Read the Full Video Transcript

Ruchika Talwar: Hi, everyone. Welcome back to your UroToday's Health Policy Center of Excellence. As always, my name is Ruchika Talwar and I'm an urologic oncologist at Vanderbilt in Nashville Tennessee. Today, I'm really excited to be joined by my friend and colleague, Dr. Amy Reed, who was a former endourology fellow at Vanderbilt and is now an endourologist at Brook Army Medical Center in San Antonio. She'll be joining us today to discuss some work she recently published exploring diversity, equity, and inclusion in clinical trials.

Thanks, Dr. Reed, for being here with us today. We really appreciate it.

Amy Reed: Thank you so much for having us. We're really excited to share our work.

Well, our paper was looking at diversity, equity, and inclusion specifically in urolithiasis clinical trials over the last two decades. And so what we know are that racial, ethnic, sex disparities can impact healthcare outcomes and this is important to focus on those to really improve patient outcomes across the board. And it's been well documented previously in oncology trials and specifically urologic oncology clinical trials that there's disproportionate patient enrollment specifically involving minority patients. So we had the question of is this also true in stone disease?

So we wanted to look at urolithiasis clinical trials over the last couple of decades to see is it truly representative of the urolithiasis population. So we sought to characterize these trends in patient enrollment by race, ethnicity, and in participant sex.

So what we did is that we queried PubMed and we found that there were 180 clinical trials over the last two decades to look at and what we looked at through each of these clinical trials was the enrollment of their patients by a patient and participant in sex, ethnicity, and sex enrollment data. And then what we wanted to compare that to was the established and known stone prevalence based off of each of those racial groups.

And so we found that 40 of these 180 clinical trials published their enrollment data, but for race and ethnicity and 105 of these 180 clinical trials published their participant sex data. And so to identify the representation and to see how adequately represented these groups were, we created a representation quotient that's been described previously where we looked at the trial enrollment of each of these trials and we divided that by the stone prevalence of each of those groups. And so with that, a representation quotient of one would be that all of these groups were adequately and proportionally represented and enrolled versus over or underrepresented. And then we stratified first by the just overall enrollment in all of these clinical trials and then stratified by study type or what they were looking for in each of these trials, their geographic location by a lay group, and then by the funding source.

And so what we found, this chart here, the four different sections here where this is looking at the representation quotient that we calculated. The red dashed line on each of these charts is the representation quotient of one. So if at that line that is a group that is adequately represented based off of what we're looking for in each of those subgroups. So anything over that line is an overrepresentation and anything under that line is an underrepresentation of that group.

So block A is just overall looking at all of those clinical trials and looking at each racial and ethnic group in each of those trials and you can see how well they're represented there. B, we stratified by funding type of all of those studies. C is when we stratified by geographic location and we broke that down by looking at the AUA group that had published the trial or if it was a multi-institutional study. And then group D, we stratified by a type, either a dietary, medical or surgical interventions or if it was an imaging study.

And so what we found is overall, white patients and participants that were enrolled were adequately represented overall and by each stratification type. Black participants or generally overrepresented in patient enrollment and then Hispanic patients and either if they classified as mixed-race or other classification in terms of race or ethnicity were generally across the board underrepresented and underenrolled. And then when we did look at sex enrollment, male and female patients were both adequately represented across the board. And of note, only one study of those 180 had a category for non-binary sex other than male and female that was reported.

So what we ultimately came to the conclusion is that specifically Hispanic mixed-race, other race patients are underenrolled or underrepresented in the trials that are reporting the data. And that is one area where we can really seek to improve recruitment and target these higher stone risk populations. And then the other component of it is that investigators should aim to improve the publication of the enrollment data. Only 25% of the studies that we found actually published the data of the patients they enrolled in the first place. So improving what was actually making it to a final manuscript and how we're communicating those results in terms of the breakdown of patients.

Ruchika Talwar: Thanks, Dr. Reed. This study study's really interesting and I think your last point there is one I just want to expand upon a bit. These sorts of studies do give us insight in where we can do better, but one big limitation is that we're not including all clinical trial data. The issue is a lot of people are not even to the stage of reporting race, ethnicity in their clinical trial data. So that's one big area where we can push for some improvement. The other thing that I found really interesting was that unlike some of the studies done in urologic oncology, Black patients were overrepresented. However, we saw a significant drop in enrollment in the Hispanic population. So that is another targetable goal that I think that we can aim towards as a urologic community. I'm curious to hear your thoughts on some potential solutions.

Amy Reed: For enrollment of these groups. So that is something that we recognize more the need for rather than necessarily being in the scope of what we're finding. First, we want to identify where those barriers are, where those goal of those gaps are, where those holes are. And I think it was surprising to us because you're absolutely right. It's a different enrollment pattern than we see with oncology trials in some ways, in an excellent way. But in others now, we're seeing underrepresentation in specifically the Hispanic population, which is of particular importance because we're also seeing data coming out where that is the highest increasing prevalence of the stone population.

So the first thing is recognizing is that this is a really underrepresented population and that's what we need to start targeting is the reasons why and why are we not reaching those patients? What can we do to target those populations to have them enrolled in these studies so we can actually apply these results and be more generalizable to that group.

Ruchika Talwar: Yeah, absolutely. And I know we're just hypothesizing here because you haven't explored that data, but I wonder how we can help overcome some of the language barrier that I presume is probably playing a role here. Now, I'm curious. It would be interesting if we were able to access the data from the trials that did not even report race, ethnicity if perhaps we would see a shift in this trend of overenrollment of minority patients because of Black patients, rather, because I think that just by reporting and making a conscious effort to make sure that race, ethnicity data is included in your study, perhaps you're of a research group that's more aware and trying to enroll minority patients across the board. So I'm curious, do you think that could be a reason for your findings?

Amy Reed: It is an interesting question and it would be really fascinating to know those results because I think you're very much onto something of if you're already thinking about it, is that going to be reflective in what your data is enrolling and showing? I am not sure it would only because we had a couple of trends that we found as far as looking over the last decades and looking at temporal trends and how things were reported. Because a big issue that we found was the categories that patients end up getting lumped into, especially if they either identified as mixed race or other, just another minority group that has very few patients and so breaking that down into having every ethnic group represented is just... it's a way of making that more concise data points for publication. So you have a lot of other groups or mixed race groups or categories.

A common trend that you would see was these binary trends. So you see it with sex and male and female patients with only one study having a non-binary option, but we're also starting to see it in racial groups where they would report white versus not white patients. And so just binary reporting of race or ethnicity, which obviously just really groups a huge population of people into just one category, which you have no idea what that's really representing and has a huge outcome. But my point was that we were commonly seeing that the trend of binary reporting was increasing over the last several decades. And so you're seeing it's more common. Who knows what's actually were collected, but the reporting is being a little more reductive or as investigators are probably trying to be more concise in reporting their data, but it's actually becoming more reductive in how they're reporting and categorizing patient demographic data.

Ruchika Talwar: Yeah, I hear your point. I think it does make sense to simplify things in terms of analyses and reporting, but it becomes challenging when interpreting these results for specific populations. So definitely something we can improve upon again and I know this study is just the start and it's exploratory and it's more thought-provoking than conclusion reaching. However, I want to congratulate you and your team for undertaking this really important study. It explores a topic where we need more literature to guide efforts so that we can provide higher quality care for all of our patients.

Amy Reed: Yes, I agree.

Ruchika Talwar: Thanks again for joining us and to our audience, thank you so much for tuning in. We'll see you next time.