Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Science

Race Cannot Be Used To Predict Heart Disease, Scientists Say (nytimes.com) 97

Doctors have long relied on a few key patient characteristics to assess risk of a heart attack or stroke, using a calculus that considers blood pressure, cholesterol, smoking and diabetes status, as well as demographics: age, sex and race. Now, the American Heart Association is taking race out of the equation. From a report: The overhaul of the widely used cardiac-risk algorithm is an acknowledgment that, unlike sex or age, race identification in and of itself is not a biological risk factor. The scientists who modified the algorithm decided from the start that race itself did not belong in clinical tools used to guide medical decision making, even though race might serve as a proxy for certain social circumstances, genetic predispositions or environmental exposures that raise the risk of cardiovascular disease.

The revision comes amid rising concern about health equity and racial bias within the U.S. health care system, and is part of a broader trend toward removing race from a variety of clinical algorithms. "We should not be using race to inform whether someone gets a treatment or doesn't get a treatment," said Dr. Sadiya Khan, a preventive cardiologist at Northwestern University Feinberg School of Medicine, who chaired the statement writing committee for the American Heart Association, or A.H.A. The statement was published on Friday [PDF] in the association's journal, Circulation. An online calculator using the new algorithm, called PREVENT, is still in development.

This discussion has been archived. No new comments can be posted.

Race Cannot Be Used To Predict Heart Disease, Scientists Say

Comments Filter:
  • Feelings Science (Score:5, Insightful)

    by Tyr07 ( 8900565 ) on Friday November 17, 2023 @03:08PM (#64012961)

    That's what this sounds like. If you're deliberately ignoring medical and scientific results in the name of equality people are going to die.

    I don't care if it's because my race typically is exposed to X that I might get X disease or not, that should make them more aware of it to better treat me. It doesn't matter to me if it's because of my genetics, or the region I typically live in.

    If you've actually found medical and scientific evidence now that it actually doesn't apply to the results, then the math should show that which would make it irrelevant and you wouldn't need to remove it from being listed.

    People are going to die for someone elses feelings.

    • I don't care if it's because my race typically is exposed to X that I might get X disease or not, ...

      Another reason to not use Twitter -- I mean, X. :-)

  • Disappointing (Score:5, Interesting)

    by Whateverthisis ( 7004192 ) on Friday November 17, 2023 @03:26PM (#64013019)
    It's a bit disappointing they're doing this. It's correct race shouldn't be a proxy for treatment, but not collecting the information doesn't necessarily solve that problem.

    Race in some ways is a doubling up of some of the other statistics. For example, more African Americans may have heart disease, but more African Americans do have diabetes than other ethnicities (14.7% of Native Americans, 12.5% for non-hispanic black, 11.8% hispanic, 9.5% asian, 7.4% white). Diabetes is a much more likely indicator of heart disease, so health information of course is more important. But there are significant lifestyle differences between the different ethnic groups in the US, and knowing that creates an opportunity for a more targeted preventative solution, as diabetes care for the white community is likely sourced from different lifestyle/diet issues than diabetes for the African American community. So if the action is "try to get people to change so they're not diabetic", a generic solution may not be the problem.

    There is a value in knowing how ethnic groups are affected by this, from a sociological point of view, as it might highlight certain avenues for improvement. Disappointing.

    • Re:Disappointing (Score:4, Interesting)

      by brunes69 ( 86786 ) <slashdot@nOSpam.keirstead.org> on Friday November 17, 2023 @03:42PM (#64013063)

      What does race even mean in 2023 though?

      What percentage of your ethnicity needs to be from Africa, and how far back, to be considered an "African-American"?

      Do we need to run an ancestry DNA analysis every time someone goes to the doctor?

      Just because two people both have dark skin does not mean their genomes are related. That may have been true 150 years ago but it certainly is not today.

      • Re:Disappointing (Score:5, Insightful)

        by Baron_Yam ( 643147 ) on Friday November 17, 2023 @03:54PM (#64013097)

        It's not binary, you can treat things as percentage odds.

        What you're looking for is the most probable cause for a condition, and diagnostically it is best to go from most likely to least, moving along the list as you eliminate each possibility in order.

        Anything else means, at best, an increased average time and number of tests before a correct diagnosis and treatment which means statistically worse outcomes.

        This is political correctness actually hurting people in ways you will be able to measure.

      • My health insurance provides genetic testing to identify known abnormalities related to particular diseases

        It also included a map of my genetic background, that was eye-opening

        I believe that this will become the new standard of care

      • by quenda ( 644621 )

        What does race even mean in 2023 though?

        It does not matter here. We don't need any deep meaning for the word. The context is the United States, which has for whatever reason collected statistics based on
        certain categories. If people can be consistently allocated to a category, and that allocation is more predictive of certain outcomes than other available information, then it is useful. It only stops being useful when we have sufficient more detailed information, such as lifestyle, diet, blood pressure, what genes are present, etc.
        Stat

    • and it's definition is fuzzy and changing at best. We're not purebloods in the slightest, and that means there's no way to really effectively use "race" as we call it (i.e. skin color and *maybe* nation of origin) effectively. Worse it discourages doctors from using actual tests to look for the real markers that can lead to heart disease & failure. And it *especially* discourages insurance companies from allowing those studies. Don't forget America's got death panels.
      • I think that the Dr Seuss story about the star-bellied sneetches [wikipedia.org]is one of the most insightful explorations of race I have encountered, particularly when it proposes a method to eliminate the physical identifiers of race

      • So I'm going to disagree with you here. It isn't something we made up.

        It is true that the concept of race has spread into so many areas that it perhaps shouldn't, and to that degree the very idea of race is confusing. There are all sorts of poor racial stereotypes that are just wrong and hurtful, and the most common issue is the difference between race and culture. Culture is a set of behaviors, but race/ethnicity is something more concrete.

        And when it comes to healthcare, race IS a thing. Sickl

  • by 93 Escort Wagon ( 326346 ) on Friday November 17, 2023 @03:29PM (#64013025)

    So if they've decided to remove this apparently useful proxy, I assume they will now start tracking all those factors for which race was providing a proxy? Because, if not, this is going to make cardiac care worse for members of those races - not better.

    • by bsolar ( 1176767 )

      So if they've decided to remove this apparently useful proxy, I assume they will now start tracking all those factors for which race was providing a proxy? Because, if not, this is going to make cardiac care worse for members of those races - not better.

      They tried to incorporate some of the underlying factors, but acknowledge the new approach is currently limited:

      However, despite interest in inclusion of measures that more directly reflect risk related to racism (eg, residential segregation, perceived racial discrimination) and additional individual- and place-based measures of social drivers (eg, income, education, residential green space), the lack of standardized assessment and capture in data sources was a key limitation. Therefore, although the PREVENT equations represent a critical step forward, integration of the social deprivation index is only a first step; the inclusion of relevant measures that represent individual experiences of discrimination, structural and systemic racism, and individual- and place-based SDOH should be a priority in risk prediction moving forward.

    • Hopefully the decision makers decide to dismiss this political activism.
  • by Harvey Manfrenjenson ( 1610637 ) on Friday November 17, 2023 @03:41PM (#64013059)

    From TFS: "The scientists who modified the algorithm decided from the start that race itself did not belong in clinical tools used to guide medical decision making, even though race might serve as a proxy for certain social circumstances, genetic predispositions or environmental exposures that raise the risk of cardiovascular disease."

    I'm not sure what this sentence is supposed to mean. Ethnic background is an independent predictor of risk for several common diseases-- for example, prostate cancer (more prevalent in Black people, less prevalent in northern Europeans, even less prevalent in Asians). For certain genetic diseases (e.g. sickle-cell disease, Tay-Sachs), ethnic background is a *strong* predictor. You ignore it at your peril.

    I don't know if ethnic background is an independent predictor of heart disease, or just a proxy for multiple other things. But that's a side issue, since the quoted sentence says "race does not belong in clinical tools used to guide medical decision-making". The prohibition isn't specific to heart disease.

    "Deciding things at the start" is a poor way to go about doing science.

    • by Calydor ( 739835 )

      That line alone reeks of the old, "Find evidence to support your hypothesis" trap.

      I'm certainly no expert in the field, and as a 40+ white European guy I probably have even less to say in this matter, but if skin color in any way whatsoever helps you determine likelihood of a disease, include skin color in your calculations!

    • The sheer number of women that have died because scientists all assumed they responded the same as men in everything from drug trials to heart attacks is staggering. Ignoring reality on feelsees only gets people killed, what happened to evidence based medicine?
      • by quenda ( 644621 )

        The sheer number of women that have died because scientists all assumed they responded the same ...

        You can replace "women" with "men" (or "elderly" or many other words) and it is still true. Simplifying assumptions are always needed. Stop with the identity wars please.

        what happened to evidence based medicine?

        Actually, I'd like to see some evidence for the scale of your vague claim, and some acknowledgement of *why* women are under-represented in early-stage clinical trials. The problem is hardly a secret.

    • by edwdig ( 47888 )

      I think they're saying that race is often used as a proxy for multiple other factors, like you suggested. And they don't think that's a good idea. They want to shift the models to more directly use the factors that really matter.

      I don't think they're trying to avoid using it in cases where it clearly does matter.

  • I think your performance in a 100 yard dash is a pretty good indicator of risk of heart attack.
  • Consider the Hispanic category. It literally means that you grew up speaking Spanish. Your parents could be Chinese, African, or Apache. Doesn't matter, you can identify as Hispanic.
    Then there's "Asian", which lumps Sri Lankans and Japanese into the same category. Probably Australian aborigines too.
    Humans in Africa have more genetic diversity that the rest of the species, so an homogeneous "Black" category almost certainly misses important variations.
    And don't get me started on mixed-race definitions.

    So no,

    • I don't know what Canadian doctors use, but the official police categories are "white" and "non-white".

      You know what? I don't care if it is genetically accurate... If you look 'hispanic' and cops are looking for you, they ought to be able to put 'hispanic' in the occurrence notes.

      • by _merlin ( 160982 )

        You can't "look Hispanic" by the US definition, because the US definition of Hispanic is people from Spanish-speaking countries in the Americas irrespective of race. A black Cuban, a native Mexican and a white Colombian are all "Hispanic" according to the US definition.

        • by kackle ( 910159 )

          A black Cuban, a native Mexican and a white Colombian are all "Hispanic" according to the US definition.

          Did they walk into a bar?

    • by quenda ( 644621 )

      Consider the Hispanic category. It literally means that you grew up speaking Spanish. Your parents could be Chinese,

      Technically, historically yes, but not in US English, where it is an ethnic term. It specifically refers to people from Latin America. Brazilians are usually included.
      Antonio Banderas in not hispanic, or "of colour". Louis CK is Hispanic.

  • The paper said they'd like to include the factors that more directly reflect the risk, but they're not able to because the data isn't available yet. This could be a good first step for making the assessments more accurate for everyone.
  • by hdyoung ( 5182939 ) on Friday November 17, 2023 @06:07PM (#64013421)
    deep into the raging furnace, here.

    Fact: one of the VERY FEW actual, documented, totally-for-real racial differences is in the amount of melanin that are produced by a person's melanocyte cells.

    Fact: there are melanocyte cells located in a person's heart.

    Ok, here's the point where I make massive disclaimer statements, to try and avoid a torch-wielding mob at my door. I'm not for or against any specific race. I'm not saying anything about "good" or "bad" or "something in between". No value judgements here. Everyone is a unique individual. Etc. etc.

    But skin-color actually correlates to melanocyte differences, and THERE ARE MELANOCYTES IN THE HEART. Are we absolutely sure race doesn't play a role in heart function? This isn't some white guy pontificating about how great it would be to classify people by race, with light skinned people at the top. I would imagine that most people, regardless of skin color, would really like to know if something about their biology points to potential heart issues in the future. Are we absolutely sure we want to apply hard-left DEI policies here and go totally racially blind?

    Jews have a well-documented genetic predisposition for Gaucher disease. White people in general have a well-understood tendency to alcoholism that's caused by their genetics. Is it a good idea to go absolutely colorblind when it comes to medical issues?

    Maybe I'm being racist here and I'm totally off base. If so, please, educate me. I can change.
    • >"Are we absolutely sure we want to apply hard-left DEI policies here and go totally racially blind?"

      Yes, far left is DEI and such. But going racially blind is NOT a left-wing concept anymore, it is a right-wing concept now. Funny how that works. Might depend on context, but, socially, if you want to not care about someone's race, you can't be a leftist.

      >"Maybe I'm being racist here and I'm totally off base. If so, please, educate me. I can change."

      You are not being racist. Although the definition

  • by imperious_rex ( 845595 ) on Friday November 17, 2023 @06:18PM (#64013443)
    If they're going to remove race from the equation in the name of equity, then I guess sex/gender (because sexism) is next? And then age (because ageism)? Gotta have equity/equality across the board, right? RIGHT?
  • I'd be perfectly fine if they said "we checked and there is NO MEANINGFUL STATISTICAL CORRELATION between race and heart disease" or it wouldn't even have to be that categorical...it could be that race suggests a relationship, but in fact this only proxies a collection of other factors which could be independently checked instead of using the imprecise aggregation of something as vague as "race".
    Then by all means, remove it.

    But they don't seem to mention anywhere that they used data to determine this , they

  • by WaffleMonster ( 969671 ) on Friday November 17, 2023 @08:11PM (#64013625)

    I was at least expecting a merit based argument for the change to be expressed. What I found instead was woke language and direct admission change were decided in advance for political reasons.

    Woke "modifiable" gibberish:

    "Although age and sex are not modifiable, they are both key components of CVD risk and are important predictors in CVD risk equations."

    "Furthermore, the inclusion of race in risk prediction may imply that differences by race are not modifiable and may reify race as a biological construct, which may worsen health disparities"

    Direct admission changes were NOT merit based:

    "Significant disparities exist whereby a disproportionate burden is experienced by individuals who identify as non-Hispanic Black, American Indian and Alaskan Native, or South Asian American individuals."

    "Because race is a social construct and an historically fraught proxy representing various lived experiences, there is the potential for the harmful interpretation that it represents a biological risk factor when included in risk prediction, which may result in race-specific treatment decisions. Therefore, it was decided a priori not to include race as a predictor in the development of PREVENT"

    "This is consistent with the growing consensus to remove the use of race from clinical algorithms broadly in medicine. Racism, and not race, structures our social and individual lived experiences, is associated with adverse SDOH, and represents a key driver of adverse CVD outcomes."

    What gets me is the obvious difference (skin color) appears to getting overlooked even though there is a substantial body of evidence linking trivially treatable impacts to relevant risk factors (e.g. diabetes and hypertension).

    https://www.mdpi.com/1660-4601... [mdpi.com]
    https://www.ncbi.nlm.nih.gov/p... [nih.gov]

    The very notion ideology should trump reality when ignorance of reality is actively harmful to people is something I don't know how to justify. It seems evil and absurd on its face.

  • I'm all for inclusion. But ignoring factors with a scientific and statistical basis because it hurts people in the feels is not the way to help under served communities or racial groups. This seems very much like the kind of attempt at inclusion that will lead to worse outcomes for those it is attempting to help.

    • Yeah. The same thing happened in the social sciences, and is now spreading into hard sciences. It doesn't matter that sex, ethnicity, or culture may be relevant to outcomes - simply ignore them, forgoing solutions that may otherwise have been found.

      It has one advantage in that we can save a lot of money on research if we simply assume differing outcomes are due to either racism, sexism, or whatever the current day istaphobia is. Factors like wealth and lifestyle choices after all rarely rise to significance

  • Seriously, being rich introduces a radically different set of health issues vs being poor. Yet, AMA removes it. Big mistake. It needs to be in there, as it will find a strong correlation between various diseases and class.
    • If only I could be honest about income on forms too.

      Race might used to have meant skin color. But is this context, race of the participant it is what box says that the individual checked. Is this verified? How? What, again, is race and how do to identify it using specific metrics that can be compared apples-to-apples so to speak or in this case the intentional differences of apples from oranges and oranges from plums? I might suggest DNA rather than identity, it might even contain nutritional identification

      • >But is this context, race of the participant it is what box says that the individual checked.

        As a proxy for social factors, this works wonderfully. For probable genetic factors, it's a lot weaker... but chances are nobody's calling themselves 'black' if their ethnicity is 'Indian'.

        What you're more likely to get is people who are of mixed heritage and have a strong personal association with one ancestral ethnic category and write that down - which still tells you they're more likely to have genes known

        • Except that race is not a social construct, no more than sex is a social construct . It is very much an issue in medical diagnosis though it is about genetics which plays less of issue than does income.
          • >Except that race is not a social construct

            You're going to need to rethink that, because race is when you call someone black, or Jewish, or white, or whatever. And that is not only loosely correlated with genetics, the definition of what puts someone in a racial category varies by where you are and the local standards.

            It is a social construct.

            • I do not need to re-think it all. You do.
              First, since when is 'jewish' a race? It is a RELIGION. For you to call someone that and think that it means race, implies that you are a racist.
              Secondly, as to black/white/etc, far too many ppl think that skin color is what makes up race. It is does not. Ancestry is what makes up race. A race is determined by a closed genetic group. Long ago that was the case. Now, in places like America, we are bulworthing things, but we still have it to a large degree. For exam
      • First, there is no reason why most ppl can be honest about income grouping. For example, 0-25k, 25-50k, 50-100k, 100-150k, 150-250k, and 250+k. And that needs to family income. Why do something like this? It makes a huge difference in what ppl eat, what/where stress and level come from, and what diseases are prevalent.
        for example, gout used to be thought of as white disease, but it is not. It is rich person disease. There are many different diseases that were falsely correlated with wrong indicators becaus
        • Well I am honest when it come to money business. But not demographic polling. You're right in that I'm kind of a dick about it. If I felt like it wasn't used against me more than for, I might be more willing to offer an honest picture of my demographics to the merely tangentially interested.
  • by markdavis ( 642305 ) on Saturday November 18, 2023 @09:36AM (#64014507)

    >"The scientists who modified the algorithm decided from the start that race itself did not belong..."

    Welcome to the new "science", where researchers decide, in advance, what data belong and don't belong, to support their feelings and social agenda. What the conclusion will be, regardless of the data. There is any wonder why people are trusting "the science" less and less.

    Disgusting.

    And for each disclosure like this, rest assured there are many, many others we don't know about.

  • Seems dumb to ignore data. Genetics play a huge part in health - including risk of heart disease.

    Really really dumb decision - can we keep politics out of science, please?
  • No it still can be used, and it does work. What the cognoscenti ruled is that it MUST not.

"I got everybody to pay up front...then I blew up their planet." "Now why didn't I think of that?" -- Post Bros. Comics

Working...