Race and medicine

This is very interesting. In 1950, black and white life expectancy at birth were very different: a white baby born that year could expect to live an average of 68.2 years, versus just 60.2 years for a black baby. But there was no significant difference between blacks and whites at retirement: at the age of 65, both could expect to live about another 13.9 years.

For the rest of the century, life expectancy at birth narrows considerably; the difference between groups is now five years, and falling. But the gap at retirement widens. White 65 year olds now get an extra 1.8 years compared to their black countrymen; at 75, the gap is a smaller, but still significant, 8 1/2 months.

It's tempting to blame this on disparity in health care access, and I wouldn't be exactly surprised to find that this is part of the case. But the subtler problem is that because Europeans are the major market, most drugs are designed for Europeans; and it turns out that there are races, at least as far as medicine is concerned. Blacks are more prone to diseases like diabetes and hypertension, and they tend to get more difficult to treat forms of some diseases such as artery blockage. While some of the medical differences are due to poverty, education, and possibly to racial disparities in treatment, many of the differences persist even when things like income and education are controlled for.

Comments

This comment has been deleted for being off topic

Posted by maggie razzi | September 29, 2007 4:54 PM

Do you have any citations for your claim?

Also, how would you control for racial disparities in access to health care if what you're testing for is racial disparity?

Posted by DivGuy | September 29, 2007 6:14 PM

Potential reasons for disparities:
1) Poorer health access related to income or maldistribution of other resources that could affect health (e.g. education)
2) Individual culturally-determined behaviors, tastes and habits
3) Lag time (relevant for wider disparities with age), where the effect of poor access twenty years ago is still having an effect
4) Racism
5) Exaggerated perception of racism by African-Americans (think Tuskeegee)
6) Shortage of racially or culturally competent care (patients prefer patients who are similar or of the same identifiable group)
7) Mistrust of health care system, independent of race of provider or perception of racism or education etc.
8) Genetics (a distasteful and potential dangerous path to take, but the racial differences in outcomes of diabetes, hypertension and obesity are compelling)

Have I missed something? I'm sure I have.

It's not enough to discuss disparities without thinking about the mechanisms by which race and ethnicity can affect health outcomes.

Posted by Zagreus Ammon | September 29, 2007 6:15 PM

I expect researchers in the field have thought of these, but a couple others come to mind:

9) Geographic distribution. People in the Northeast and the Southeast probably face different enough environments to matter, when you're summing outcomes over millions of people. Also, you can just have different quality of care available when you live in different areas; small towns in rural areas aren't likely to have a great selection of specialists.

10) Differences in education or literacy. If you can't read the doctor's instructions, you might not take the medicine properly. I'd expect this to be an especially big issue for people who don't speak English.

11) Differences in broad knowledge of warning signs. If you don't know that fast-growing moles with irregular boundaries are a danger sign, maybe you wait too long to get that skin cancer treated.

12) Employer or other organization activities. My employer has a once-a-year low-cost screening for a bunch of stuff and flu-shots, some employers have health clubs on site, even dieting help on site. (I've heard really amazing things about Microsoft in this regard.)

I'm sure there are others.

Posted by albatross | September 29, 2007 11:48 PM

http://www.gnxp.com/blog/2006/02/iq-and-health.php

Posted by Unknown Healer | September 30, 2007 1:07 AM

Maybe you should just avoid talking about race, Megan.

Posted by brad | September 30, 2007 2:20 AM

Up until Social Security I bet Vital Statistics obtained age at death by asking the family. And in 1935 Social Security probably got age by asking the people (within reason). I wouldn't be surprised if many blacks over 65 in 1950 had no good idea of how old they were (the black schools available in the late 1800s were even worse than they were in 1940). As a result, the numbers might be seriously messed up.

Posted by Rachel | September 30, 2007 8:20 AM

This is a bit off-topic, but one thing that you need to keep in mind when dealing with life expectancy is that it's measured backwards. To figure out life expectancy for someone BORN today, they need to take data from people who have already DIED. This gives you a number, but it doesn't tell you anything about how long a baby born today should expect to live.

So when comparing the white/black life expectancy for today, you need to realize that that has almost nothing to do with recent medicine and almost everything to do with medicine from 50+ years ago. But what's interesting is that the life expectancy at 65 is measuring effects of recent advances (because you don't have to wait as long for someone age 65 to die).

I don't know what it all means, but it becomes very difficult to say what's causing things here.

Posted by Josh | September 30, 2007 8:57 AM

This comment deleted for racist trolling

Posted by Jozef | September 30, 2007 9:07 AM

This comment deleted for random trolling

Posted by Shorter Megan McCardle/Steve Sailer/Robert Stacy McCain | September 30, 2007 11:58 AM

Remember to add into the equation that there are known physiological differences between races, apart from specific disease predispositions, e.g. iron absorption and metabolism, calcium metabolism, bone density. These facts have been suppressed by political correctness.

Posted by Hugo Lambrechts | September 30, 2007 12:18 PM

Clarification from someone who does this stuff for a living:

It's not an issue of "there are races". The issue is that there are genetic variants, many of which are distributed unequally across (what patients report as) race. Given perfect data, "race" wouldn't be the issue, but roughly correlated information you have is worth more than a perfectly informative datum that you don't.

So it's not true that "there are races", just like Genome Project heavy breathing about how all human variation is randomly distributed across all of humanity isn't true either.

Posted by CC | September 30, 2007 1:39 PM

CC, do I understand you aright? Race doesn't exist. But "race" is a source of correlates such that it will be fruitful to act as if "race" does exist until, at some time in the future, more detailed info is available.

How does that differ operationally from saying "until we have near-perfect info on you as an individual, the best we can do pro tem is treat you - in part - as a member of a race?" Is that what you mean?

Posted by dearieme | September 30, 2007 4:20 PM

dearieme:

I'd put the "in part" in bold, and maybe blink tags, but otherwise that seems like a reasonable restatement of my point.

Posted by CC | September 30, 2007 7:33 PM

It should be uncontroversial to point out that extended families exist. Everybody belongs to many different extended families -- your mom's, your dad's, paternal grandmother's, etc.

Some larger extended families, however, are partly inbred -- when you go back 40 generations in your family tree to, roughly, 1000 AD, there are a trillion openings for direct biological ancestors. So, it's certain that some of your ancestors did double-duty. This means that large extended families tend to be partly closed off genealogically. And that's the technical definition -- "partly inbred extended families" -- of what we refer to as racial groups.

Posted by Steve Sailer | October 2, 2007 8:43 PM