Ted talk discussion (The problem with race based medicine)

[WingedOx]

---

Members do not see this ad.

That's literally the only post in this thread that mentioned admissions in any form. The other comments and the talk itself were referring to race in medical practice.

Ugh I meant to write "you wouldn't have clicked if the title didn't talk about admissions." Sorry, I've been on call all weekend.

 

[ridethecliche]

Her talk isn't about medical school admissions at all, it's about doctors using race as a factor in diagnosing and treating patients and whether or not this practice is useful and beneficial.

It appears that her reasons for stating that this practice is not useful is because the statistics don't apply to every single patient of a race/identity and because the underlying principles aren't understood.

Doesn't that invalidate the entire field of public health and statistics as well?

/end tongue in cheek.

 

[PB&Jam]

Ugh I meant to write "you wouldn't have clicked if the title didn't talk about admissions." Sorry, I've been on call all weekend.

The title also doesn't talk about admissions...
Sorry you've been on call all weekend, that sucks.

 

[PB&Jam]

It appears that her reasons for stating that this practice is not useful is because the statistics don't apply to every single patient of a race/identity and because the underlying principles aren't understood.

Doesn't that invalidate the entire field of public health and statistics as well?

/end tongue in cheek.

I don't agree with her at all. I was merely pointing out that her talk was not about admissions.

 

[terra330]

Considering this exact topic has come up in my genetics, general bio, and 2 public health classes, and they have all echoed what I've stated, I'll have to continue to disagree with you. I get it, you really want to be a doctor so naturally your confirmation bias and lack of understanding on the subject is clouding your judgment.

Is it so hard to accept that race is both a social construct and correlated with genetics? No one is saying any particular race is more or less human than the other. Just that populations were geographically separated for a long time, and certain groups accumulated certain traits. Sometimes these traits can be relevant when delivering healthcare.

But what do I know, I'm apparently just race baiting because I agree that blanket statements about an entire race of people aren't the best way to approach patient care.

So what do you suggest? Ignoring race altogether would just result in blanket statements applying to the entire species. The goal in medicine is to cater treatment as best to an individual as possible. Basing treatment plans based on our genomes would be ideal, however it's not really feasible at this point in time (though that will likely rapidly change in the coming years)
1) It's still expensive, especially if you want results ASAP
2) Even though we know many conditions are hereditary, identifying which genes cause those condition is rather tough. Even for something as simple as curly/straight hair, we haven't figured out how to identify that with just a genome with 100% accuracy.

So in the absence of genetic testing, next best thing is family history. However, family history has its limits in that records are far from perfect, we only have access to a generation or two prior to our own, and sometimes due to the genetic lottery, you'll be the first in your family's recent history to exhibit some issues.

So when you can't get the information you need from genetic testing or family history, the next best option is your race. No it's not perfect, and yes, now that mixed race couples are more common than in the past, it will likely become less useful within a couple generations. Hopefully by then genetic tests will be sophisticated enough and cheap enough for that to no longer be an issue.

 

[ridethecliche]

I don't agree with her at all. I was merely pointing out that her talk was not about admissions.

I was responding to the second part of your statement not anything about admissions!

 

[benjamin94559]

you pc bro?

but seriously this talk will be used by the most pc

 

[benjamin94559]

I guess I would argue that we have data based on how people self identify so we´ve kinda sort controlled for that variable

 

[whatever5]

I'm having trouble understanding what she's upset about.

At first, she complains that she doesn't know whether to check caucasian or black in a survey, and says that this makes the study "bad science" because apparently it would be way too difficult to just check both. She then goes on to complain that certain estimates yield different levels based on race, but the difference between estimates seems so negligible that I'm not sure why it's a point of concern.

Basically, it sounds like she wants doctors to ignore people's race completely, which I think is just as dumb as the notion that we should ignore people's races in real life and be like "I just see people man, I don't see color". Being of a certain race makes you more susceptible to certain diseases, which she acknowledges, though she cites the cause as sociocultural rather than genetic - this may be true in some cases, but who cares? The way I see it, her argument is no different than saying its unethical to test an injection drug user for HIV or to screen a Jewish couple for Tay-Sachs.

 

[LuckyMD2b]

Is it so hard to accept that race is both a social construct and correlated with genetics? No one is saying any particular race is more or less human than the other. Just that populations were geographically separated for a long time, and certain groups accumulated certain traits. Sometimes these traits can be relevant when delivering healthcare.

I find it hard to understand why physicians are so wed to the use of a social construct as marker of biological activity.

Can you state the exact extent to which race is correlated to genetics?

Biologically, please explain to me why this woman should be considered in the same category as a person who is 100% of African ancestry when >50% of her heritage is European. In the United States this woman is black, but she would not be characterized as black in Brazil. A person’s race can change depending on location and time. How can such a variable be in any way reliable from a scientific standpoint?

So what do you suggest? Ignoring race altogether would just result in blanket statements applying to the entire species. The goal in medicine is to cater treatment as best to an individual as possible. Basing treatment plans based on our genomes would be ideal, however it's not really feasible at this point in time (though that will likely rapidly change in the coming years)
1) It's still expensive, especially if you want results ASAP
2) Even though we know many conditions are hereditary, identifying which genes cause those condition is rather tough. Even for something as simple as curly/straight hair, we haven't figured out how to identify that with just a genome with 100% accuracy.

So in the absence of genetic testing, next best thing is family history. However, family history has its limits in that records are far from perfect, we only have access to a generation or two prior to our own, and sometimes due to the genetic lottery, you'll be the first in your family's recent history to exhibit some issues.

So when you can't get the information you need from genetic testing or family history, the next best option is your race. No it's not perfect, and yes, now that mixed race couples are more common than in the past, it will likely become less useful within a couple generations. Hopefully by then genetic tests will be sophisticated enough and cheap enough for that to no longer be an issue.

Many of the commenters on this thread have attacked this woman and her argument without addressing any of the very valid points she raises in her talk. As an orthopedic surgeon I would like to know the biological basis for black patients needing less pain medications for long bone fractures than white patients. When we begin to use a factor to treat patients our first duty is to ensure that it will do them no harm, and this is an example of how race has a negative impact on patient care.

 

[LuckyMD2b]

I'm having trouble understanding what she's upset about.

At first, she complains that she doesn't know whether to check caucasian or black in a survey, and says that this makes the study "bad science" because apparently it would be way too difficult to just check both. She then goes on to complain that certain estimates yield different levels based on race, but the difference between estimates seems so negligible that I'm not sure why it's a point of concern.

Basically, it sounds like she wants doctors to ignore people's race completely, which I think is just as dumb as the notion that we should ignore people's races in real life and be like "I just see people man, I don't see color". Being of a certain race makes you more susceptible to certain diseases, which she acknowledges, though she cites the cause as sociocultural rather than genetic - this may be true in some cases, but who cares? The way I see it, her argument is no different than saying its unethical to test an injection drug user for HIV or to screen a Jewish couple for Tay-Sachs.

You are referring to the difference between GFR estimates seems so negligible that “I’m not sure why it’s a point of concern.” If the difference between estimates are negligible why should they be used at all?

She isn’t making an argument as to the ethics of using race, she is arguing that it isn’t scientific; which is absolutely correct.

Intravenous drug use is a direct risk factor for HIV transmission, your comparison is illogical.

You say, "...she cites the cause as sociocultural rather than genetic - this may be true in some cases, but who cares?"

It amazes me that the same people who will argue for the scientific basis for the importance of considering race are also arguing against considering the many profound social impacts of race.

 

[Cytarabine]

You are referring to the difference between GFR estimates seems so negligible that “I’m not sure why it’s a point of concern.” If the difference between estimates are negligible why should they be used at all?

She isn’t making an argument as to the ethics of using race, she is arguing that it isn’t scientific; which is absolutely correct.

Intravenous drug use is a direct risk factor for HIV transmission, your comparison is illogical.

You say, "...she cites the cause as sociocultural rather than genetic - this may be true in some cases, but who cares?"

It amazes me that the same people who will argue for the scientific basis for the importance of considering race are also arguing against considering the many profound social impacts of race.

Let's says there are 2 drugs, A and B, that treat a given condition. In an RCT, you have 2 subgroups, one a sample of the obviously heterogenous "black people in the United states" and one "white people in the United states". The study has a statistically significant finding of decent effect size that drug A works better in the black population and drug B works better in the white population. Explain to me why me preferentially prescribing drug A as a first line therapy for black patients and drug B as a first line therapy for white patients is (a) inappropriate and (b) not evidence based

 

[WingedOx]

The title also doesn't talk about admissions...
Sorry you've been on call all weekend, that sucks.

doh... double brainfart.

 

[Psai]

Considering this exact topic has come up in my genetics, general bio, and 2 public health classes, and they have all echoed what I've stated, I'll have to continue to disagree with you. I get it, you really want to be a doctor so naturally your confirmation bias and lack of understanding on the subject is clouding your judgment. But what do I know, I'm apparently just race baiting because I agree that blanket statements about an entire race of people aren't the best way to approach patient care.

Yeah im gonna keep wondering if my black and hispanic patients with chest pain are sicklers and it won't really come up for whites. Btw taking a few classes in something gives you approximately no extra legitimacy when speaking about a subject

 

[LuckyMD2b]

Let's says there are 2 drugs, A and B, that treat a given condition. In an RCT, you have 2 subgroups, one a sample of the obviously heterogenous "black people in the United states" and one "white people in the United states". The study has a statistically significant finding of decent effect size that drug A works better in the black population and drug B works better in the white population. Explain to me why me preferentially prescribing drug A as a first line therapy for black patients and drug B as a first line therapy for white patients is (a) inappropriate and (b) not evidence based

You are responding to my specific questions regarding real life situations, with a hypothetical question involving a scenario that you made up and represents an extremely uncommon situation in standard medical practice. Interesting…

In your hypothetical example, before making any policy changes with regards to first line therapy for specific populations, I would expect the following conditions to be met

1. The results of this study should be reproducible and/or consistent with the results of previous studies.

2. I would also need the authors to provide some hypothesis as to what would be the biological mechanism that they believe results in such a dramatic difference between the two populations.

3. Potential confounders should be clearly identified and addressed prior to making a final policy recommendation

At that point I would be satisfied that this example would be appropriate and evidence based.

I, however, would still argue that given the fact that there is more genetic diversity within racial groups than between groups; there would be some individuals within each population who will be better candidates for the other drug. I would strongly caution a future physician from using race as the only factor in determining treatment for an individual patient.

Now that I answered your question, maybe you could give a shot at answering why black patients should get less pain medication for the same injury as white patients?

 

[LuckyMD2b]

Yeah im gonna keep wondering if my black and hispanic patients with chest pain are sicklers and it won't really come up for whites. Btw taking a few classes in something gives you approximately no extra legitimacy when speaking about a subject

First Hispanic is an ethnicity describing a very diverse group of people, there are Hispanics who would be considered white, and those who may call themselves black, or Native American and everywhere in between.

second there is a test for sickle cell. You don't have to use race as a factor in determining if the black patient with chest pain maybe having a heart attack or sickle cell crisis. You can get the test and find out, or simply ask the patient.

 

[getdown]

You are responding to my specific questions regarding real life situations, with a hypothetical question involving a scenario that you made up and represents an extremely uncommon situation in standard medical practice. Interesting…

In your hypothetical example, before making any policy changes with regards to first line therapy for specific populations, I would expect the following conditions to be met

1. The results of this study should be reproducible and/or consistent with the results of previous studies.

2. I would also need the authors to provide some hypothesis as to what would be the biological mechanism that they believe results in such a dramatic difference between the two populations.

3. Potential cofounders should be clearly identified and addressed prior to making a final policy recommendation

At that point I would be satisfied that this example would be appropriate and evidence based.

I, however, would still argue that given the fact that there is more genetic diversity within racial groups than between groups; there would be some individuals within each population who will be better candidates for the other drug. I would strongly caution a future physician from using race as the only factor in determining treatment for an individual patient.

Now that I answered your question, maybe you could give a shot at answering why black patients should get less pain medication for the same injury as white patients?

I don't think any of us ever said using race as the only factor in determining therapy was the way to go. However, race is a proxy for genetic differences early in human evolution when people in Europe and Africa evolved different genetic predospisitions. White Europeans have a predisposition for developing melanoma and certain inherited genetic diseases. If I were an IM/FM doc and an old white person came in with spots on their skin I'd be more willing to look for melanoma then if a darker skinned person came in. African's developed sickle cell to combat malaria. Now if a black person came into an ED with chest pain that's going to be higher in the differential then if a white/asian person came in complaining of chest pain. There are other example and studies I've already posted in the previous page regarding anti=hypertensives and rate for prostate cancer in African Americans that you can look up if interested.

However, I don't think I've ever heard of this claim that black patients received less pain medication than white patients. I've been working with you orthopods for several years now and I've never seen this happen. The first time I heard this was when the lady in the video mentioned it. Could there be personal bias regarding this? Yes, probably. But if it did occur, it's not exactly medically accepted. If you have a paper or source on this post it up so we can see.

 

[LuckyMD2b]

I don't think any of us ever said using race as the only factor in determining therapy was the way to go. However, race is a proxy for genetic differences early in human evolution when people in Europe and Africa evolved different genetic predospisitions. White Europeans have a predisposition for developing melanoma and certain inherited genetic diseases. If I were an IM/FM doc and an old white person came in with spots on their skin I'd be more willing to look for melanoma then if a darker skinned person came in. African's developed sickle cell to combat malaria. Now if a black person came into an ED with chest pain that's going to be higher in the differential then if a white/asian person came in complaining of chest pain. There are other example and studies I've already posted in the previous page regarding anti=hypertensives and rate for prostate cancer in African Americans that you can look up if interested.

However, I don't think I've ever heard of this claim that black patients received less pain medication than white patients. I've been working with you orthopods for several years now and I've never seen this happen. The first time I heard this was when the lady in the video mentioned it. Could there be personal bias regarding this? Yes, probably. But if it did occur, it's not exactly medically accepted. If you have a paper or source on this post it up so we can see.

I am an orthopedic surgeon I agree it is not medically accepted to give patient's less medication based on race. I'm sure this disparity varies depending on institution and social make up of the populations served.

But there are studies that shows that it happens.

American Journal of Public Health, Dec 2003 2067-73 "Racial and Ethnic Disparities in Emergency Department Analgesic Prescription"

Clinical Orthopaedics and Related Research, Jul 2011, 1859-70 "Defining Racial and Ethnic Disparities in Pain Management"
"Racial/ethnic minorities consistently receive less adequate treatment for acute and chronic pain than non-Hispanic whites, even after controlling for age, gender, and pain intensity."

You argue that if a black person comes in with chest pain, sickle cell should be on the differential. Ok, but sickle cell disease affects only 1 in 500 African-Americans. The chest pain is much more likely to be coming for another cause. Would you delay potentially life saving interventions for the <1% chance that this is sickle cell. Also you don't need to use race as a marker for potentially having sickle cell there is a test! I strongly believe most people over the age of 12 with sickle cell are aware of their diagnosis, and you just have to ask them. You don't need to assume that the black person with chest pain is probably a sickler.

Melanoma is more common in Europeans agreed. But black people can still get it as well.

I am not arguing that there aren't differences between racial groups when it comes to certain diseases. That is a fact. My issue and I think the purpose of the TED talk is to say there are better ways of identifying a patient's risk for certain illnesses, and if we just assume that race best because it is the most convenient then we may be doing our patients a disservice.

 

[medic86]

I am an orthopedic surgeon I agree it is not medically accepted to give patient's less medication based on race. I'm sure this disparity varies depending on institution and social make up of the populations served.

But there are studies that shows that it happens.

American Journal of Public Health, Dec 2003 2067-73 "Racial and Ethnic Disparities in Emergency Department Analgesic Prescription"

Clinical Orthopaedics and Related Research, Jul 2011, 1859-70 "Defining Racial and Ethnic Disparities in Pain Management"
"Racial/ethnic minorities consistently receive less adequate treatment for acute and chronic pain than non-Hispanic whites, even after controlling for age, gender, and pain intensity."

You argue that if a black person comes in with chest pain, sickle cell should be on the differential. Ok, but sickle cell disease affects only 1 in 500 African-Americans. The chest pain is much more likely to be coming for another cause. Would you delay potentially life saving interventions for the <1% chance that this is sickle cell. Also you don't need to use race as a marker for potentially having sickle cell there is a test! I strongly believe most people over the age of 12 with sickle cell are aware of their diagnosis, and you just have to ask them. You don't need to assume that the black person with chest pain is probably a sickler.

Melanoma is more common in Europeans agreed. But black people can still get it as well.

I am not arguing that there aren't differences between racial groups when it comes to certain diseases. That is a fact. My issue and I think the purpose of the TED talk is to say there are better ways of identifying a patient's risk for certain illnesses, and if we just assume that race best because it is the most convenient then we may be doing our patients a disservice.

Do you seriously think an EM doc will delay lifesaving tests/treatments (like getting an ekg) by just considering something on the differential?

Why does every poster on that particular side of the argument indirectly contend that doctors/their colleagues are generally idiots?

 

[ridethecliche]

The issue is that race is a placeholder/proxy variable right now since we don't have more information and that's the best predictor we have right now. It's not the only thing that matters, but till we can more readily ascertain a genetic basis and the costs become more reasonable, it's something that's important to think about in terms of risk.

 

[terra330]

Clinical Orthopaedics and Related Research, Jul 2011, 1859-70 "Defining Racial and Ethnic Disparities in Pain Management"
"Racial/ethnic minorities consistently receive less adequate treatment for acute and chronic pain than non-Hispanic whites, even after controlling for age, gender, and pain intensity."

Did they account for income or insurance? I often have to remind my primary care doc that I have kick ass insurance and that its okay to prescribe the medication he mentioned (he'll often mention that X medication is good but he's not sure if my insurance would cover it, I guess because most of his patients are lower SES)

 

[Psai]

I am an orthopedic surgeon I agree it is not medically accepted to give patient's less medication based on race. I'm sure this disparity varies depending on institution and social make up of the populations served.

But there are studies that shows that it happens.

American Journal of Public Health, Dec 2003 2067-73 "Racial and Ethnic Disparities in Emergency Department Analgesic Prescription"

Clinical Orthopaedics and Related Research, Jul 2011, 1859-70 "Defining Racial and Ethnic Disparities in Pain Management"
"Racial/ethnic minorities consistently receive less adequate treatment for acute and chronic pain than non-Hispanic whites, even after controlling for age, gender, and pain intensity."

You argue that if a black person comes in with chest pain, sickle cell should be on the differential. Ok, but sickle cell disease affects only 1 in 500 African-Americans. The chest pain is much more likely to be coming for another cause. Would you delay potentially life saving interventions for the <1% chance that this is sickle cell. Also you don't need to use race as a marker for potentially having sickle cell there is a test! I strongly believe most people over the age of 12 with sickle cell are aware of their diagnosis, and you just have to ask them. You don't need to assume that the black person with chest pain is probably a sickler.

Melanoma is more common in Europeans agreed. But black people can still get it as well.

I am not arguing that there aren't differences between racial groups when it comes to certain diseases. That is a fact. My issue and I think the purpose of the TED talk is to say there are better ways of identifying a patient's risk for certain illnesses, and if we just assume that race best because it is the most convenient then we may be doing our patients a disservice.

I find it hard to believe that you are actually a doctor

 

[idontwatchgreysanatomy]

First Hispanic is an ethnicity describing a very diverse group of people, there are Hispanics who would be considered white, and those who may call themselves black, or Native American and everywhere in between.

second there is a test for sickle cell. You don't have to use race as a factor in determining if the black patient with chest pain maybe having a heart attack or sickle cell crisis. You can get the test and find out, or simply ask the patient.

Hey now, he's taken a couple classes about this subject so he thinks it makes him an expert lol.

 

[LuckyMD2b]

I find it hard to believe that you are actually a doctor

If I'm not a doctor you must be a pretty pathetic medical student, as you have yet to provide one coherent argument to refute anything I have said.

 

[LuckyMD2b]

Do you seriously think an EM doc will delay lifesaving tests/treatments (like getting an ekg) by just considering something on the differential?

Why does every poster on that particular side of the argument indirectly contend that doctors/their colleagues are generally idiots?

Of course I don't think EM docs will delay lifesaving treatments because of a differential, that's almost as absurd as using a socially determined construct to try to predict a biological trait.

 

[idontwatchgreysanatomy]

Of course I don't think EM docs will delay lifesaving treatments because of a differential, that's almost as absurd as using a socially determined construct to try to predict a biological trait.

Got eem

 

[medic86]

Got eem

Yeah, not really. Especially since s/he just dismantled his/her previous assertion that was being used to further an argument.

 

[LuckyMD2b]

Yeah, not really. Especially since s/he just dismantled his/her previous assertion that was being used to further an argument.

I'm going to need you to re-read the posts again big guy. That previous "assertion" was a rhetorical question to highlight the absurdity of using sickle cell in the differential diagnosis of chest pain in an African-American patient.

 

[whatever5]

You are referring to the difference between GFR estimates seems so negligible that “I’m not sure why it’s a point of concern.” If the difference between estimates are negligible why should they be used at all?

She isn’t making an argument as to the ethics of using race, she is arguing that it isn’t scientific; which is absolutely correct.

Intravenous drug use is a direct risk factor for HIV transmission, your comparison is illogical.

You say, "...she cites the cause as sociocultural rather than genetic - this may be true in some cases, but who cares?"

It amazes me that the same people who will argue for the scientific basis for the importance of considering race are also arguing against considering the many profound social impacts of race.

There doesn't seem to be a reason to use it at all, but again it's a non-issue and not an example of the downfalls of race based medicine.

Using race in a clinical evaluation is as scientific as using statistics. You know person A belongs to group B, and group B has a higher likelihood of suffering from conditions X, Y, Z - so you keep in mind to screen person A for X, Y and Z. Where's the harm? To a clinician, it doesn't matter if the people in group B have a higher likelihood of suffering from certain conditions due to genetic or sociocultural causes, what's important is that the clinician doesn't miss anything.

FWIW, I'm pretty much what the internet would call a SJW and there are many real reasons to be concerned about race based medicine, but this lady brings up barely any of them (the only one i remember is the pain study). Instead, she is advocating for some idyllic world where doctors have all the time and money in the world to sequence your entire genome rather than using a statistical shortcut.

 

[LuckyMD2b]

There doesn't seem to be a reason to use it at all, but again it's a non-issue and not an example of the downfalls of race based medicine.

Using race in a clinical evaluation is as scientific as using statistics. You know person A belongs to group B, and group B has a higher likelihood of suffering from conditions X, Y, Z - so you keep in mind to screen person A for X, Y and Z. Where's the harm? To a clinician, it doesn't matter if the people in group B have a higher likelihood of suffering from certain conditions due to genetic or sociocultural causes, what's important is that the clinician doesn't miss anything.

FWIW, I'm pretty much what the internet would call a SJW and there are many real reasons to be concerned about race based medicine, but this lady brings up barely any of them (the only one i remember is the pain study). Instead, she is advocating for some idyllic world where doctors have all the time and money in the world to sequence your entire genome rather than using a statistical shortcut.

Do you want your physician to make treatment decisions for you based on statistical shortcuts?

I don't.

Your last creatinine was 0.6, today it is 1.2. If you were white then your GFR would be in the abnormal range and we would start treatment, but you're black so it's ok. We'll just keep an eye on it. See you next time, please see the secretary on the way out to pay your copay.

I understand that there are differences in the prevalence and outcomes for many diseases based on race and in those cases it is more than appropriate to have different recommendations for screening tests, diagnosis and in some cases treatment. But for the vast majority of cases I think that a patient's race is better a predictor of social and economic issues rather than biological, and we should keep that in mind to ensure that we aren't reinforcing racial disparities.

Race isn't like ABO blood type. There is no test to scientifically distinguish between groups, because there is such a high degree of variation in the measurement then we have to consider the uncertainty in any results that we draw from the use of the variable.

 

[Psai]

You have to be kidding me with this nonsense

 

[getdown]

Do you want your physician to make treatment decisions for you based on statistical shortcuts?

I don't.

Your last creatinine was 0.6, today it is 1.2. If you were white then your GFR would be in the abnormal range and we would start treatment, but you're black so it's ok. We'll just keep an eye on it. See you next time, please see the secretary on the way out to pay your copay.

I understand that there are differences in the prevalence and outcomes for many diseases based on race and in those cases it is more than appropriate to have different recommendations for screening tests, diagnosis and in some cases treatment. But for the vast majority of cases I think that a patient's race is better a predictor of social and economic issues rather than biological, and we should keep that in mind to ensure that we aren't reinforcing racial disparities.

Race isn't like ABO blood type. There is no test to scientifically distinguish between groups, because there is such a high degree of variation in the measurement then we have to consider the uncertainty in any results that we draw from the use of the variable.

We already do treat our patients based on statistical shortcuts. That's why there are all these studies out there comparing different drugs against each other and doing sub-group analysis on them to see which patients are going to be most benefited from them. Will all patients in that subgroup respond the same way? No. But are you saying we should spend thousands of dollars and delay treatment so that we can workup every patient from scratch and eliminate the perception that race should not be used in medicine? Ideally, that'd be nice but realistically quite unlikely until someone finds a cheap genomic test.

As for your sickle cell disease statement. One should definitely consider it as a differential. How does considering it as a differential delay life saving treatment? Someone with chest pain you're already ordering EKG, trops, CBC, BMP, CXR per usual before you even see the patient. This is more likely especially if you're working in areas (D.C., the south) that have a higher percentage of African Americans. And like you said you can easily figure it out by asking them which is probably better than just assuming they don't have it since it's <1% of the population. Which would be a disservice to their medical care.

Any reasonable doctor who see their patient's Cr jump from 0.6 to 1.2 will put pause and work it up. To assume most will just attributed to race variability is just silly at best and malpractice at worst. Now if their baseline is around 1-1.1 then no worries, but if the baseline is 0.5-0.6 then worry.

Look, ultimately this argument as a whole is pretty stupid. No matter which side of the line you stand on the ultimate goal is to take the best care possible for your patients. If you're using race for their good i.e considering disease that are more prevalent in that patient population, treatment options and screening options that are more beneficial in that patient population this shouldn't be an issue. But if you're using race a surrogate to NOT make reasonable treatment plans like in the pain med paper then yeah it's totally wrong. But really, we're all on the same team, just looking at a variable from two different perspectives.