The Government Should Stop Mandating the Use of "Race" in Medical and Scientific Studies

The FDA and the NIH each require as a condition of funding that scientists break down their data by a totally unscientific "race" classifications.


I was asked to contribute to a symposium on race, racism, and administrative law at Notice and Comment, the blog of the Yale Journal of Regulation. Given my current research on the American law of race, some of which I have blogged here, I had a lot to choose from; the most egregious forms of racial classification in the U.S. are largely the product of administrative decisions, rather than legislation. I ultimately decided to write about how the FDA and NIH require the use of ridiculously unscientific "racial" categories, adopted by the OMB for entirely different purposes in 1977, in biomedical research, and why this use of "race" should be abolished. Here's a taste, but you can read the whole thing at this link:

As of this writing, the federal government is considering using race and ethnicity to allocate access to a new Coronavirus vaccine to combat Covid-19 when one becomes available. More specifically, the government is considering giving preference to African Americans and Latinos because they have been disproportionately affected by the pandemic.

There are obvious dangers to allocating medical resources by race … especially in a politically sensitive an area such as vaccines, where the public is already all-too-prone to accept various conspiracy theories and quackery that leads them to oppose vaccination. Instead of expanding the use of race in this way, science and medicine should be moving away from considering race and ethnicity at all.

Unfortunately, the FDA and NIH have mandated the use of race and ethnicity since the late 1990s. As a result of this mandate, the use of race has become so common in the scientific and medical communities that most people in the field fail to consider whether there is any justification for doing so. As one scientist reports, "we don't tend … to think a lot about that [race] variable, what it means, how it's defined, how it's being used. We just sort of use it blindly."

This is very unfortunate, because, in addition to other problems discussed below, the FDA and NIH mandated that the "race variable' be based on the arbitrary (but now standard in American life) racial and ethnic classifications established by the Office of Management and Budget in 1977 for civil rights enforcement purposes. At the time, the OMB warned that the "classifications should not be interpreted as being scientific or anthropological in nature." This did not stop the FDA and NIH from institutionalizing them into medical and scientific research….

Any discussion of race in science and medicine must start with the recognition that variations in DNA that may have scientific or medical implications are not specific to race, as such, but to geographical distance between different populations. Additionally, there is no known example of polymorphism that is found exclusively in any particular "racial" group….

Even if at one time race may have been useful as a crude proxy for genetic heterogeneity, as DNA testing has become more available and much less expensive, race is a poor substitute for looking at actual discernible genetic differences between people. "Pooling people in race silos," an editorial in Nature Biotechnology declared, "is akin to zoologists grouping racoons, tigers, and okapis on the basis that they are all stripey."…

The OMB category of Asian, meanwhile, is absurdly non-specific and unscientific. It includes people with origins everywhere from the Philippines to the Indian subcontinent. There are vast differences among the various ethnic groups that comprise the two billion or so people who live within the Indian subcontinent, much less between South Asians and East Asians.

Hispanic/Latino is an even more problematic category. Latinos' origins can be any combination of African, Asian, European, and Indigenous. Nor are they culturally homogenous. There is no reason to believe that data about Dominican residents of New York City is applicable to indigenous Mexican farm workers in California….

[S]upport for the idea that we should allow the government to use research based on arbitrary, scientifically ridiculous OMB racial categories to allocate medical resources to people based on those categories seems both fantastical and an unjustified triumph of unscientific racialist thinking. Unfortunately, this is what NIH's and FDA's imposition of the OMB categories into scientific research has wrought.

NEXT: The View from Inside Portland's Federal Courthouse

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. 1) I disagree with allocating vaccines based on race. It would make more sense to allocate based on risk/jobs. This would have the intended effect without discriminating against non-blacks/Latinos.

    2) No, the use of race should not be abolished. It sounds like the categories need to be updated, but they shouldn’t be abolished. There’s medicines that have been found to work better for blacks than whites. To ignore the genetics factor in medicine is a big mistake. So don’t scrap the category, update it instead. Maybe it’ll only be useful in 10% of studies, maybe more, maybe less. Any way you look at it, in medicine, more data is on the population being studied is better when trying to determine the effectivity.

    1. “black” and “white” are really lousy and leaky classifications of “race”.
      Relying on them for ANYTHING is dangerous, especially medical.

    2. The flip side of this, of course, is in any attempt to make the vaccine mandatory. In prioritizing Black/Hispanic people to get it, is not there a related acceptance that they more need it? And hence a civil rights violation would meet the strict scrutiny standard?

      Worse, imagine a Black/Hispanic Only quarantine. That’s entirely rational once you establish that they are at higher chance of having it….

      Oh Brave New World….

    3. 2/ so use actual genetics, rather than ‘race’, which is hardly a genetic classification.

  2. Exactly…There are certain drugs and diseases that have a strong genetic component, and certain races have stronger or not response to some drugs… This is important medically…

    As far as vaccines, it should be based on threat avoidance…high risk individuals get it first, others as needed.

    1. Scientifically, there is no such thing as “certain races,” so “certain races” cannot have a stronger or not response to drugs. People with certain genetic makeups may have stronger or not response to drugs, but the correlation with what is commonly known as race and such makeups is so weak as to be effectively useless. There is, for example, a lot of genetic heterogeneity among Africans. Ethiopians are closer genetically to Jews and Saudi Arabians than to sub-Saharan Africans. Africa has the tallest and shortest ethnic groups in the world, a product of genetics, who live relatively near each other. It gets even more complicated in the U.S. given that the average European admixture in the “African American” population is somewhere between 7 and 24%. A drug that works better on African Americans may be picking up something small that relates to African-descended people in general, or it make be picking up a large effect on a small ethnic subgroup. There is no way to know, maybe any claim, for example, that a drug works “better” on African Americans speculative. And, as noted, the categories of Asian and Hispanic have even less scientific salience (white, too).

      1. It may be helpful if you drop the meaningless term “scientifically” and instead narrow the claim to the life sciences, or more specifically biology, since that appears to be the framework you are working within. Perhaps something to the effect of, “researchers working in the field of genetics have persuasively demonstrated that, from a biological perspective, the concept of race does not capture a meaningful way to differentiate discrete groups of people” (citation). I just made that sentence up, but I’m sure you can find a journal article saying something close that and just plug in the citation. Consider that one a freebie.

        Something along that line would give you a more scholarly voice, as opposed to just saying “scientifically there is no such thing as….”

        1. I don’t think there’s a problem with his phrasing except in the most technical sense. In the most technical sense, trying to defend a ‘scientific basis of race’ is “not even wrong” – the assumptions are so bad that the truth or falsity can’t even be evaluated, and instead it’s meaningless gibberish. Summarizing that as “scientifically there is no such thing as race” is a perfectly acceptable gloss.

    2. You and BillyG need to re-read the article. Bernstein is not saying that there is no genetic component to diseases or treatments. Bernstein is saying that broad definitions of “race” that are in use are utterly inadequate to that purpose.

      He implies (and I’m inclined to agree) that if you did redefine race to something that was medically relevant, it would be so different that calling it “race” would no longer make sense.

      1. Good summary, thanks.

  3. DNA testing has become more available and much less expensive

    Yes, the cost (and speed) of mapping someone’s genome has dropped by a crazy degree since the turn of the century. But inquiring minds want to know: are we going to map everybody’s genome before giving them the Coronavirus vaccine? If not, isn’t a – possibly improved – crude mapping based on social understandings of race better than nothing?

    1. Well, if they can identify the relative genetic variations, rather than having to sequence somebody’s entire genome, you could get by with just looking for specific markers. At that level of detail, genetic testing is pretty darn cheap.

      So I agree with the proposal in principle, though I’m not sure we’re there yet technically.

    2. No, it’s not.

    3. OK, smart guy — what race is Obama?

      1. Obama had just as much claim on being the 44th white President as the first black one.

        1. everyone knows that the first black president was Bill Clinton

        2. Yes. Now which race will Martinned assign to him as a vital medical property?

    4. I think you could do the relevant DNA testing. But if for some reason it couldn’t be done, asking about *specific ancestry* would be a far better proxy than ‘race’.

      (ie, “My ancestors were predominantly Swedish and German” instead of “White” conveys a lot more useful information).

      1. Begin with blood typing. It’s cheap, results are available instantly, and it is known to correlate with resistance to some diseases. Although most racial or ethnic groups include all the common blood types, the ratios between these types vary greatly with ethnic origin, enough that blood type could account for most of the ethnic differences in susceptibility to various diseases.

  4. In her analysis of 204 biomedical research articles that used race or ethnicity as variables, Catherine Lee (2009) writes, “[D]espite seeing the importance of race or ethnicity in their research, authors rarely defined or operationalized the concepts adequately” (1184). Apparently, law professor pundits who comment on the subject are no different, as Bernstein has not even attempted to operationalize these key terms in his argument. Each individual reader is therefore left to fill-in the gap using their own understanding of the terms. One can only wonder how that will play out on a political blog.

    Lee, Catherine. 2009. “”Race” and “Ethnicity” in Biomedical Research: How Do Scientists Construct and Explain Differences in Health?.” Social Science & Medicine 68: 1183-90.

    1. Did you actually read the article linked? I didn’t define the “racial” terms because researchers are required to used the OMB definitions.

  5. As age seems to be the greatest factor, perhaps we should sort by age.

    1. Age correlates with the biggest consequences (death) but for those catching it I think the biggest correlation is (lower) income. I’m not sure which to go by first. Provide it to the over ~60ish first or start at the bottom of the income ladder and work up or some balance of the two?

      1. Wait, why would you start with the old people? Surely you’d want to do them last, since the vaccine will add the fewest healthy life-years there?

        1. As far as cost benefit analysis is concerned, absolutely.

          Then again, humanity is past due for a culling, so why bother. Where’s Thanos when you need him?

        2. You’d do the elderly first because the over whelming majority of fatalities are elderly. A three year old probably doesn’t need the vaccine at all, except to prevent the spread. And since the likelihood of dying from the disease drops off so precipitously from 75+ to 65-74, and again for 55-64, you’d start at the top and move down, to maximize the amount of healthy life-years saved. As an example, in California a majority of the deaths were people over the age of 74, even though they made up only ~5% of total reported cases. The CFR for people under the age of 50 is like .01%. (The IFR is lower). If there’s a shortage of vaccines, you really wouldn’t want to give any of them to people below the age of, say, 35. Probably not below 50.

          1. Except that death is not the only concern.

      2. Income correlates with lifestyle, including things like handwashing.

    2. Yes, but which age group achieves the goal. Depends on whether kids are a significant vector (probably not), in which case vaccinating the vector could be most efficacious, or not.

  6. “As of this writing, the federal government is considering using race and ethnicity to allocate access to a new Coronavirus vaccine to combat Covid-19 when one becomes available…..There are obvious dangers to allocating medical resources by race …”

    I was similarly surprised when reading in the NYT a couple weeks ago the social justice handwringing about whether it would be right to allocate vaccine to blacks and hispanics first. This seems the perfect plan to implement if one wanted to 1) alienate blacks and hispanics, populations with historical distrust of the medical and government establishment, who could be made to feel that they are guinea pigs in the vaccine experiment, and 2) alienate whites who might see the SJW class moving others to the front of the line.
    In other words, allocation based on race might be just the ticket to ensure Trump wins.

    1. You’ve got me thinking about an interesting study. Take all the SJW proposals which are based on race. Replace race with “people with lower income” or something similar. Observe how that factors into how favorably people view the proposals.

    2. Putting aside the rest of my post, of course allocated vaccines by race is the absolute worst of all worlds. Many members of minority groups will believe they are being “experimented upon,” and will refuse vaccination, and non-minorities will be angry that others are able to get the vaccine before them based on “race.”

      1. If the allocation is simply access to the vaccine, you’d only have volunteer guinea pigs. Maybe the effect is the same. We’re not talking about forced vaccinations, are we?

        I happen to think it’s a mistake to hand it out based on race, too. We know enough about complicating factors to allocate based on need.

        1. “Voluntary” is a funny way to describe any mandatory government program, and thinking that a covid vaccine will not be mandatory is a funny reading of history.

          1. If we’re positing a world in which there are not enough vaccines to cover the entire population, why wouldn’t it begin voluntarily?

            1. You mean why wouldn’t the rationing begin voluntarily? And if ungrateful wretches turn down the gift from GodGovernment Almighty, they’ll just be forced to take it.

            2. “why wouldn’t it begin voluntarily?” Because government is run by power-mad people who’d rather make everything that is not forbidden mandatory. For instance, in WWII, Pearl Harbor inspired more men to volunteer than the armed forces could train or equip in the first year – but since a draft system existed, they ran all recruiting through it.

  7. Race (genetics) can have a big effect on drugs and vaccines. Now the left probably finds that inherently dangerous as they have pushed race as a social construct for the last century. But, any scientist will tell you that genetic markers in various races can mean dramatically different results with certain treatments.

    1. Jimmy, just curious, do you actually understand the science behind the claim that race is a social construct?

      Nobody disputes that there are differences in the amount of skin melanin. The problem is that the term “race” is ill defined, and therefore of limited usefulness at best, and only rarely tells us anything useful. For example, people from India are Aryans and therefore more closely related to white Europeans than they are to black Africans, even though they have dark skin. Further, there has been enough inter-mixing over the millenia — for example, anyone whose ancestors were living in Europe in the 13th century almost certainly has at least some Mongol blood — that there are very few “purebreds” among us anyway.

      In this case, the issue really is genetics rather than race, and unlike race, genetics is something that is well defined and does provide useful information.

      1. “Jimmy, just curious, do you actually understand the science behind the claim that race is a social construct?”

        There is no science behind it. It’s social “science”, hence “social construct”. It has absolutely 0 to do with science, like many of the leftist’s pet projects.

        1. Most biologists and anthropologists would disagree with you, but what do they know?

          1. ” The problem is that the term “race” is ill defined…

            In this case, the issue really is genetics rather than race, and unlike race, genetics is something that is well defined and does provide useful information.”

            You’re confusing your concluding statement with implying that “race as a social construct” is scientific. “Race” is not scientific. It’s anti-science. And “social sciences” aren’t science, they’re religion.

            Genetics are absolutely useful and scientific. Race is not, because race is make-believe.

            But, sure, “they” would disagree with me. Terrible attempt at an appeal to authority.

      2. I read population geneticists and I haven’t seen any of them make a claim like that about Europeans and Genghis Khan. We can detect a genetic signature of Genghis and his descendants, but it’s concentrated in Asia.

        Population geneticists have discovered that Europeans are a hybrid of populations as diverged as modern Europeans compared to modern Chinese, but that’s a hybrid of populations that go back a LOT further than Genghis.

    2. You too — tell me, what race is Obama, medically speaking? Forget the damned social construct — this is a vaccine which doesn’t give a flying rat’s ass about yru social construct.

  8. We don’t have the infrastructure for individually genetically bespoke treatments yet.

    Given that, what other heuristic do you suggest?

    1. In the absence of a known genetic marker that can be utilized, I’ll defer to the “stripey” example. I quote a geneticist pointing out somewhere else is that the only reason “race” seems salient, is that the relevant studies assume such a thing exists, and then look for medical correlations. If they didn’t start out with that background (and scientifically unjustified) assumption, it would never occur to them to think that all people who look, say, vaguely or more like they have African descent should be put in the same group for research purposes.

      1. But don’t we need some heuristic for genetic similarity to use until genetic tests are not yet de rigour for a standard workup?

        I am not doing some Socratic thing – I genuinely don’t have an expectation as to the best policy answer here.
        Assuming race is indeed a bad heuristic, what’s other good quick ones to use…BMI, maybe…not coming up with much else.

        1. I have plotted covid-19 mortality vs. BMI for about 3 dozen countries. There is no significant correlation except that countries with low average BMI have significantly (10 or more years) younger populations and that bunch falls into a separate population without age or BMI correleattion

          1. I wasn’t speaking just about COVID, but about medical research generally.

            But that’s good to know.

        2. Male and female are pretty good places to start, albeit currently under attack. A medical system that can’t ask whether someone is biologically male or female, but can make decisions based on a scientifically ridiculous category like Hispanic, is verging on quackery.

          1. I don’t think they’re under attack for medical research purposes.
            I assumed gender went without saying.

            I wouldn’t call it quackery, since there are clinically provable beneficial results in treatment protocols. But yeah, spending some resources interrogating whether this is ‘stripey’ confirmation bias seems completely fair.

            1. It’s not impossible for granular data on ancestry to be a useful, albeit very imperfect, proxy for genetics. You just have to be very careful with it.

              1. geographical ancestry that is, not “race” ancestry.

              2. It’s not impossible, but I concur that past protocols were almost certainly insufficient to determine how useful and best practices, due to assumptions about the exact correlation between genetics and race.

                Not quackery (indeed, quackery is more about medical practice than research), but the usual prejudices we strive to get rid of sneaking in to research.

    2. I’d suggest getting as detailed of an ancestry as possible (ie, to ancestral region/population), but that relies on testing being tracked on the basis of ancestry too. That increased granularity, however, would tend towards preserving some real signal, unlike the agglomeration of quite disparate groups under OMB ‘race’ definitions.

      (Fun fact, people of sub-saharan african ancestry represent more genetic variation than the rest of the human population on the globe, mostly because sub-saharan africa encompasses groups who are more disparate than elsewhere in the world. If you classify them all as ‘black’, that means some ‘black’ people are genetically more different from some other ‘black’ people than white Europeans are from Han Chinese).

      1. Note: ancestry is useful, but not sufficient. See:

        (In short, some individuals will be more similar to other populations than their own, especially when considering a small number of genes or phenotypes, which would always be the case when guessing at specific drug suitability).

        1. The point of the OP is that these studies may be in doubt since it’s possible using any categorization to tailor treatment would have a beneficial result. (And the control is, of course, no categorization at all).

          1. I don’t think human genetic phylogenies are in doubt, or other human genetic comparison studies, since those aren’t done on the basis of “race”, especially not as described by OMB. This study ultimately supports his point (that ‘race’ isn’t a very good categorization), but it doesn’t do anything with race.

            This study works with ‘populations’, which are definitely a biologically real thing.

            1. No, but genetic tests are not ubiquitous enough for genetically bespoke treatment options to be a thing yet.

              As such, we need to use heuristics for now. Race is a seemingly obvious one, but it’s never really been interrogated scientifically. And indeed there are reasons to believe it’s a pretty crap categorization to use.

  9. This seems to miss something rather large. There were social and legal racial classifications that led to very different realities for those so assigned. The categories lasted longer than many people’s lifetimes meaning that individuals categorized saw themselves and their children treated differently for generations. Racial ‘science’ was one often relied on justification for the legal and social categories.

    One consequence of these categorizations is different experiences for groups differently categorized. Even after legal categorization both social effects and lingering effects of the legal categorization of people (and ancestors) in the past create different levels of resources in dealing with, experience with and trust in scientific and medical institutions. These themselves then can have negative impacts for impacted groups.

    Therefore, ensuring that people from the past categorizations are being included in, reached out to, and ‘looked out for’ seems to me not only justified even if all of good will now agree the categorizations were never scientifically justified and should not have been the basis of legal and social disadvantage, but perhaps morally mandated. The purpose is simply to recognize that while the categorizations should never have been instituted, they were, to people’s detriment, and we want to look out for the people that fell under the categorization going forward.

  10. Failure to take account of race is medical malpractice. Stick to the law, ok?

    Sally Satel, M.D., “I Am a Racially Profiling Doctor”, New York Times, May 3, 2002.

    1. The idea that racially-specific drug reactions could be confirmation bias is not one to be discarded lightly.

    2. Satel is 100% wrong. Among others who say so is Sally Satel, who a few years earlier wrote an article critical of the notion that the FDA and NIH should be considering race in research. The only thing her two articles have in common is a reflexive opposition to the politically correct impulse of the moment. (And for that matter, Satel really has no more expertise on this than anyone else who reads the relevant literature. Being an “MD” doesn’t mean you know a damn thing about race and genetics.)

      And, btw, since it’s the law that requires researchers to use “racial” categories defined by law, this is law.

      1. Here is the earlier Satel article: Sally L. Satel, Can Science Survive Research by Quota, USA Today Magazine, Sept. 1994, at 76.

        1. In the earlier piece, she criticized the growing movement to require researchers to recruit minority (and female) subjects for their studies. But everyone acknowledges that minority subjects were well “underrepresented” compared to their % of the population in studies at the time, to the extent it would have been largely impossible to find “racial” effects in studies if they existed. So if Satel believed that race is an important, relevant factor in medicine, as she purported to in 2002, then why in 1994 was she arguing against rules that would make sure racial effects would be measurable?

          1. Sickle cell.
            Lactose intolerance.

            You can’t be serious.

            1. These are all genetically inherited conditions. None of them are exclusive to any race or ethnic group. There are certain ethnic/national groups, Ashkenazic Jews and Icelanders among them, that are so interrelated that they will have much higher rates of certain diseases inherited from the groups’ “founders” than others. But being an Ashkenazic Jew or Icelander is not a “race” as race is used in medicine.

              As for sickle cell in particular, the fact that you don’t know that it’s not a “racial” disease calls into question your competence on the general subject. “It is common in Africa, in Mediterranean countries (such as Greece, Turkey, and Italy), the Arabian Peninsula, India, Spanish-speaking regions in South and Central America, and parts of the Caribbean. In all those regions, both dark and light skin people can carry copies of the sickle cell genes.”

              1. DON’T ask race.
                You might as well don’t ask if the patient smokes, or drinks (because not everyone who smokes is likely to have heart disease and not everyone who drinks too much might have fatty liver).
                Go ahead, Dr Bernstein.
                Might be hard to find malpractice insurance.

          2. Perhaps in the intervening years she learned some new information.

      2. On the strictly legal aspect of this question, we can grant you your expertise. But to the same extent we can grant Satel her expertise on the medical aspect of the issue.

  11. As to the argument that there can be considerable variation in subcategories of racial categories therefore the use of the latter are worthless (?)…If a study found that there were a X% increase (or decrease) in hate crimes against ‘Jews,’ would it be a useless statistic because, after all, we don’t know if, perhaps, the entire increase were only for certain sub-categories of Jews (are we talking Ethiopian, Ashkenazi, Sephardic, etc? Reform, Conservative, Orthodox, etc?). I can’t think of many sociological categories that can’t be subdivided down further (and maybe further still). I mean, if read that ‘Republicans’ are more likely to give to charity, is this a worthless statistic because one could drill down into endless sub-categories (Libertarians, which can further be divided down into so many types, Social Conservatives, who might be Mid-western conservative Catholics, Floridian Southern Baptists, etc., etc.,). Does this mean, in a kind of Xeno’s paradox style argument, these commonly and long held variables and their operationalizations are worthless?

    1. We don’t use, and certainly aren’t required to use, any of those categories in medicine. No doctor has ever asked me whether I belong to any of those categories, except Ashkenazi Jewish for the purpose of screening for genetic diseases when my wife was pregnant.

  12. If I had a nickel everytime someone told me race doesn’t exist because no official body set a firm definition for it….

    Races ie groups of genetically interrelated peoples don’t stop existing because people prefer they don’t and they don’t have a rigid definition anymore than lakes stop existing because they don’t have a rigid delineation from a pond.

    I could sympathize with claims that we need a better classification system but really, nobody has come up with a better alternative. In fact the Left which which is behind these efforts to get rid of the concept of race references race in its rhetoric more than anyone else.

    1. Given your take on “the left”, you should know, as you would if you actually had read the link, that it’s the left that is currently most insistent that “race” be used in medical research.

  13. David Bernstein is not a scientist or a doctor. Doctors currently use race, even crude self-reports, because the alternative results in worse predictions of patient health outcomes in a number of areas. It would be even better if doctors had genomes for every individual, but we aren’t there quite yet and can’t expect to be in the near term. Most medical research is currently done on the majority population, which is to say white. Not taking into account medical differences across populations and assuming data derived from mostly white samples is just as valid for others is just going to lead to worse outcomes for minorities. The government’s classifications are flawed and not what a scientist would come up with, but throwing away data is worse. This isn’t sufficient to show that race/ethnicity should be mandatory, since many things which are generally good ideas don’t need to be required, but Bernstein’s argument, though popular, is wrong.

    Racoons, tigers & okapis actually are more related to each other than they are to most other species: they are all mammals. We can surmise from their DNA how far back they all diverged. Self-reported race matches quite well to cluster analyses of DNA, because it’s indicative of recent (compared to the rest of humanity) shared ancestry.

    1. I encourage you and everyone else following this to read, Javier Perez-Rodriguez & Alejandro de la Fuente, Now is the Time for a Postracial Medicine: Biomedical Research, The National Institutes of Health, and the Perpetuation of Scientific Racism, 17 Am. J. Bioethics 36, 41 (2017).
      Here is just one problem with doctors “using” race. Doctors rely on self-identification. One study showed that one-third of a large national sample of interviewees reported a different ethnicity or race a year after the initial interview. Americans’ self-identified race on census forms often varies from decade to decade. So if a patient comes in and tells the doctor I’m African American, or white, or whatever, you have no idea what they means, if anything. And people who identify as African Americans, or Native Americans, can be anywhere from 100% to virtually zero of African or indigenous origin. How could that possibly be helpful in an individual case, even if you assumed that racial background was salient?

  14. There is no “medical” aspect. There is a scientific aspect. Doctors aren’t geneticists, biologists, or otherwise “experts” in the relevant matters. Doctors can pick up whatever they get trained in (i.e. conventional but unscientific medical wisdom, which is often wrong, see episiotomies) and, like anyone else, they can read the relevant studies. Satel, in short, had no more inherent expertise in the underlying scientific issues than I do. Actual scientists, particularly geneticists, seem on average much less keen on using race in medicine than practicing doctors (who again are far from being experts in genetics) do, and much more aware of the many ways race is misused by doctors like Satel. In fact, if you read up on Satel’s article, many scientists criticized it harshly, and some of the examples she gave were, to say the least, poorly chosen.

  15. Professor Bernstein,

    What do you make of the FDA’s approval of BiDil, the heart failure medication for self-identified African Americans?

    The Lancet summarized the criticism, which I understand you share, as follows: “If race were used as a proxy for ancestral African genomics, it should be a continuous function (10%, 30%, 70%, etc.) It makes no scientific sense to map a continuous function onto a dichomotous variable.”

    But is this criticism really valid? As Justice Holmes observed long ago, night and day have no sharp dividing line between them, only
    shades of gray, and yet they are highly useful concepts. Society constructs categories out of continua based on somewhat vague and uncertain criteria all the time.

    Moreover, it’s a natural human tendency to exaggerate the vagueness in classifications we don’t like in other grounds, but find it perfectly acceptable in categories we otherwise feel more comfortable with. Grades, inspections of all kinds, court judgments, all manner of human judgments take complex and broad subjects and squeeze them into narrow categories – pass/fail, guilty/not guilty, hire/not hire, etc. – by imperfect criteria, for purposes of making a decision.

    Why is race, and its vagueness and imperfection in classifying complex genomic information into workable categories distinguishable by an ordinary observer without sophisticated expertise or complex tools, any different?

    If race has no meaning at all, then why does BiDil work in studies of self-identified African Americans, and not in studies of general heart failure patients?

Please to post comments