 All right, I think we're going to go ahead and get started. Welcome to the latest edition of our interdisciplinary faculty seminar. It is a pleasure for me to introduce Professor Sonia Starr, who is the Julius Krieger Professor of Law and Criminology at the University of Chicago Law School. Her work focuses primarily on issues related to racial and socioeconomic inequality across various areas, including the criminal process, treatment of people with criminal records, employment discrimination, and DEI issues in education and health care, including a particular focus on statistical discrimination and algorithmic fairness issues. Her research methods include quantitative empirical analysis, as well as doctrinal and theoretical analyses of legal issues. She teaches courses related to criminal law, constitutional law, racial inequality, and serves as editor of the Journal of Legal Studies. Professor Starr will speak today on race and clinical practice guidelines, a legal perspective. So join me in welcoming her to the McLean Center. Thank you so much, Dr. Angeles. And thank you all for being here and to the McLean Center staff for all of their hard work. So I'm going to start with this. Wait, why isn't this advancing? It was advancing a second ago. There we go. OK. So I'm going to start with this caveat. I'm a law professor. I'm going to talk about the law so it might sound a lot like legal advice. The law that I'm going to talk about is within my academic expertise. But I think that medical providers should familiarize yourself with these areas of law. But I don't want to breach the ethical norms of my own profession. And so I am required to tell you that I am not your lawyer. And if you want medical advice, you should go to somebody who is and that I'm not even a member of the Illinois bar. And so this is academic commentary and not legal advice. So OK, so today before I turn to the law, I want to spend some time on an overview of the sorts of practices that I'm focused on, which have been the subject of extensive debate in medicine itself, in other health professions like neuropsychology and in medical ethics. Less so in law, so I'm trying to add a legal perspective. Some of you may already be familiar with some of these practices either because they intersect with your own area of practice or because you've been following the debate about racialized clinical practice guidelines in medical fields. But some of you may not be familiar. So I want to spend the first part of my talk just kind of going over some examples and laying the groundwork for what we are talking about. So there's various ways that race gets taken into account in modern medicine. And I've assembled here a loose and probably not exhaustive taxonomy. So the main focus of my talk today is going to be what you see here under Roman numeral 1, which are instances in which an individual patient's race is taken into account in determining something related to the course of their individual treatment, whether it's a specific treatment recommendation, a diagnostic decision, a prognosis, et cetera. And in particular, within those categories, and so everything that falls under part one here is what in the law we would call disparate treatment, which is to say taking race into account, race itself into account in deciding how to treat a patient, as opposed to what lawyers call disparate impact discrimination, which is subtler and refers to racial disparities that are produced by reliance on race-poor mated characteristics with nominal races. So here I'm talking about sort of the least subtle race in which racial bias is produced into practice. That is when race itself is basically used as a variable. So mostly, I'm going to focus on things that fall into categories 1A and B here. Race norming or correction of observed values or use of race as a risk predictor or a predictor of treatment success. So these are examples of where sort of quantified uses of race in which some sort of algorithm is explicitly using race as a variable in order to make some kind of prediction or a recommendation. So those often fall into the category of what's labeled clinical practice dialogizer CP or CPGs. They can be used for diagnostic decisions, prognosis, or care recommendations. And those have been the central focus of the recent ethics debate. And before I get to the law, to the legal perspective, let me note that there's a slightly subtle distinction between A and B here. Most of the debate, most examples that have been used that have been centered in the public debate has concerned race norming category A here, which is what I'm going to mostly focus on myself sometimes referred to as race correction. And what that means is that some characteristic is observed for a patient, like, say, how much air they can breathe out or their performance on a cognitive test and dementia diagnosis or the quantity of some chemical in their blood. And then the characterization of whether that measurement is abnormal or how abnormal it is adjusted based on race. It's done relative to the population distribution of that characteristic broken down by race. So you ask, is it normal for a member of the person's race? Race norming has been particularly criticized because it often has the perverse and widely critiqued result that members of racial groups in which adverse health traits are more common. The people who disparately suffer some health condition become less likely rather than more likely to be able to access care. And I'll give an example of that in a minute. There's a couple of academic pieces out there that defend the use of race and clinical algorithms. Well, there's a bunch that are specific to particular algorithms. But there's a couple of pieces out there that sort of generically try to defend the process. Most of them don't really focus on race norming, which is often harder to defend. Instead, they focus on category B, which is the use of race as a risk predictor. And here, again, race is being plugged into a formula in a way that can shape care decisions, but it's not really a matter of norming. It's about taking population level differences and the prevalence of some health characteristic into account in predicting the outcome, say, the outcome of some medical procedure or something like that. This practice is still going to be legally still racially disparate treatment and also has raised some controversies. But it's worth noting that the consequence of employing race in that way is often the opposite of what happens in race norming examples, because often it elevates the risk assignment that is given to higher risk groups as opposed to masking the higher risk that certain groups experience. For that reason, it might have a better chance of overcoming legal challenges on grounds of medical necessity, et cetera. Even so, can be critiqued for overly focusing on race as opposed to other mediating characteristics like socioeconomics. More on that in a minute. Let me take a second briefly before I move on about the other two categories on this slide. So first, category 1C here. So I'm going to focus today as has the medical ethics debate on quantified algorithms. But of course, it's possible for race discrimination in medicine to play out in less formal and less quantified ways as well. And that would raise similar legal issues, although often would be harder to prove, since there's no algorithm written down that, say, a potential plaintiff could point to in court. And there's all kinds of reasons that health care providers might discriminate based on race or mechanisms by which they might do so, just as there are to explain discrimination in any other area of society. But one that I want to highlight here briefly is the prevalence of medical myths about essential racial differences. Ideas that remain common in our modern world, even though they are grounded in a vision of race as a biological category that is largely quite outmoded academically and which also has very problematic roots that sort of had its high point in the slavery era and is highly tied to the history of white supremacy. So doctors, even highly educated ones, are not immune to, or even, well, I guess all doctors and even recently educated ones are not immune to these myths. So for instance, a widely cited study from 2016 by Hoffman et al showed that 40% of first year medical students and 25% of fourth year medical students believed falsely that black people have thicker skin than white people do. The authors reasonably suggest that this prevalent false belief might be one reason for the widely documented fact that black patients who present similarly with self-reported pain symptoms have more trouble getting access to pain medications than white patients with the same symptoms do. So the myth that black people are more impervious to pain has specifically an ugly history. It was invoked in the slavery era to justify his brutal treatment. The prevalence of myths like this is tied to a view of race as fundamentally a biological category rather than as a social category. And that vision of race has proven sticky, right? Hard to get rid of entirely, even though it has been undermined time and time again by evidence that for the most part, race isn't really a meaningful biological category. Which is not to say that there's no genetic differences between any world populations, but the way particularly in the United States where people of different races often represent wildly disparate groups actually that fall within categories of race as we have constructed them, in which multiracial identities are common. Like, these are not silos in which health traits are very readily categorized. And in fact, although there are many population-level health disparities that exist according to race, most of them are much more strongly explained by mediating socioeconomic variables like racial stratification in poverty and in health access. So finally, let me mention this last category, which I'll return to at the end, which we might describe as race consciousness in medicine rather than as racially disparate treatment. So here I'm talking about medical or health policy decisions that are conscious of racial disparities and seek to reduce them, but not by making the treatment of individual patients turn on their individual race, but rather by making overall policy decisions that are informed by information about the hurdles that particular groups face to treatment and that seek to reduce those hurdles or seek to avoid adopting rules or standards or practices that don't have a strong medical justification but have racially disparate consequences. So I'll turn back to this category at the end of my talk. But the preview basically is that this category of race consciousness is legally quite different from individually racially disparate treatment. And my view is that, at least under current law, that although there is a sort of radical colorblindness movement that's pushing to extract all race consciousness from position making, that at least under the current law, that kind of thing is much more easily legally defensible. OK, so having gone through this taxonomy, I want to jump back to category 1a, where I'm going to spend most of my time and to give one example. So I'm going to focus on this one example that's well explored in medical literature, in the current ethics debate, and even in sociological and historical literature, which is the measurement of lung capacity using an instrument called the spirometer, which is used for many purposes in pulmonology and for triage and ERs when people come in reporting breathing difficulty. So patients will breathe into a spirometer. It measures how much air they can exhale, which is called the lung exhalation volume. But in a modern spirometer, the readout doesn't just tell the doctor the actual exhalation volume. Rather, it shows a figure that is normalized by race as well as a handful of other demographic characteristics, age, sex, and height usually. So basically, it shows how much the patient can breathe relative to others of their race, instead of relative to people of all races. So this is like a classic race-norming example. These images are from a book that's not by a doctor, but it's by Lundy Braun, she's a sociologist. I think this is an excellent history of science book. Like it talks about how spirometry measurements came to be racialized and the surprising ways in which spirometry has influenced a bunch of different areas of medical practice and other social phenomena, including legal accountability for workplace pollution, for example. So on the right, this is also an image from her book. You can see, so that's like a really old spirometer. That's like a newer spirometer in which you can actually see physical switches, one of which is labeled race. And the newest spirometers have more modern digital interfaces and may call less attention to the race adjustment than that physical switch does, but they are functionally similar in the algorithms that are provided. So in a paper by Gaffney et al. in 2020, the authors work through, give this illustrative example, which I will borrow here. So assume you have two patients, they're both male, six feet tall, and 40 years old, and they both come in complaining of shortness of breath and they both have an actual exhalation volume into the spirometer of 3.5 liters. So if the patient is white, this volume, then that's what it's called FUE 1, will be flagged as low by modern spirometers. That's because it's 78% of the average that's predicted for white men who are six feet tall and 40 years old. If that patient is black, a spirometer that has a race correction will categorize the same FEV 1 as being in the normal range. 91% of what's predicted, which is considered within the realm of normal. Now you might think, okay, but surely this adjustment wouldn't be happening unless it helps doctors to make better care decisions, but there doesn't seem to be strong evidence of that. Just, again, this is not my own judgment, this seems to be what the medical literature has been recently concluding. There's a couple of studies that assess the prognostic significance of FEV 1, adjusted and unjusted by race. And both of these studies find that the unadjusted volume of actual amount of air that our hypothetical patient is exhaling is the thing that predicts their later prognosis, right? And that you get worse predictions if you make the adjustment by race, that there's basically no racial difference in the prognosis of FEV 1. So why on earth are they making this correction? Well, it's not that it comes out of nowhere empirically, right? It is the case that black people on average holding the other demographic factors constant have lower lung exhalation volume. But you can't just stop with that average comparison. You have to ask, why is this so? Why do black people on average exhale less air? The race correction implies a bio-luxury. It implies that it is normal for black people to exhale less air. And thus, it is not a health concern unless a patient comes in and presents with an FEV 1 that is not normal for black people, right? But there's overwhelming reason to believe that the answer to this question is principally socio-economic and environmental as opposed to a biological. So black patients on average again at a population level face health disparities from the womb, right? Like because of less natal care, which might affect their lungs later on. The biggest factor probably in explaining this particular disparity is differences in pollution exposure, both in the home and in the workplace throughout people's lives. Black people have far higher rates of asthma and basically every pulmonary condition as a result. But that point doesn't make these differences normal, right? It makes it a problematic health disparity. And so what the race correction does by treating it as though this disparity is normal, is like it takes something that, there's something wrong with recognizing disparity, right? Like with having data that shows that the disparity exists, but rather than treat it as a problem to be addressed, for example, by making sure that black patients get access to pulmonary care that they need. Instead, this normalizes it quite literally, right? Uses it as the benchmark by which individual patients' health will be assessed. So this is not a small correction. And so it has consequences that are not small, right? So extra menino do a study, like they draw data that included spirometry measurements from a broad-based population health site. So not a pulmonary patient sample, but they're just looking for essentially the prevalence of pulmonary issues in the general population. And they find that if you use the race correction, right? So if you use the numbers that the spirometer spits out normally, you find unsurprisingly, because of the race correction, you find unsurprisingly that the rates of pulmonary impairment are similar across white and black populations. You have in the vicinity of 9% of all people in the population showing some degree of lung impairment and a little less than 1% showing moderate to severe lung impairment. But what happens if you don't use the race correction, right? If you don't use the race correction, by the way, it leaves there's no correction to the white values. Those are always considered left unadjusted. If you don't make that race connect correction to the black figures, then you get a measurement of 37% showing some degree of impairment and 1.7% showing moderate to severe impairment. So that means that the race correction leads to three quarters of mild impairments and over half of more serious impairments in black patients being missed. So that's quite consequential. Okay, so I'm not gonna work through other examples in that kind of detail, but I just wanted to flag a few other recent controversies. One in neuropsychology has probably gotten the most attention of anything else. And that's because it was at the core of a stand-alone role in the family and it's a settlement for concussions. So the concussion settlement required patients to get assessed for dementia. And the tests that they used, the NFL was sort of quietly behind the scenes rejecting, telling the settlement administrators to reject any submission of test results that had not been subject to appropriate demographic adjustments, which included race morming. Effectively, and the NFL ultimately didn't deny this characterization, this race morming process assumes that the black players started at a lower cognitive performance level and therefore they have to show much more secure actual impairment in order to qualify for it now. And the NFL defended itself by saying, well, like this is just the norm in neuropsychology and just following the rules. And although that's a little bit oversimplified because it's perfectly vocal within neuropsychology there not, and it was quite cynical of the NFL. Many physicians and psychologists that submitted assessments had them sent back because they didn't need to stand in the NFL and imposed. Still, they're not entirely wrong. These are pretty much the athletes in neuropsychology so they did a considerable debate there. So the NFL did eventually back off of it under a massive public pressure once the two players sued them and they got extensive media coverage. So in nephrology there's been multiple different debates about race correction. The most prominent one involves the use of phlegm and the blood to estimate kidney function. And the race correction there has the effect of reducing kidney disease diagnosis in black patients. In obstetrics there was a calculator like a popular tool that was used to calculate successes of vaginal birth after C-section. And that often there's a pretty simple higher rates of failure for black defendant, or not defendant, sorry, I teach criminal law, so. So sometimes I might say the wrong things. So black patients were predicted to fail and it steers them toward C-sections. The calculator I'll point out is not really an example of race warming. It's more of an example of that second category that I pointed to of race as a risk prediction category. And arguably isn't from a solid idea of biological difference, but rather just a recognition of disparity in outcomes which I think is a disparity that really exists and that is largely socio-economically mediated or mediated by differences in access to care. And so those who defended it at the time said, look, if black patients attempting B-backs really do have worse outcomes, shouldn't we be honest and say so if we're giving them medical advice about what path to pursue. But there were various examples with this answer so it got some pushback in the literature in that field in obstetrics. And what is that? It identified a health disparity that was real, but rather than respond by trying to improve care and outcomes for black patients and say like what do we need to get our black patients to a place where they can have similar success with B-backs as white patients do. And instead responded by steering them away from a care path that could benefit them. So you sections have higher rates of complications and much longer recovery time. So and second, it was question begging, right? So if any given hospital or doctor believes in its own ability to provide equal care to its own black and white patients, right? Then why hospital ever give different advice to its white and black patients about what to do based on race, simply based on the fact that in the aggregate in society there are care differentials, mostly across facilities. And if a particular provider of facilities seems like that is the problem to be addressed rather than just by steering the care of patients. So all these criticisms are not coming from me. They're from experts in their respective fields. And indeed the last two of these examples, the Nephrology Guidelines and Obstetrics are both examples of where the clinical practice guidelines have been changed within the past two years precisely because of these problems. And so there have been changes in some other fields like some in cardiology. We're just announced this past month. The kidney, the change in the kidney algorithm, it's worth noting is an example of where the solution that was settled upon was not simply removing race from the algorithm because a lot of people pushed back and said that it actually is less predictive as a measure. Creatinine levels are used as a proxy essentially for something that's difficult to measure. I'm going to mispronounce glum, you're a real lure, filtration rates. And so that proxy people argued was just not as good for black patients. But what ended up happening is, hey, basically found a difference was as effective and equally effective across racial groups to be used for the same like estimated purpose. And so that was adopted instead. Okay, so that's the background on the medical and medical ethics debate. I've said little about the law so far. But I think we now have to, today as it's been playing out in medical fields, has also not said very much about the law, right? So I have up here since it's to some good overviews of the medical ethics debate. They don't talk about the law at all. Legal solid is also not favoring with attention. There's a professor called Dorothy Roberts at Penn who has been a strong critic in the legal academy, although she's quite interdisciplinary and her approach is a little more sociological when I'm trying to add in my work, including this recent paper called statistical discrimination, is really a more traditional legal fact kind of to express what is the law I actually say about these practices. In this piece, I use this as an example along with examples of what we call statistical discrimination in other areas beyond healthcare. Okay, so to me it's a little strange that this debate has said so little about the law. Even though it's almost like medicine is heavily operating in the shadow of law, like fear of litigation, and also the direct effects of regulation, right? Affect many aspects of medical practice. So why not this one? My assumption is that basically every doctor knows that they're not supposed to discriminate based on breaks, right? If a patient, say a black patient came to your door and you said, sorry, I don't treat black patients, you know that you would be breaking the law, right? And yet the fact that many people are effectively being turned away from care or doing different direct mutations on the basis of their race because an algorithm includes race as a variable doesn't seem to like get people's antennas activated. Like there might be something wrong here, but I'm not arguing that there is something legally wrong here, right? So let's talk about sources of law, like what law governs the use of race in medicine? Well, first about a quarter of medical practice in the US occurs within public facilities, which are governed directly by the United States Constitution and in particular by the Eco Protection Clause of the 14th Amendment, which is the main constitutional tool that restricts the Constitution. Second, basically every other public medical facility in the country is governed by title six of the Civil Rights Act of 1964, which governs all entities that receive federal funds and applies very broadly. So like if any portion of a hospital and portion of a university receives federal funds, then the whole university or the full hospital is governed by title six. Section 1557 of the Patient Protection and Affordable Care Act also bars discrimination in healthcare. There are federal agency regulations implementing those laws, which could come to include, I should say, a proposed Department of Health and Human Services rule interpreting section 1557, which has some particular comments about clinical algorithms, which I'll get to at the end. And then states and localities have their own anti-discrimination laws. And these are all clear and untapped with each other ever, but the fact that the community of federal prostitution always comes with federal statutes from the regulations, all took the law of state and local law. But for the most part, these are non-identical, but fundamentally cutting in the same direction protections for individuals from discrimination. I want to focus on the constitutional system. We're going to ask, are you guys who work in the private facility? Well, it's because the Supreme Court has interpreted title six of the Civil Rights Act, which does apply to the University of Chicago and basically every health facility to provide ways of discrimination, protections that parallel the equal protection of law. So that is why, for example, in the recent affirmative action cases, the Supreme Court made no distinction at all between Harvard and the University of North Carolina to the defense in those cases. It just applied the constitutional standards to both of them. And that's because it pre-existing doctrine that says title six, federal forms of recipients are basically governed by constitutional law to raise discrimination. So I'm going to talk about the constitutional standard, but it does apply to private actors in healthcare. And I'm going to focus here on the law as the Supreme Court that is, which is sort of for better or for worse. I've been a critic of some of the courts, colorblindness type doctrines, but they're the law and you should know the law as it exists, right? Okay, so under the Equal Protection Act, the standard that applies to individual prostitutions, the standard of review is called strict scrutiny. Strict scrutiny means essentially that a court will look very close and then requiring a very demanding justification for racial classifications. Strict scrutiny applies when individuals are treated differently for one another based on a suspect classification. So I have highlighted things in a bunch of colors here, because I want to highlight some different things about that rule. So the first is that, this is about individuals being treated differently. The Supreme Court has a very individualistic model of what equality is to be. The Supreme Court never likes approaches to equality that talk about equality as between groups. They're very focused on individuals being treated equally without respect to their individual characteristics. So for that reason, we'll talk about their type of problem in the Supreme Court, which is the meaning of the reason is fair to people predicted after seeing a possibility to work with the Supreme Court. The Supreme Court is not planning to act in an assessment of what fairness means, or rather that concept of fairness doesn't figure into its conception of equality under the equal protection clause. And it's so equally important. These are the decision makers keeping the right factors and not just the families who treated the Supreme Court along with the potentially prohibited factors. And are they treating like individuals alike, right? So then this idea of different inner disparate treatment this I flagged at the beginning, disparate treatment discrimination means either express classifications like a law or a policy or that expressly categorizes based on race and trees people accordingly, or some form of intentional discrimination to be discriminated between the administration of official racial law. Disparate impact on a group alone does not under the Supreme Court's doctrine give rise to a prostitution. They're widely criticized by law scholars, but it's been a law for four years. It's not likely to change. So basically the court focuses on examples in which decision makers have treated like individuals different. So then the idea that it has to be based on a suspect classification. This is essentially a causation standard, right? So race or another suspect criterion doesn't need to be the only factor that the decision maker considered. It's not like, the plaintiff doesn't have to prove that the algorithm kicked every black patient out and allowed everybody to sit in at first for the one word factors. But if it can make a difference, right? If it's one of the factors that's considered so that people who are otherwise equal but are on the borderline might be treated differently because of their race, then that's gonna get it subject to strict scrutiny and a plaintiff may be entitled to a process in which it can actually focus the appeal. So then what is a suspect classification? For our purposes, what's important is race, also religion, national origin, sometimes citizenship status. Notice that gender, sex is subject to a lesser standard and categories like age are not subject to elevated scrutiny at all. So it's, and most types of black, I don't get what it's called, it's called rational system use. There's a difference, let's say there's a difference in the number of people that are on the borderline, so there's a difference in the number of people that are on the borderline. Generally problematic discrimination is totally prohibited in some contexts of employment, but it's not like the arguments against race and class and religion is what we have to do. I'm gonna apply this to all of the people that are on the borderline. It's just that special legal standards of pride and race is involved, and partly the reason for that is because of the ugly history of racial history and usually hasn't gone well to be full of government decision makers, or in this case, private, but federally funded decision makers to make decisions based. Okay, so what actually is the strict scrutiny test then? The rule is that disparate treatment based on race will only be upheld if it is narrowly tailored to advance a compelling state interest. This is a very demanding standard. Overwhelmingly, if strict scrutiny applies, courts will strike well, 100% of it's green. There are instances in which it's been upheld for decades, affirmative action of higher education was one of those instances. And the idea was that some affirmative action policies were narrowly tailored to advance a compelling interest in educational diversity. The court recently overruled that. So if anything, I think they've been moving more toward the fatal impact approach. The narrowing requirement requires ruling out of race-neutral alternatives. So the decision, the defendant in this case in which it's been ruled out, that they really comprehensively searched for race-neutral ways to accomplish the thing they were trying to accomplish and that all of the reasonably available such things would have failed to accomplish their interests. And that interest has to be compelled, right? And it's the defendant's burden to show that they considered alternatives, a good reason to reject it. And that their interest is compelled. And the court has been increasingly reluctant to recognize compelling state interests as opposed to the opposition's opinion affirmative action rather than increasing disparities. So that's the upside of the recent affirmative action. Okay, so if we can practice guidelines, how are they gonna do? I think often at least, maybe not, I'm not saying none of them can survive this test, but I think that there's a lot of obstacles that we just thought we could do to prepare to defend against the very search and review of the court, which means they're gonna need strong evidence. And sometimes I think at least that evidence is not gonna be there. So strict certainty is gonna be automatically triggered because they are racial classifications that's explicit in the algorithm. That means that the court needs to not be able to do anything. But if they can't help them, that's fine. And it's not there, and then there's a solution, right? The clinical practice guidelines are public. And so this is relatively easy to prove compared to other racial discrimination cases. I think proving a compelling interest would probably be difficult in many of the cases because the evidence is at best hotly contested and the strict scrutiny standard is not deferential to the expert's advice. It does not say in the list if that's what these courts have nothing to say about it. It actually says that you guys have a way to show that if you wanna use race, that there's a really strong justification for it. And then the other thing that I think these, they're these like recent studies and sometimes changes in the official CPGs based on those studies have illustrated the effectiveness of these mutual offerings, right? And so, for instance, that's the key to the end of the problem being in the guideline for estimated review in effect for a few years now. It's gonna seem that like some, I don't think it's exactly number one factor, but some non-trivial factor in fact is still used to get this test run and that's because it's just slow and definitely effective to change their balance. So if you're a person who has a degree or a degree of success in diagnosed and deficient stress, the set and process and improvement risk factors. So, but now that the official guideline has changed, anyone who continues to use the old racial lens to guidelines is gonna have a really, really difficult time to check it out in court because we now know that a race needs to go off. Is that a long line of Supreme Court doctrine? What would be the state versus the world discrimination when it's based on race or other categories that receive special scrutiny? So what are you guys typically saying? What are you guys whispering to me? What's it like you said? The big difference between what they call these cases for the real time, I don't think that's. What it says, to put it in the first place, which is the use of, if you take the first place, you have a class, you can take something that you really need. But what I don't like the best is that, you know, my data sharing and I think that's is a good way to get from the less developed, and have some sort of community. And even when it's used as justification, say for workplace discrimination, some economists defend this economically efficient, right, like why not use all of the information that you have available. It's actually been a couple of economists who have died into the debate about clinical practice guidelines and said, you know, essentially, just like everywhere else, why not use all the information that we have available. That's the idea that we want every predictor, that some predictors are, like even if they weren't associated with some sort of data that supported a generalization, that they're just too problematic to commit basically, that it promotes disparity and needs to be associated with, you know, like a toxic history and therefore just shouldn't be used. And so there's a long line of cases holding that such generalizations, general, that statistical generalizations, although it's trying to take that out to statistical generalizations, to take that away. The key doctrine on this was actually developed in the sex discrimination context in a series of cases. Mostly in the 1970s, a lot of them involved benefit administration, I mean, this is, the government has factored certain things, processes, that things known as something that men were likely to need or it said, you still get structure of the policies that it doesn't accurately describe. In a case in the 1990s, a lot of people were offered to do that. And then in the 1980s, all kinds of behavioral advocacy and physical health, all of these physical standards and training cycles and so forth, and you can ask centers in general, and then you can ask them to get the work that they have to do and talk to them. And the court, it's a novel, it's a cycle and it's true. But I believe that in real life, there's a huge commitment to do what is best for themselves to do. So that is true. I really am trying to, I think the point that, There was a law that basically sent different drinking ages for men and women. And the state submitted what was from them. And that part was like, not all of them. It can be that kind of notation. So most of the cases involved gender because it used to be the first period of time in which there was still a prominent domestic gender decision in the streaming law. And that was the period in which this law was developed. But in the cases in which it's come up, the focus we did, the focus here, is to be stronger for this than for gender and sex. And before that, I said that for sex streaming, which means the point that the development of the system in general, which is particularly for the basis of pregnancy or breastfeeding, basically, physical sex differences often associated with the delivery of the process. And maybe I said, but not in my back-insist case. But basically, the center system, I thought, would not be well-allowed that kind of generalization in the race context. In a pretty recent case, they threw out like these babies. That's what I believe is the case. Because the expert in this sense, in this sense, I think about physical abuse, in effect, this is the case we need to get the case out. And they didn't care whether they're about the merits of the statistical evidence. They were just like, no, not going to try to justify the difference on the basis of that kind of generalization. So why is all this relevant? Well, what did you want to move on with? And I'm sure, in the literature, you can find examples of examples in which there really does appear to be some group-based difference that a racial investment is really necessary to have. And there are some cases that say that this is too often to happen with other people. There's more of a theory of the language. Even people who are the strongest defenders of racialized CPGs will concede that point, right? And there are some things that we can take for granted. And the implication, even though the court has not, I should say this court has not decided a case like this in the context of clinical practice, that's right. So I'm extrapolating from the way that it's treated in the case of Eli to justify it. I'm giving you your whole, there are these reasons, but I'm saying that in the literature, you can see that there's some impact plot. There's some difference in average values that are observed. OK, so does all this mean that, you know, that this means that we can do it? You're just going to say, yeah, I don't think this means that. Yeah, all of that in the practice of policy making, for example, needs to ignore race. The Supreme Court doctrine certainly favors a color-blindness norm when it comes to individual level racial class. I think there's something in the case of race is more policy level, right? So it's not about treating individuals differently, like Professor Gunn-Tay, or individual race, but rather making broad health policy decisions, including in the design of other factors, whether it's the lack of awareness, that racial awareness, and the awareness of the experience, and the goal of trying to reduce all of these types of practices, not to the point that you're right, courts do strike down kind of patiently useful to get individuals some benefits. And it is based on what can be used for increasing this experience, meaning same people of color from white spaces, for example. And the companies are motivated by what is called the minority, which is the gas that you've been abraded against before. Even though, of course, there is a minority like in the COVID-19 pandemic, there's millions happening in the country, and the majority of them are in the U.S. and the majority of them are in the U.S. and the majority of them are in the U.S. and the majority of them are in the U.S. and the majority of them are in the U.S. and the majority of them are in the U.S. and the majority of them are in the U.S. and the majority of them are in the U.S. or center, in the U.S. as a lot of the people are. So, I'm trying to understand if, let's say, a lot of these organizations are trying to try to And their practical practice time lives come out and say that the black patients have an eye risk for referendum When those libraries are sent down, there's always, when everyone practice that language, there's always this screamer, that this is not going to find a standard there, but rather a general reputation of practitioner. And so it seems to me that if a patient said, I got radioactive iodine because of the race-like bad lines, I'm gonna sue someone based on human presence. If they go to sue an organization that put out the guidelines, they would say, we don't practice medicine. He was put out a guideline for us. If they sue the practitioner, practitioners would say, I didn't make my decision based on the guideline. I knew it was, I made it in a way to like sit, just like the guidelines told me to just pass it. Right. Okay. So I'm just curious, and I'm not arguing from an actual point of view, it would be inappropriate to utilize the time, just learn from a really good perspective. How do these things work out? Yeah. So I think that's a couple of points here. One of them is, oh, that's a good point. So I didn't get to my last couple slides. Well, one of which was about this new proposed, the period has been closed on it. And I don't know whether it's ever going to go into law, but it does have like some of the kidney guideline on the NFL case. Somebody is asking, what is the response when some of that's more than 50? And it says several times that we're going to be able to. And so basically it's saying I think they're talking to more than providers that you can use the price overly to go by. It's confusing. But I think they're sort of warning that, the terms are like it's provable that when you see the requirements, you can use it yourself. This was the facilities. So the ACA, which is what this regulation is in terms of the money. I mean, I think there's a lot of institutions that receive federal funds. And so there, so under Title VI, the plaintiff would see the hospital on that. And so there might be questions in particular cases about like, who is responsible, right? The upside of this regulation is kind of worrying about the response both in terms of how it could happen. Yeah, Jeremy, I think that's not the question of should I just kind of like die or is it just not really like that? Yeah, so I think the ACA, and since I think institutions and individual doctors have responsibilities, and so I think it's possible. I mean, this might just involve like other legal doctors who are employer-oriented, policy-oriented, and you get to, you might need to go to OVN, but let's say some money, some money, some money, some money is not going to happen. Yeah, so I think that's interesting because there haven't been a lot of lawsuits so far. That's the other side. And so these things are servicing in courts. I have yet to see a lawsuit that's just a plain lawsuit against a provider by a patient. You, as software and quality, okay, any other questions? All right. Thank you. Okay. Any other questions? All right. Thank you. Should I end this soon? Sorry. I will.