role of technology and human touch to advance equity and characterized how we were doing from a recruitment discovery perspective at our environment. And we are part of a larger health system, UCSF Health, which has resources to provide technology, multi-channel recruitment campaigns, most of which are not applied in oncology, but actually in other disease settings. And then with an our cancer center the vast majority of our recruitment approaches were what we classified as inReach, so recruiting patients after they arrived to our cancer center, rather than outreach, really targeting, engaging patients who haven't yet arrived to our door. When we looked at the use of technology and you see here I broke down our recruitment practices at our cancer center For each disease group, I classified it either as in reach, so recruiting from patients arriving at our cancer center versus outreach, recruiting patients from our communities. And then I also flagged when technology was used, and what you can see here is the vast majority of our recruitment was done in-house only among patients that came to us and very limited use of technology. We were enriched for technology in GU because that's my clinical area of interest, and also the most common way that we recruited patients to trials was word of mouth. Was we told our peers what trial was opened when during our tumor boards or during our weekly conference meetings. And we also had a sheet of paper that we posted on a cork board. This was our robust means of communicating trial information at a comprehensive cancer center. So we knew that this type of informational silo was likely also a factor that's driving inequity.


Now when we think about inequities in clinical trials we also think about the context in the preclinical to clinical phase. So I know we have a lot of study sponsors folks who work with the study sponsor here. And this is really critically important to think about. We all know race is a social construct and not a biological variable, of course. And genetic variation in ancestry are distinct from race. However, race is actually a pretty useful proxy for this concept we've been talking about, which is social and structural determinants of health. And that's the context in which we follow it as we know that there are lived experiences that may be shared by individuals with a common race.


So that's why we follow race now when we think about data analysis in a preclinical phase. It's really important to characterize the population for which you were able to capture that data and to ask questions about the generalizability of your data set and to identify where your data is underpowered to make conclusions about safety and toxicity across diverse populations. So going back to that early slide where I showed you that the goal was representation. That was, that's what the dictum has become since the 1993 NIH Revitalization Act. But the goal is actually much larger than that, it's actually oversampling minority populations. So you are powered to make really valid conclusions about safety and efficacy across populations. That's really the Nirvana state. That's what we're trying to achieve. 


When we think about clinical trial design, of course looking at the epidemiology of the patients who have the condition in which you are trying to answer a research question around, look at the demographics of the research team as our team has published that. Racial ethnic concordance influences, attitudes and behaviors towards clinical trials. We actually demonstrated that in a randomized clinical trial. Look at your eligibility criteria and ask the question, where is my criteria actually systematically excluding certain populations? Look at the burden of your protocol and how much of a long commute you're asking your patient to participate in, and then think about study sites and materials and the topography of the study site, how how easy is it for the patient to actually navigate that study site and the accessibility when it comes to content. Other study sponsor considerations, just echoing these themes is the, these infrastructure, the trial design, trial site selection, and then lastly, this notion of community partnership, which really fundamentally is listening to providers working in our communities, their pain points and what trials they perhaps need, to fulfill what they're seeing clinically to be a demand among their patient population. And also having an I ear on the ground for advocacy groups and other organizations.


So we've been thinking about recruitment through this talk thinking about recruitment practices, and I just wanna highlight that given the heterogeneity clinical trials are so variable given the heterogeneity research infrastructure and the US health system. The meetings of inclusion and research are multiple, and inclusion by itself does not ensure equity. And so I close with this very commonly used slide because I think it simplifies everything we've talked about in a digestible way, which is this talk has been about equity, of course. And when you think about these terms, let's first talk about inequality. What is inequality? And inequality is unequal access to opportunity. So here you see a tree with apples leaning towards one person, dropping apples for that person. Okay, so someone else is in a circumstance of disadvantage here. What is equality? Equality is evenly distributed tools. To access a circumstance. And what you can see here is evenly distributed tools does not necessarily make someone really overcome their circumstance of disadvantage. Equity is custom tools to help improve that circumstance. And now you see there's a custom tool of a higher ladder to help that person reach an apple, but yet that tree is still leaning to one. And so justice is actually what we're trying to achieve. And that's the north star here, which is to address the structural causes of these circumstances of disadvantage so that we can all access these opportunities. And let's think about that in the context of clinical trials, there's so much work to do, but really working together we can help achieve this mission. So with that, I'll take any question.


10 Key Takeaways about Health Inequity in Cancer Research

  1. Health inequity in cancer research refers to disparities in the prevalence, diagnosis, treatment, and outcomes of cancer among different populations, often based on race, ethnicity, socioeconomic status, and geographic location.

  2. Health inequity in cancer research is a complex issue that is influenced by multiple factors, including access to healthcare, cultural and societal norms, and historical and systemic biases.

  3. Health inequity in cancer research has significant implications for patient outcomes and physical health, as those who experience disparities may receive lower quality care, have poorer treatment outcomes, and higher mortality rates.

  4. Addressing health inequity in cancer research requires a multi-faceted approach that includes increasing access to healthcare, addressing implicit bias in the healthcare system, and increasing diversity in clinical trials and research.

  5. One of the key steps in addressing health inequity in cancer research is to better understand the root causes of health disparities through research, data collection, and analysis.

  6. Addressing health inequity in cancer research requires collaboration and engagement among stakeholders, including patients, healthcare providers, researchers, policymakers, and advocacy groups.

  7. Strategies for addressing health inequity in cancer research must be tailored to specific populations and contexts, as the factors contributing to health disparities can vary widely depending on location, cultural norms, and other social determinants of health.

  8. Increasing diversity among researchers and healthcare providers is critical to addressing health inequity in cancer research, as it can help to identify and address implicit biases and improve cultural competence.

  9. Policies that promote equitable access to healthcare, such as Medicaid expansion, can help to reduce health disparities in cancer research by increasing access to cancer screening, diagnosis, and treatment.

  10. Addressing health inequity in cancer research is an ongoing process that requires sustained commitment and action from all stakeholders to achieve meaningful change.


Hala Borno, MD - About The Author, Credentials, and Affiliations

Hala Borno, MD is a medical oncologist and assistant professor of medicine at the University of California, San Francisco (UCSF) at the Helen Diller Family Comprehensive Cancer Center. Her clinical practice focuses on treating people with advanced gastrointestinal cancers. Her research interests include health disparities in cancer care and coming up with new ways to treat these cancers. Dr. Borno got her medical degree from the University of California, Los Angeles (UCLA). She did her residency in internal medicine at the University of California, San Francisco (UCSF), where she was also the chief resident. After that, she did a fellowship in medical oncology at the Johns Hopkins Hospital and a fellowship in health services research at UCSF. Dr. Borno has been honored for her work in cancer research and patient care. She received the American Society of Clinical Oncology (ASCO) Conquer Cancer Foundation Merit Award and was named a Forbes 30 Under 30 in Healthcare in 2020.