Now we're ready to move on, and just really delighted to bring up our next panel. If you guys want to transition up here with me, we're doing a session this morning that's a little bit of a review of a session we did with support from the Doris Duke Foundation, the Gordon Betty Moore Foundation, in consultation with the National Academy of Medicine back in June on this same topic, on the topic of reconsidering race and clinical algorithms, a really important topic for our specialty societies. And we invited some of our former speakers to join us today to reprise their prior sessions, but we also invited someone new to join us today. So I'm going to turn things over to Dr. Cindy Escobar-Alvarez from the Doris Duke Foundation, who will introduce her panelists, and this will be an extraordinary session, I promise. Thank you so much. Thank you. Well, good morning, everyone, and thank you for having us here for this important discussion. Really delighted to be here. And let me start here. So I'm Cindy Escobar-Alvarez, Doris Duke Foundation, and we believe at Doris Duke that doctors and patients deserve information that they can trust. And there are, let's see, big green button. Every day there are, and you all know this very well, that there are interactions between patients and doctors to diagnose disease, to assess risks, to determine what treatment course to take. And sometimes that those decisions are, there's a consideration of race. There is a race factor or a race coefficient that may be factored into those decisions. Sometimes we believe perhaps to the detriment of the decision or detriment of health equity is what I would say. And when you look at where that guidance comes from, some of it comes from what people learn in medical school, some of it comes from medical societies, and that has just, is just what has been agreed upon over the years. But when you really look at where does the root cause of it, where is this coming from? Many times it's coming from research, right? And studies that were done a long time ago where scientists loved data and they came across race as a variable and factored it in in a way that could perhaps facilitate decision making or would improve the prediction of a model as it was assessed at the time. Now we believe that that practice needs to be revisited, the research practice, or why was that consideration put into place at the time? And to do that we need, it's an effort of many, it's not an effort of any one particular group. For example, funders and publishers can encourage it, medical societies can also encourage their members and especially those who develop guidelines, clinical guidelines, have a pulse on what is that underlying evidence that may need to be revisited. So at Doris Duke we believe this is very important so that we can have better tools for patients and doctors. And earlier this year announced grants to $10 million to several groups that are addressing this in different ways. Some of them that are members of CMS, CMSS, and they are demonstrating that we need research and we can answer these questions of how do we reconsider race in clinical algorithms through research. And they of course together have a reach of thousands of individuals, health professionals, and researchers who can make that assessment for what do we do and how do we get the word out about how to reconsider race in a way that furthers health equity. Earlier this year we also partnered with CMSS as well as others to talk about this issue, to learn more about it. Why is it important? What can we do about it? What are some groups already doing about these? And what does AI also can do to help us further our knowledge and sharpen these tools that need to be revisited? And I invite you all, CMSS has this really great report that summarizes those discussions, our learnings, and more important, recommendations. Recommendations for all. I love that this is very much in keeping with the theme of the meeting this week of let's drive meaningful change together. So let's do that by identifying recommendations to pursue. There's some very specific ones for medical societies. Share about what you're doing, share about your actions, and spread the word. And certainly we are trying to do that at this Adores Duke so you can follow us on our several media channels to hear what is happening. And with that, let me stop there and I will introduce our panel who really will speak to why this is important in very concrete ways. So I will introduce the three of them first and then turn it to our first speaker, Dr. David Jones, who is a psychiatrist and historian and a Bernard Ackerman Professor of the Culture of Medicine at Harvard University, where his teachings at Harvard College and Harvard Medical School explore the history of medicine, medical ethics, and social medicine. He's currently at work on several other histories, is very busy on the evolution of coronary artery surgery, of heart disease and cardiac therapeutics in India, and of the threats of air pollution and global warming to health. Then we have Dr. Joseph Wright at the American Academy of Pediatrics, who is a pediatric ER physician and the first chief health equity officer at the American Academy of Pediatrics. He oversees and implements the academy's national strategy to promote and achieve health equity for all children, as well as efforts for the academy to become an equitable, diverse, and inclusive organization. He previously was a chief health equity officer of a 12-hospital, 30,000-employee University of Maryland medical system and was professor and chair of pediatrics at the Howard University College of Medicine, as well as senior VP at Children's National Hospital. And last, we will hear also from Dr. LaTunza Adams, who is a clinical pathologist, medical educator, mentor, and medical advocate, and associate professor at Oregon Health and Science University. She teaches didactics to pathology residents, is a co-director of the pathology residency program, and a director of microbiology rotations. So with that, I'll pass it to you, Dr. Jones. That's not my slide deck. Okay, good, there we go. There was a disclosure slide a minute ago. I don't have anything to disclose, nothing to worry about there. So as Cindy said, we're here to talk about one of the most basic questions facing doctors across the medical specialties, which is should we provide different healthcare to individual patients of different races and ethnicities because of their race and ethnicity. My own work on this has focused on the issue of clinical algorithms that alter the outcome based on the race or ethnicity of the patient, but this is a problem that is pervasive throughout medicine. I've been asked to kick things off by talking about the history of how we've ended up in the situation where we are now, where race plays a very important role in many of the clinical algorithms, treatment guidelines, and diagnostic tests of American medicine. The problem is a very old one. The idea that different humans, different types of humans have differences that were relevant to medicine goes back to the very earliest years of colonization in what became this country. Thomas Jefferson, for instance, famously described many differences between the white people and the enslaved Africans with whom he was familiar. He described many specific differences in physiology, including perhaps most famously the claim that the pulmonary apparatus of Africans was defective compared to that of white people. Doctors followed that lead when pulmonary function tests or when spirometers were introduced into American medicine in the mid-19th century. U.S. physicians described differences in the measurements they saw between white people and in their language between full blacks, mulattos, and Indians. Something similar happened with x-rays. William Wrenchen famously described the first x-ray in November in 1895. Within two months, American physicians were arguing that black people required higher exposures because their skin was supposedly thicker and their bones were supposedly denser than other humans. This intuition got built into American radiology. General Electric, which produced most of the x-ray devices used in this country in the mid-20th century, advised radiology technicians to adjust the exposures they gave to patients whether the patients were deemed to be normal, easy to penetrate, or hard to penetrate. Black and brown people who were declared to be hard to penetrate required higher voltages and exposures than other humans. This caused a bit of a scandal in May 1968 when it came up in passing in Senate hearings about radiation safety that featured Ralph Nader. In response to the outcry about this practice, the National Center for Radiological Health looked into this question, decided that there was no good evidence base for race-specific exposures in diagnostic testing, and recommended that the practice be stopped. But the race adjustments for pulmonary function tests, which also had gone back to the 19th century, persisted. And meanwhile, many other race-adjusted tests continued to proliferate in American medicine from the mid-20th century until quite recently. Now the history of how all of these race adjustments got introduced has not been well studied. From what I have seen, there was never controversy when someone recommended that a race adjustment factor be introduced into diagnostic practice. There has only been controversy recently as advocates have argued that these race adjustments should be removed. Now I came to recognize this problem slowly. I had trained in psychiatry, a field that famously has very few diagnostic tests or clinical algorithms, but even there I became aware of some of the many race-based practices that exist in medicine, especially in psychiatry relating to race-based recommendations for drug dosing of psychiatric drugs. And it doesn't take too much work to dig into the psychiatric literature to realize that many of these race-based practices were not that well justified. I reengaged with these questions more recently in spring 2016 when I was approached by a series of first-year medical students at Harvard Medical School who were concerned about how race was discussed in the curriculum at Harvard. Specifically they were puzzled to see that race adjustments existed for several common diagnostic tests, estimated glomerular filtration rate, pulmonary function tests, and the FRAX tool that is used for the diagnosis of osteoporosis. They wanted to dive into the data sets to see if these practices were well justified. And working with them, we decided to pursue several lines of inquiry. For instance, the historian in me was interested in a straightforward one of tracing backwards So, as a historian, I was interested in tracing the citations backwards through time to see where these practices came from. And if you do that, you quickly recognize a problem. Many of the rationales for the various race adjustments could be traced back to the old days of 19th century scientific racism, with slavery-era claims about thick skin, dense bones, musculature, or, as I said before, the defective pulmonary apparatus that had been cited by Thomas Jefferson. Another medical student, Darshali Vyas, approached me about a different tool that she had encountered on her obstetrics rotation, an algorithm that predicts the risk of attempting a C-section in a woman or a person who has previously had—sorry, a risk of attempting vaginal delivery in a person who has previously had a cesarean section. This tool, you can see the formula there, adjusts for both African-American race and Hispanic race—that's their language—and again, you could trace this thinking backwards through the literature, again, to slavery-era claims about racial differences in the shape of the pelvis, the idea being that white women had a pelvis that was ideally suited for childbirth, and African women had a pelvis that was ideally suited for physical labor. Now, as we were doing this work, it became clear that there were many people who were interested in this question and pursuing critiques. There were groups at University of Pennsylvania, at UCSF, at University of Washington, and they were all working to cast doubt on the use of race of several of these tools. And for a variety of reasons, one tool, the estimated glomerular filtration rate, got the most attention. Now, as these critiques were starting to emerge, the group of students at Harvard and the students at many other universities quickly became frustrated because of the nature of the resistance that they were getting. Going toe-to-toe with some of the people who had created these tools was a very frustrating exercise, trying to debate with the people about the evidence-based rationale. We weren't equipped, as a group of medical students and a historian, to make the case on the grounds of the data sets that justify these practices or not. And the other thing we realized is that hospitals had begun to ask us to do modeling studies. What would be the impact if the race correction was dropped from EGFR before they were willing to consider making a change? But of course, hospitals hadn't demanded modeling studies earlier when the race correction was introduced to begin with. And this reflects a pervasive asymmetry in American medical thought. Race is often assumed to be relevant unless it can be proven otherwise. So in response to this pushback, I developed a new strategy, one that didn't focus on critiques of the individual tools, but looked at a much bigger picture approach to critique many of the problems that each of the tools has in common. Many of the tools, as I mentioned earlier, have a weak evidence base. Many of them can be traced back to 19th century theories of scientific racism. The implementation of these tools often leaves much to be desired. They rely on a crude distinction between black people and white people, or black people and non-black people. And often, if a tool is used as directed, they would actually exacerbate health disparities that already exist in this country. The simplest way to demonstrate this is with one of the tools, the stone score, that was recommended for use by emergency room physicians to predict the risk that someone has a kidney stone if they present to the ER with flank pain. The tool rates five factors on a scale of 0 to 3. 13 is the highest score, indicating the highest risk. Now some of the factors, if you look at them there, make sense. If you have blood in your urine, you get three points. You shouldn't have blood in your urine. Stones cause blood in your urine, so that one is perfectly sensible. But what about race? If you're not black, you also get three points. That's as significant a factor as hematuria. Now maybe this is possible. You could go back to the data set that generated the stone score to begin with and try to understand how this was manifest in the data set. But it's certainly implausible at first pass. And then worse, if you use this tool, it will systematically direct medical resources away from black people and towards white people, because it's the white people with a higher score, other things being equal because of their race, who will be recommended for a CAT scan and a further diagnostic evaluation. So working with one of the students, Darshali Vyas, who is by this point an intern at MGH, we decided to look into this. And so as she did her internship rotations, she accumulated a list of race-adjusted tools that she came across as she admitted patients to the wards. And so by the fall, she had come up with a list of about 25 different tools that incorporated race. And so the two of us sifted through these and picked a list of 13 that seemed most promising to analyze. The tool gave us enough information that we could see how race was used. All the tools we chose were ones that, if used as directed, would make disparities worse in the way the stone score could. And so we decided to shoot for the moon and submitted to the New England Journal of Medicine. And much to my surprise, they accepted this. The article was working its way through production at the Journal in May 2020, when George Floyd was murdered by police in Minneapolis. New England Journal of Medicine decided to fast-track this article. And it was published in June while the race riots were still active, the race protests were still active in this country. Now the article, in part because of this timing, generated a lot of media attention. But I want to emphasize a few points here. First, we were not the first people to discover this problem. There were many people who were working in parallel on this problem because the problem was obvious enough eventually to be recognized by many different kinds of people. We just happened to submit a review at the right time. We did not call for race tools to be abandoned. Many of our critics have said that's what we did. That's not what we did. We said that these tools should be reconsidered. People needed to look very closely to see if any of them were justified. Now of course, our suspicion was that if you looked at these tools closely, you would see that they were not justified, but strictly speaking, all we called for was a reconsideration of how race was used in these tools. Now a lot has happened in the past three years, and you'll hear more about that from the two other people on the panel. So I'll leave that story there. Another thing that was very useful about this is I benefited from a well-placed ally, one of my colleagues, Michelle Morse, who I knew because she had been a hospitalist at the Brigham and Women's Hospital in Boston, at this time was a legislative fellow embedded within the House Ways and Means Committee of Richard Neal, a Democrat from Massachusetts. And so Michelle had been working on these problems before she went to Washington, D.C., and used her position on the House Ways and Means Committee to really push this problem onto the agenda of both the House and the Senate. She was able to orchestrate hearings in the summer of 2020, and then was one of the instigators behind a request for information that was sent by House Ways and Means to the leadership of various medical societies, asking them to explain exactly what they were doing with race in these tools and why. Now because this letter came from the House Ways and Means Committee, which controls the American budget, they had a 100% response rate to this request for information. These have been posted online, so you can look at them. Many of the societies, notably the American Medical Association, disavowed racism in medicine and committed themselves to anti-racist reforms. Many other groups established working groups to examine their tools carefully. The two that have gotten the most attention were the working group that reconsidered EGFR and the other one more recently that reconsidered pulmonary function tests. Both of these have now officially recommended that race be used from these tools, but implementation is difficult. There are roughly 6,000 hospitals in this country. Each hospital will individually have to recode its electronic medical records and its laboratory computer systems to get race out of EGFR, and the same thing will have to happen with pulmonary function tests. All the spirometers that exist in this country are going to need to be reprogrammed to get race removed from these tools. And then there have been a few medical societies who have defended how they have used race in their tools. I'll be happy to say more about that in the Q&A. Now while obviously gratified by this progress, I also realize that the basic problem in American medicine persists, which is the way that we use race and ethnic categories. Some of the tools we looked at used the official Office of Management and Budget categories of five races and two ethnicities, but most of them relied simply on the black-white distinction. There is no scientific justification for dividing the human population into two groups, black and non-black. There's just no basis for that. And the basic problem here is you can't make good medical science based on a distinction that itself is fundamentally flawed. It's just not a useful way to parse human diversity. What medicine needs to do is develop much more sophisticated ways of thinking about the difficult problems, the difficult challenges of human diversity. We could develop better ways to characterize the human population instead of relying on a simplistic black-white distinction. We could dispense of the use of these self-identified categories and invest and try to figure out how to make more sense of ancestry through ancestry informative markers, but that will be problematic for various reasons. There's been a lot of recent work criticizing concepts of ancestry and how they've been used in medicine. Many people have recommended shifting away from genetics and inheritance and focusing on experiences of racism. That would be productive, but then again you'd need to develop tools that would somehow allow you to capture and operationalize how people have experienced racism, and that won't be easy to do. While I think it's important to keep race in the descriptive statistics of medicine so that we can map health inequities, I'm not recommending that we sweep all the problems with race in this country under the carpet. It's just a question of what we do with it. It is essential that we continue to track the many race, ethnic, gender, class disparities that exist in this country, but we should be very wary when we use that kind of data to impact or inform the care of individual patients, as if an individual person who self-identifies as black somehow embodies all of the characteristics that have been described in that population of black people in this country. We have to be very sensitive to these huge problems of miscategorization. Human populations are becoming more complex as old social barriers break down. There are many people in this country who would identify as multiple races. If you were to take someone like Barack Obama, I suspect that every hospital in this country would have coded him as black, he self-identifies as black, and they would have race-corrected his EGFR, and race-correcting the EGFR of someone like Barack Obama makes too many errors for me to catalog here. Again, I'd be willing to explain that more in the Q&A. There is no doubt that people of different races experience different kinds of health care and health outcomes in this country. I'm not denying that. The question is, what's the best way for medicine to respond to these differences and to understand this? When you see race differences in health or disease or treatment outcomes, is that because of some race-based difference in biology? Is it about a race-based difference in exposure to social determinants of health? Is it a race-based difference in the ability to comply with physical therapy or cardiac rehab, a difference that might reflect the structural racism in the American health care system, or whether you have the time to follow up in your life with physical therapy three times a week? Some people can do that. Some people can't. The ability to comply with those kinds of treatment programs depends on issues of race and class and income and everything else. Meanwhile, any effort that we invest in increasing our efforts to understand these problems of race and racism ought to be matched by efforts to increase the sophistication with which we handle things like socioeconomic status. We need longitudinal data sets that control for socioeconomic exposures over a lifetime and sophisticated analyses that allow physicians to understand the contributions of the many aspects of human experience to health disease and health outcomes. Now I understand that it will be very difficult to overhaul the knowledge base of American medicine, which is entirely structured at this moment by these simple race distinctions, especially of black and non-black. We need much better data if we're going to understand these problems and make progress. I understand that this is not easy to do, but I'll remind everyone that we have, as a population, as a community, we have done many difficult things in the past. We have sent people to the moon and brought them back alive. If American medicine sets its mind on the goal of doing better with the challenge of human diversity, we will be able to do so. Thank you. Thank you, David, you're always a hard act to follow. Well, good morning, everyone. Again, my name is Joseph Wright, and I am a pediatric emergency physician, and I really am delighted to be here this morning, and I thank CMSS for the invitation. I have nothing to disclose. But I do want to acknowledge the Doris Duke Foundation, as Cindy mentioned in her introductions. The American Academy of Pediatrics is one of the grantees whose work is being supported by the Doris Duke Foundation in this area, and I also want to acknowledge our project team. I am here representing them this morning as principal investigator on this project. I also, in absentia, want to acknowledge our CEO, Mark Del Monte, who, as you heard from Helen, is one of the incoming board members here at CMSS. So the American Academy of Pediatrics is focused on the health and well-being of infants, children, and adolescents, also young adults. And I mention this because pediatrics has long legs. And if we get things right for kids, a rising tide floats all boats. Everyone benefits. However, the converse is, if we get things wrong, the developmental trajectory of the kinds of assumptions and misinterpretations of which David spoke have lifelong impact. In terms of my own journey, as Cindy mentioned, I am a pediatric emergency physician, and I mention that because I am often personally challenged when I think about the emergency department and acute care settings where the use of clinical practice guidelines is designed to shortcut and to really build upon implicit biases that we all have. And there has been a long and deep literature around the inequities that are embedded in acute care medicine and emergency medicine in particular. And I think about my own training and think about the acceptance of status quo and not challenging dogma in the way that David's students have in this generation. And it has really emboldened me as my career has moved towards now to the far right of this slide where my job is to focus on equity science. And I emphasize that term science because as David mentioned, the perpetuation of racialized and unsubstantiated science is at the root of what we see in so many inequities that result in health disparities and we have to acknowledge that and have to be willing to dig deep to understand that. So the AAP is a policy-driven organization and I share with you this series of policies over the last several years that have focused on this work. I have been fortunate as a board member of the AAP before accepting this position to have had a front row seat on the early steps of the American Academy of Pediatrics equity journey and you can see here pre the murder of George Floyd that the impact of racism and child on child and adolescent health was our foundational policy statement that continues to be the most highly cited and authoritative piece on this subject matter and really led to following the murder of George Floyd the accounting and reconciliation of our organization around our own journey and looking in the mirror and seeing where we have made mistakes historically and moving forward that work has now led to an emphasis on the work that I'll be describing today. Before I dive into the project itself let me just share with you the pediatric UTI story because it is one that has evolved over the course of the last three years and I think gives us some insight as to what the medical specialty societies can do and should be thinking about. So this is a clinical practice guideline that was originally published in 2011 and reaffirmed in 2016 by the board and for any of you who has had a young child under the age of two with a fever. This is a very common workup and a very common diagnostic exercise and you can see here embedded in the clinical practice guidelines algorithm that there is a dichotomization of what to do based on race and in fact that CPG clinical practice guideline led to the development of a calculator a calculator that all learners have on their devices and this calculator as you can see here again incorporates racial identification as a dichotomizing or differentiating tool in order to perform the standard test of a characterized urine specimen. In 2020 one of our members called this out and here you see the article which was published in JAMA Pediatrics that challenged the use of race in this clinical practice guideline and race was inappropriately inserted as a default biologic proxy here in lieu of incompletely explained and incompletely explained epidemiologic observation. Now I know these investigators well. These are my colleagues. This work has been done in the emergency department setting and I do not believe that they woke up one day and said we're going to discriminate against little black girls. However they did not go deeply enough into explaining what they were seeing and lobbed in race as a variable that got them the answer that they were seeking. And so I have had the opportunity to speak to my colleagues to present to them just like I'm doing to you today. And this is not about intent. I'm not here to call you a racist which is what the initial knee jerk reaction was. But it's about impact. If you cannot tell me that we are not generating a cohort of little black girls who will grow up with scarred urinary tracts then we've got to apply equitable care to all children. So we took this as the board retired the state retired the statement in 2021 and is feverishly working on a new clinical practice guideline. However in the meantime the authors of the calculator I showed you the UTI calc 1.0 went back into the laboratory and looked at their own data and published this meta analysis in June of 2021 where they not only reexamined their work but work from around the world and arrived at this statement that in replacing the variable of race with the clinical variables of history of UTI and duration of fever the accuracy the predictive accuracy was similar with the previous models. And I say academic humility because when we first approached this group who are very well known very well known clinical researchers in our field they were resistant. They did not want to take the calculator down. This is proprietary tool it did not cost to access their tool. But again was widely used and after publication of this meta analysis and this reexamination of the worldwide data they have since posted UTI calc 3.0 where you can see that the race variable has been removed and replaced by these two clinical variables. Now one might ask the question why was were these two clinical variables not included in the logistic regression modeling in the first place. But that's the subject of another presentation. But this is again the kind of thinking the kind of reexamination of the science that is necessary from within our specialty societies that is necessary to reexamine and get at more equitable answers in our clinical practice guidelines. So that's the UTI story. Our approach with the Doris Duke Foundation grant is more broadly oriented. Our project is really focused on designing an approach to eliminating race based medicine and in all of our clinical algorithms and inform how we develop future recommendations not just within the AAP but but being able to work across specialty societies. Again there are many opportunities with how we deliver clinical care in the pediatric age group that have implications for other societies in later that that are focused on individuals in later life and to a point that David made. We have to also be willing to test what we are proposing. And so one of the opportunities that we're using the Doris Duke resource to support is a testing of a new UTI clinical practice guideline in a clinical setting which I'll describe in just a moment. So the theoretic framework and the project phasing is as you see here. We are right now in the discovery phase. The AAP publishes not only clinical practice guidelines but policy statements clinical reports technical reports. It's a large body of work that needs to be frankly scrubbed from stem to stern. But how do we go about doing that. How do we generate an initial pool of policies through which to get started. And again as David said in his room we've got to get started somewhere. This is a huge lift and I know it sounds cliche but a journey of a thousand miles begins with but a single step. And so as we think about how we revise either existing statements or tweak statements that are coming up the academy has a periodicity of every three to five years of looking at all of our our statements for review. Our goal is to establish clear criteria and rationale for examining these statements the impact the visibility or lack thereof of inequities that produce disparities. We are partnering with a clinical academic clinical partner to again test in a high volume pediatric acute care setting. One that we select from that candidate pool and then very important which is even beyond what we have stated to Doris Duke that we will do is this idea of making sure that there is a bibliometric footprint because there is a paucity of information in the scientific literature that that lays this out the work of David and his students is literally laying the ground the foundation for this work. So we all as grantees have an obligation to build on that thought leadership and to really describe from methodology standpoint why this is important to talk about new terminology. We are introducing in our project the concept of a race conscious approach again to the point David made. And this is a this is a terrible analogy for a pediatrician to make but we can't throw the baby out with the bathwater. We absolutely absolutely have to continue to examine the impact of race and on on outcomes. And so this is really important when we talk about moving from a race based to a race conscious approach. So what have we done. We've seeded a rapid revision team the RRT. These are members from a variety of specialties from generalist to subspecialist who are all academically based. You see their attribution institutions there and and importantly we have also seated with these individuals who are are largely mid career scholars who have devoted their scholarship to equity science also a member from our family partnership network. This is incredibly important. All policy development at the academy does include input from families. So Miss Lightfoot who has previously done work with us is embedded in this team as well and we are excited to get started. We've had a few meetings already and what we're what we are faced with in our patient facing policies is a large I'm sorry I thought that might have OK I guess they didn't get the most recent version loaded but nevertheless the large compendium of evidence based guidelines that drive pediatric practice and what you see here is that the clinical practice guidelines constitute the smallest number but the most important number because the impact for instance of a of a urinary tract infection clinical practice guideline is huge. Clinical reports are the next largest number and then other policies that are patient facing include technical reports and endorse clinical practice guidelines. And I emphasize that because the AAP does endorse other societies practice guidelines. And so this is where we have interface and collaboration across sister societies. We are not tackling the endorsed CPGs just yet but we hope down the line that we will be able to interface. So the starting point is to review 153 policies to get out of the gate. What the RRT's activities and progress to date look like is that we're refining the methodology to identify race based AAP policy. There is not a uniform systematic way to do this. So we are in the process of refining this tool that you see here we're using the red cap the red cap platform in which to assign each policy independently to two rapid revision team members. And by the end of the year we will hone it down to three policies that we will prioritize and revise. And when I say that we are not trying to replace the policy development apparatus that is already in place at the American Academy of Pediatrics. We would hope that the rapid revision team supplements the authoring groups at the academy and also the peer review mechanism that looks at the policies which is actually the board. So what kinds of hurdles have we experienced and this is obvious that there are limitations in how we go about searching what is inherently biased the typical key search term or even natural language processing tools don't really get it. They are just the tip of the iceberg. So we developed through a modified Delphi process a set of criteria and this is a screenshot of our of our whiteboard that we through we worked on the problematic definition of race and statements is often not obvious. And so we went through a live exercise of applying this rubric that you see here on the slide to a clinical report and that was very fascinating to see the range of interpretations as we move through that exercise and we recognize that when we look at the total denominator of policy some 400 plus statements that again this is a heavy lift but one where if we identify a systematic approach we can then apply it as statements come up for their regular review to all of these statements that constitute are constituted in this this compendium. This is the 23rd edition here and as I said contains more than 400 policies. So this is really what from a visual standpoint I'm a simple emergency medicine guy. This is what we're faced with. The things that appear in algorithms and equations are easy. That's what's above the surface. So getting at those opportunities is really a small part of the work ahead. What is more challenging is the more insidiously embedded historically based bias and discrimination that David described and we are trying to apply a race conscious approach to get at finding out in the context of not knowing what we don't know. So this is the work ahead. I've introduced a term that perhaps some of you have not yet become familiar with. It's again an example of where we have to expand the literature moving from race based to a race conscious medicine. This is literally the in the Lancet 2020 and it's interesting these are students from Yale who published this seminal paper and I think it's I don't think it's coincidental that the folks who have approached David the folks who I am now working with these early career folks are leading the way and pushing all of us to do the right thing here. And when we talk about the road to a race conscious approach it's important to understand some very basic premises that are inherent in the way that we teach medicine and you see them laid out on the bottom of this slide that race is all too often embedded as a biologic construct and that is step one job one as we begin to change the way that medicine is taught. So what are our opportunities here. Again standardizing an approach and describing a methodology. That's really what we are tasked to do with the Doris Duke Foundation support and then scale that process for broad application with all the authoring bodies within the AAP first but what's most exciting for me is the potential for partnerships across specialty societies that incorporate that developmental life course perspective that I described earlier. For instance we have a very extensive statement on the management of hypertension in children. One of the other grantees in our cohort is the American Heart Association. So at some point it would be wonderful to align what we are doing for children with regard to early factors for cardiovascular disease with the recommendations of the AHA. I'm just just saying Cindy I'm not you know that's the next grant right. OK good. So in the background as I wrap this up in the background of all of this is this work that we are rolling out for our members. Words matter equipping our authors and contributors and what I'm most proud of with this particular contribution to the AAP board policy manual is that it recognizes that this work is dynamic and fluid which is sometimes a hard thing to express in the context of a big specialty society. All too often we are we lay down the law and then there's an inertia. We can't change that or we can't revise this and until until the next cycle. This was a step by the board to simply say that this is this is a dynamic process and evolving process so we have to we have to be flexible as well and we have to be able to evolve the very words in the language that we use. So I'm proud of the board first of all for embarking on this path which is you know rather new I've been a member of this organization for 30 years and this level of flexibility has not always been the character of the organization or for any of our organizations and also just released about a month ago right here across the street at the convention center we just had our national conference and exhibition is this compendium and this is notably the editor the editors here are the original authors of the policy statement from 2019 on the impact of racism on children adolescents have developed this compendium that I had the privilege of writing the forward for and I share this with you because this is the kind of for our learners it is organized in a way that is modular so it can be used for learners for instance as part of a noon conference yet it's also encyclopedic in its nature if folks really want to use it as a reference tool and these are the kinds of offerings that we need to begin the rollout to our members to begin to socialize the language the concepts and as I often say we often are preaching to the choir when we have a very large congregation in our case 67,000 members who have to be read in to this work so that concludes my remarks that's a picture of our our initial in-person meeting and very excited to work with these young folks and again want to thank and acknowledge the Doris Duke Foundation for getting this started thank you good morning. My name is Dr. Latanza Adams and I'm from Portland, Oregon, so the change in the climate has affected my voice. I'm a little froggish right now. Like I said, that's a little not better, okay? Excuse me. Again, my name is Dr. Latanza Adams. I am from the VA Portland Health Care System and I'm also associated with OHSU, Oregon Health Sciences University, as was discussed earlier. I was asked by CMMS to tell the story of my father's journey as a consequence of race-based testing and to elaborate on what the College of American Pathologists is doing to address this issue. I call my portion called The Human Impact. As far as disclosures, I have none, and because I do work for the government, I have to read this. The views and opinions presented within this work are those of the author and does not represent the official positions of my employer or the United States government. Charles Clinton Adams. Who was this man? This whole started with me joining what's called PCEC and the College of American Pathologists. It's the Professional Community Engagement Committee. Excuse me. We were trying to figure out one of the goals is to increase the presence of pathologists in the community and education and kind of bridge that gap. One of the initiatives is KnowYourPathologist, yourpathologist.org. We were initiating trying to figure out what we could do to reach the public. During this time, it was a huge change in the EGFR calculation. I happen to mention that because we were implementing the change in our laboratory. I told my story, and they said, well, we would like to put that on there as a way to engage with the public. They wanted to know about this. I just wanted to give a brief because while we talk about all these societies and the things that we're changing, it's really easy to get into the minutia and the technicalities of it. At the end, there's a human. There's a patient. There's a person. One of those was my father. He was born in Gary, Indiana. During his high school years, he was an apprenticeship for mortuary school. Really, I wouldn't be who I am today, literally, if it wasn't for him. Before he went off to Vietnam, he was, like I said, an apprenticeship for mortuary school. What he would do is, my actual first autopsy was with my father. I was seven years old. My little Mary Goldfish died. I was upset about it. He said, well, you want to find out why he died? I said, yeah. He said, well, we're going to do an autopsy. I said, well, what's that? He said, well, it's a way we can find out why people and things, what happened to them. We got a piece of towel paper, and we laid my little fish out, got a little utility knife, and sure enough, we found one of the little pebbles at the bottom of the aquarium stuck in his gills. He said, see, he passed away from respiratory failure. I said, that's amazing. Then just to go on, in the summer I would do academic programs, and there was a raccoon that was hit on the side of the road. Every day he would tell me, if they don't remove him, this will happen, that will happen. I learned the stages of decomposition from him. You can say my path towards pathology was kind of unconsciously geared that way because of him. He had a strong work ethic. He did not, before he could continue his studies in being a mortician, he got drafted to Vietnam. While in Vietnam, he was subject to grenade detonation, and he had shrapnel in his back, and he almost lost his arm. He had to be airlifted to Fitzsimmons Hospital in Colorado. Afterwards, when he returned, he became a fireman. He said he'd seen too much death, and he wanted to become a person that saved lives. He was a fireman, and before he got sick, he had 28 years on the job and was the fire chief of the Gary Fire Department. That's just a little bit about him. That's my mother, cannot not mention her. Her name is Beverly Adams, and that little peanut, that's me. About my father and his journey, why I'm here, he was a diabetic and had hypertension. In June of 1997, I heard a thud on the floor, and when I opened my bedroom door, it was him. He suffered in a stroke. From that, he did therapy. He did physical therapy, occupational, and his whole goal was to get back to work. He wanted to get back to work. He got better, and his cognitive skills improved, but then we started to notice between 1997 and December, especially the year 1998, he started emotionally not being the same and cognitively not being the same, and we didn't know what was going on. My mother, she is a retired phlebotomist, and she noticed that his behavior was similar to the behavior of the patients that had kidney problems. She kept bringing it up that, hey, something's wrong with him. He needs to see a nephrologist. During that time, my father's insurance was through the city with his fire department, and they didn't allow you to just say, let me call up a nephrologist to go and see this person. He had to be referred, and people kept saying that, oh, he's just depressed because he was a fireman, and he did all these great physical things, and he's not able to do them anymore, and they tried to put him on antidepressants, and he wasn't having it because he didn't need it, but time kept going on. My mother kept pushing. No one would listen to her, and finally it got to the point he couldn't even remember the simple alarm code. It was to the point we had to make it 1, 2, 3, 4, and he couldn't remember that, and it was just something was not right, and we just went. We gave up with our health care area and went to Chicago, and we went to the University of Chicago emergency room, and immediately they came back and said, your husband needs dialysis, emergency dialysis. He's in end-stage renal failure, and that's how he was diagnosed. Afterwards, he was put on peritoneal dialysis. That lasted for a while, but during that time, heparin was under investigation for the FDA, and with peritoneal dialysis, sometimes you can get what's called a fibrin clot in the port, and recommended treatment was to put another bag in him. So about this time, we had like three bags in him, direction of the clinic. We had to bring them in. They put another bag in him. Eventually, he developed peritonitis, really bad infection. He had to surgically get things removed. He suffered more strokes, was in the hospital that summer, and had to permanently use hemodialysis for his hormone dialysis. And from 1999 to end of 2000, he just had a series of events that did not, wasn't positive. He never really acclimated to the hemodialysis. When we first was diagnosed with renal failure, we went to Northwestern to say that, hey, we are willing to do a transplant. I was going to give him my kidney, and they shut us down and said, well, no, that's nice, but he doesn't have enough reserve right now for that. So let's get him stronger, bring him back, and then we'll discuss it. Well, we never got that chance. He developed MRSA infection. He had fluid on the lungs and ended up having to go to ICU, and three days later he passed away. Now, whether the change in the algorithm back then would have mattered and prevented that course of events, I don't know, but he would have gotten a chance. Now that I'm on this side of health care and looking back, from at least in 1997, they should have known something. Something should have appeared. You wouldn't go to direct renal failure in that short of time. So something wasn't catching their eye for that. Not one time was kidney failure or kidney chronic, anything with his kidneys, was brought to our attention. And it could be that he didn't present sick enough for any interventions at that time. One of the things with the EGFR equation is that for blacks, you have to appear much more sicker for your GFR to be to the point where you'd be considered for any interventions or transplants or anything like that because it was falsely elevated. And so while you clinically may be sick, your numbers didn't fit the criteria, so you weren't considered for that. But in this situation, at least he would have been given a chance. This year at CAP, I met with the CEO of the Kidney Foundation and one of their representatives that they have come and speak. And one of the things that she mentioned that was very profound as a result of the change in the EGFR equation, what's being done is that they are doing retrospect looks at people's values in EGFRs, and they're re-stratifying where they are on the kidney recipient list. And she mentioned that she had a very close friend that before the change, she was nowhere near getting a kidney. After the change, immediately she got a call and was able to get a transplant. So this is a profound thing, and I think one of the things that is amazing about this is, as the panelists said, it was not a question of adding this race into it, but taking it out is a crazy feat. So to have this change to be so wide and to be so pushing in this momentum is a very good thing because it's something that has questioned, I know perplexed me from my training, like why are these factors here? Like where does it go back? But, you know, you're in the midst of the training. You just kind of like blindly accept it because you're trying to get through the next step in your process. So as a pathologist, what is CAP doing? What is the college doing? So from the creation of the video on my story, we created a video telling my father's story. And this was then amplified to different ads targeting areas of people of color, those areas of high-risk kidney diseases. These are in like your Hulu pre-ad streams. At the annual conference that year in New Orleans, there was a large press briefing related to that and the awareness and the benefits of non-race-based testing. We have patient education initiatives on www.yourpathologist.org, patient advocacy partnerships with the National Kidney Foundation with content. Also, there was a creation of the podcast Addressing and Eliminating Bias in Laboratory Medicine. Also, there were advocate statements and calls to Congress that resulted in an invitation to the Congressional Black Caucus Health Brain Trust, where there was discussing health equities and disparities. Also, there are many media opportunities that came from this. Me and myself, I did a health special with the local news station for Channel 5 or NBC Back Home. Also, there was a news special done in Louisville, Kentucky, with our media past president, Emily Volk, and other radio opportunities. I just wanted to end that race is a social, not biological, construct. That's my father. Also, too, just as a pathologist, what we can do, we all mention the different things, for instance, changing of the UTI makeup and how you approach it. One of the things that we do and I love about pathology is that my field, especially clinical pathologist, touches all of you all. One day, I'm a rheumatologist. Another day, I'm a cardiologist, pediatrician. It's a great opportunity to be involved. Whatever you do, you have to come to us because we're the ones that have to change the equation. We have to change the lab reports. That's why it's important to reach out to us as pathologists. There's a big push in us getting from out of the basement, out of the microscope, into your lives. I'm that type of pathologist that I am on the committees. When I learned of this, I reached out and I said, hey, what are we doing? I was very fortunate to know that for the National VA, they are putting this in place. It was my job to make sure we had the testing appropriate, my job to make sure that we had the wording for you all, and then especially with the ease of patients obtaining their reports, make sure that it was available and easy for them to comprehend as well. As well for cystatin testing, to make sure we had that ability to do so and other hospitals could send to us. Our role is very important, so use us. We are wanting to be a part of this process, and I think what's important is that we all are in it together. Thank you very much. Excuse my voice. Have a good day. Well, thank you to our speakers for all of that, and thank you, Dr. Adams, for sharing your personal story with all of us. We have some time for a few questions, and please find your ways to the microphones if you have questions, and I'll get us started here with one that I think struck me about, that came throughout all the presentations, and that it seems at each point you've all encountered resistance to revisiting the evidence base, and why do you think that is? And I know who wants to start. It's a question for you. One thing Joseph Wright had mentioned, which is that the people who have created these tools, I have every reason to believe, are 100% conscientious people who are trying to do the right kind of evidence-based medicine. And then when they get criticized as the way that we've criticized them, many of them feel like they have been accused of being racist, and that obviously triggers hard feelings for reasons I'm totally sympathetic with. And it becomes a difficult strategic challenge for the advocates, which is how do you engage with people who have done something with good intentions that might have had bad consequences, and you want to engage them to change things without saying we think you're wrong or we're a bad person. It makes it a very difficult dynamic. And I think the one way around that is to say, well, look, why is it that race emerged as a signal as people did these multivariable analyses of their data sets? Part of it is because it's coded in every data set in this country. You cannot find health data in this country that's not coded as black, nonblack, or better ways than that. But find me a data set that's coded by income. Find me a data set that's coded by zip code. We just don't have those kinds of data sets. So the data that we have is asymmetrical, and that pushes certain kinds of variables, especially race, to the forefront. And that's what researchers have been responding to. And that's why I said it's so important to change the data sets that we have to allow researchers to find things that are the true signals. In the case of the UTI calc, you know, the duration of symptoms, a history of prior UTI. If we had better data sets, people would be able to do analyses that would avoid the pitfalls. And just to amplify on that, you know, just a very human element, the scientists are very personally tied to their work. And the getting beyond the initial resistance of criticism, as David said, one of the ‑‑ we were talking beforehand. What I am finding is that in socializing this work, we tactically have to be very much focused on the evidence, on the science. Not that the moral imperative is not obviously incredibly important. But in order to get people's attention, and in order to bring people like our clinical researchers at the University of Pittsburgh around the UTI calculator, in order to get these folks to either reexamine or examine more closely and critically and more authentically the factors that influence the outcomes that we see, it's very important that in audiences like this that we skew to the science, to the evidence. And while one would hope that the presence of historical inequities and the disparities that we know well would be enough, when what we are experiencing and engaging in the academic ecosystem, we have to be as focused on the science as these individuals believe that they were. So that's often a difficult challenge when folks are first approached. I'm really loud, though. I can just be loud. Thank you. Hi, Elizabeth Dry. I'm from the National Quality Forum, and also trained as a pediatrician. And I just first want to thank the panel. Just an unbelievable, powerful set of presentations together. And thank you, Dr. Adams, for sharing your personal story and inviting us to work with you and pathologists going forward. I wanted to raise something which I'm a little hesitant to raise, because every time you raise this at a conference recently, you potentially go into, like, a whole other space, which is artificial intelligence and our work that many of us are focusing on right now to make sure we get the use of that in medicine right. And I'm really almost feeling like this is the easier piece of that work because the algorithms that we have and the way they're put together to date is more transparent a lot of the times than what we'll be doing with AI. And plus the people putting together are the people in this room and colleagues who are in the research setting and will be moving more towards work driven by data scientists and really large language models where we can't necessarily see underneath them. So I wonder if you could speak to the opportunity we have right now to try to make sure as we go forward that we address these issues. There's time pressure on it as the National Academies does its current look at NQF. We're also convening an AI panel focused on potential biases and quality measures. So I would love your thoughts about how we can get ahead of this even as we have so much work to do here and your leadership has been so critical. Yeah, I am concerned on the iceberg slide about what's below the surface. And until we have a better handle on how bias and discrimination shows up in the denominator of patient facing guidelines, a better handle on what unsubstantiated or racialized science has contributed by diving into those references that have been kicked down the road for decades and continue to be cited until we have a better understanding of how that is showing up. My concern is that all AI will do is amplify the bias that already exists. So I almost feel like the work we're doing now is we've got to clean up the reference base. And that's why I emphasize the importance of building out the bibliometric footprint on what drives the development of much more scientifically authentic guidelines. But I hear you. I think the presenters before us talked about how we are glacial with regard to how we don't keep up with the science with regard to our guidelines. So I think there has to be a meeting in the middle. At this point, I think it's too large a leap to suggest that AI is a solution for what we have as a huge, and I say this, I don't mean to be glib, it's taken 400 years to get here and we're not going to fix this tomorrow. So we have to be very deliberate and we have to be very careful and we have to be very thoughtful about the steps that we take now so that at some point we can become more nimble and faster, quite honestly, in fixing these. Two quick things I would add is, one, since the inner workings of these algorithms are often black boxed because of the way the AI works, the best thing we can do is to look at their outputs and see what kinds of things they're doing. And so if you develop a tool and you look at what it would then do with individual patients, pay very close attention if you think this tool is actually making disparities worse in the way that the Stone score does. And if it does, then you really need to look carefully to see if you stand by the analysis it's producing because you can get in a lot of trouble. And the best way you can head that off prophylactically, again, is through the data set. If the training data has patient populations stratified by age, male, female, and black, non-black, the algorithm is going to take those variables seriously in its output. If you give it a much more complicated, nuanced data set that has 50 or 100, 200 variables, physicians often don't have access to those data sets. The health insurers do. I had a very interesting conversation with the chief information officer at Aetna. He said, if you want social demographic data, we have it. We know a vast amount about our subscribers, which, of course, stresses us all out. But the data exists. They have data on income. They have data on occupation. They have data on street address. They know everything about us. And if you use high-resolution data, again, it's not going to affect the race. Race is a co-variable with nearly every other exposure in this country. Income correlates with race. Educational attainment correlates with race. Place of residence correlates with race. You can't remove that from society. It is part of our society. But if you give the algorithms better data that take more variables into account, you're more likely to get a better algorithm. Thank you. We have time for one more question. Hi. Todd Ibrahim with the American Society of Nephrology. I just wanted to thank the panel and all the work that you've been doing in this arena. I thought I'd provide sort of an update or maybe just to build on what you've said within the EGFR. Can you not hear me? Better. Sorry. Just to sort of build on what's happening with EGFR. So one point in the sort of nomenclature I've started using in my head because we are starting to get a lot of press calls. So Dr. Wright's point about really letting the science drive this, which is what we tried to do. There's still a narrative out there, just a fundamental misunderstanding about race not being biological. And what I've started saying to reporters as they call it or try to do both sides of the story is that there are not two sides of the story. That the world is round. It's not flat. And we don't argue that. And I think that's a really important concept here. And while I agree with Dr. Jones that there was an intention in terms of the people developing these algorithms, there are a group of physicians, many of whom are very senior physicians, who fundamentally are not accepting that reality. And I don't quite know how to have that discussion. So that would be a question just to circle back to. I think for Dr. Adams, just to publicly thank her because her efforts working with the National Kidney Foundation, who is our partner in this, and they're the big patient advocacy organization. So they've been working directly to increase adoption in the labs. It's taken about three years for more than 50% of the labs to adopt the new standard. That doesn't sound great, but the previous standard, the racist algorithm, it took 10 years to adopt. So the fact that it's been accelerated is because of efforts like Dr. Adams's. So just to publicly acknowledge that. I think the final point to make, and I'm probably going to drift into some problems here, is that at the end of her presentation, Dr. Adams mentioned what's happening. There were two algorithms in Dr. Jones's article related to kidney function. One was the EGFR, which we're trying to address. The other related to transplant. And it was both the ability of black Americans to be transplanted, to be prioritized on the list, as well as to be prioritized as living kidney donors. And what's happened is that's been slow-walked by UNOS, and I don't know how else to say that. So while they've publicly made the commitment, they haven't done that in their activities. And so any medical specialty society that works with UNOS is a favor if you could just communicate to them that you'd like an update and you'd like them to move faster. To me, it's unacceptable that the rest of the community has moved quickly in this arena, as quickly as possible. More than 50% of the labs have adopted this mechanism, et cetera. But on a really important issue related to transplant, for reasons I don't really understand, and maybe their intentions are good, although I don't think that's true, they have moved slowly. And I think we have a responsibility to the patients and to future people with kidney disease, with kidney failure, to have them transplanted, and this applies to other transplanted organs. So just to raise that. I don't know if you wanted to respond to the first question or not. Thanks again. I just want to thank the nephrology community for really getting out front in a very visible way and to the point that we've all made about the importance of transparency and developing the bibliometric footprint. And I say that because the society community is an accelerant. In other words, I've spent the last 10 years in the C-suites of three health systems. And until the societies state unequivocally the policy, the clinical policy, it's hard to get the attention of decision makers in health systems. And so the work of the nephrology community and to your question about what to do with some of the, we'll just say the senior subject matter experts who are often convened to develop these policies. One thing I noticed with regard to the American Thoracic Society's recent statement from the spring that recommended moving away from race-specific equations in PFT laboratories, they published the dissenting opinions. And that was fascinating to me because we don't often do that. Generally, there's a consensus and we don't know what happened in the room. And I just want to read this because I find it fascinating. In that statement, there were three people who disagreed with that recommendation from the ATS. And they state that the recommendation conflicts with their view of heritable differences and body proportions between people that were set by different ancestral climates. Now, the fact that that is in the literature, it now allows us to come at that statement with fact, with science. And so, again, I applaud the transparency of the societies for publishing the proceedings, how we got to this recommendation and in the case of the ATS in a way that's remarkably transparent that could very easily be countered. And also, what's also interesting to me is understanding the disconnect. I have only begun to appreciate how instrumental the transplant list community is relative to moving this work forward. I had seen recently that there were black patients in the Cedars-Sinai system who were pushing a class action suit because there had not been movement on the transplant list in response to the corrections. And so, there's so much that we all have to learn. And I just wanted to make that comment. I may have missed the question, but I just wanted to. Thank you so much for that. So, clearly, we could speak all day about these. And thanks to our speakers. And if we can be with one thing is that these corrections, these issues, we have to look for them, right? They're not evident. So, let's all try to do that. And thanks again to our speakers. And thank you all for having us here. I just want to thank our panelists and thanks to Sydney for moderating. That was an extraordinary panel. Again, we've just got up on the screen the report that we did from our session in June. You can scan to read the report. There's an entire section in that report on AI, by the way, for those who are asking about AI. We've got a little bit of time. We extended to session. It will begin at 9.40. So, if you want to go to your concurrent sessions. Please note you're welcome to attend the Diagnostic Excellence sessions including the poster walk that's beginning now, as well as the Diagnostic Excellence panel discussion at 11. Those are the grantees that got funded as part of our efforts from the Johnny Hartford Foundation and the Moore Foundation. So, thanks again.

race in clinical algorithms

negative impact

healthcare

reevaluation of practices

outdated research

scientific racism

evidence-based medicine

health equity

collaboration

stakeholders

clinical algorithms

kidney stones