Hacker News new | comments | show | ask | jobs | submit login
The Widely Divergent Life Expectancy in Every County in the US (thesoundingline.com)
45 points by Four_Star 5 days ago | hide | past | web | favorite | 39 comments





Note that’s it’s also incredibly localized, even in stereotypically “wealthy” areas. For example, life expectancy varies by >10 years across neighborhoods in San Francisco County (https://www.cityofberkeley.info/uploadedFiles/Health_Human_S..., slide 32).

For anyone who wants to check out the direct correlation mentioned in the article between this heatmap and the obesity heatmap, here is a link: https://maxmasnick.com/2011/11/15/obesity_by_county/

Unlike the map linked in the article this obesity heatmap is also by county so it makes for an excellent comparison.


Interesting they don't mention their extreme example (Oglala Lakota County) is an American Indian reservation.

because it shouldn't matter.

The separation of powers from any reservation to the Federal Trust is completely arbitrary and is only active when it is convenient. Congress can do anything it wants to a reservation and faces zero consequences. Host States can and actually do almost anything to reservations and face zero consequences.

and that can extend to building up and improving conditions closer to a base line. even when recognizing that each reservation is entirely different, despite some common outcomes.


It matters in terms of understanding how much of the gap in lifespan could be addressed specifically by solely addressing issues on reservations.

> by solely addressing issues on reservations.

The IHS itself is the culprit. It's hard to improve life expectancy when your only access to medical care is through a system that's not only overcrowded and underfunded, but which has truly zero incentive to provide quality medical care to its patients (and a long, documented history of corruption and outright exploitation of its patients).


It is one culprit of many.

> because it shouldn't matter.

It absolutely, 100% does matter, because Native American reservations[0] receive their healthcare exclusively through the IHS. The IHS is a completely separate system that is exclusively used by Native Americans, and it has massive documented logistical and ethical problems going back all the way to its founding.

It's downright deceptive not to mention that the data is so heavily shaped by this structural problem, because it completely changes the message people take away from it.

[0] including Oglala Lakota County


Healthcare does not represent a 10 year life expectancy age gap alone.

Healthcare has the largest influence on life expectancy from infant mortality, but the 0.5% increase only changes life expectancy by ~6 months.


> Healthcare has the largest influence on life expectancy from infant mortality, but the 0.5% increase only changes life expectancy by ~6 months.

That is simply not true for the IHS. The IHS is way, way worse than anything you can imagine in the rest of the US. It has literally committed genocide (intentionally) under the ruse of medical care, and genocide has a funny way of decreasing life expectancy.


I am not defending IHS, if you say it's doing a poor job then it likely is.

However, other factors are still important. EX: "At 43%, adult obesity rates in Oglala Lakota County were 17 percentage points higher than rates of adult obesity nationally and 13 points higher than statewide rates."

Further, you can find several other counties with rather similar life expectancy results that don't have IHS. So, simply pointing at IHS is not the complete story.


[dead]


Since you won't stop posting unsubstantive flamebait like we've asked many times before, we've banned the account.

Isn’t this basically just a proxy of poverty and drug use?

I think a more interesting map would be life expectancy of those over 70.


This data is meaningless for analysis without exposing the racial breakdown. That South Dakota county is almost certainly nearly all American Indians. The big red patch in the south is very likely largely influenced by lower African American lifespans.

How does that make it meaningless? Yes, certain segments of the population are even more underserved than others, but that’s just more evidence for the fact that we’re doing a crap job of keeping people healthy.

For one thing, it makes the title a bit deceptive. If the variance in lifespan is actually caused by something other than location, then where you live doesn’t actually affect your lifespan.

Because race is genetic and genetic variations lead to wildly disparate "outcomes." Your zip code is one of the weakest influences on your health and lifespan, relative to other scientifically proven factors.

4 of the top 5 counties are above 80% Native American. The remaining one is mostly prison inmates.

None of the top 5 has any real correlation to a geographic area per se.

Thus "Depending on Where You Live" isn't really a great title.


> How does that make it meaningless? Yes, certain segments of the population are even more underserved than others, but that’s just more evidence for the fact that we’re doing a crap job of keeping people healthy.

Because by not breaking out the single largest compounding factor, it's masking the real problem underlying this issue: there's a massive disparity in outcomes for people of different races.


Perhaps poverty, access to healthcare, or education is a bigger factor? Do you have evidence race is the "real problem"?

In Washington state, the counties with the worst mortality are 95% white, with 5% native (Grays Harbor and Okanagan)

(poverty, education, race are of course very correlated)


Race in fact does play a big factor in health disparities. For example, African Americans are at much higher risk for heart health related issues. And when heart disease is the number 1 killer of people of all races.. well that just doesn't bode well for the life expectancy of AA.

And it's not just because minority populations have worse access to healthcare. There are in fact genetic components that predispose different races to different diseases.

Japanese people, Glaucoma. AA, heart disease. White people, Celiac.

Basically there's multiple components: Race, Healthcare access, Poverty, Education, Local Cultural proclivities for: diet, exercise, etc.

https://jasn.asnjournals.org/content/26/2/247.full


> Perhaps poverty, access to healthcare, or education is a bigger factor? Do you have evidence race is the "real problem"?

You're asking if I have evidence that the IHS is an absolute disaster, which impacts mortality rates and life expectancy?

Yes, that's a problem that's been very well-documented for the past seventy years.


> The big red patch in the south is very likely largely influenced by lower African American lifespans.

There is certainly a strong correlation[0].

[0]: https://kieranhealy.org/blog/archives/2015/06/12/americas-ur...


How shocking that a massively multi-ethnic country that contains 5% of the human population has very large gaps in life expectancy if you measure by county.

If you look at European countries, a much coarser level of measurement the gap is 83-71, 12 years.

https://en.m.wikipedia.org/wiki/List_of_European_countries_b...


It is shocking because we don't officially acknowledge the existence of social classes in the USA.

All men are created equal here. It's counter to our national narrative to think of communities and areas where your life is fucked just because you're born in the wrong spot.


Or fucked because you were born to poor uneducated parents in the right spot...

It's pretty shocking, because I don't think most Americans would suspect there is that much regional variation within the country. The US is frequently regarded as basically monolithic, with people moving from one place to another for jobs, family, cost-of-living reasons, basically assuming that life in one part will be largely similar to life in another part. This assumption is not backed up by data.

(That said, I would be interested in an analysis of whether or not your life expectancy varies with where you live, or where you are born. I.e. if I am from Boston but move to Mobile, do I end up with a Bostonian's life expectancy or a Mobilian's? How long does it take for the shorter life expectancy of someone from Mobile to override the longer life expectancy of someone from Boston? Or does it never do that? All would be interesting questions for research.)


So roughly half as much for a continent with over twice the population and 44 sovereign countries? The difference between areas in South Dakota and Colorado is bigger than that between Luxembourg and Moldova

Going by state in the US the difference is 6.3. If you go by all US states and territories it’ll go up to 7.9. The smaller your unit of measurement the greater the range will be. Counties are already prettt ridiculous given that some have populations in the millions and some I’m sure are in the hundreds.

If you want to compare European life expectancies you might be able to do it by Eurostat statistical region but none will have populations in the single digit thousands.


The extreme range in the US is specifically driven by a few counties are either all, or mostly, Native American Reservations with very specific issues around unemployment, alcoholism, etc.

> Native American Reservations with very specific issues around unemployment, alcoholism, etc.

Native Americans typically receive their healthcare from the IHS, which is known for being an absolute disaster. It's not surprising (or news) that life expectancy is much lower for a population which receives its healthcare exclusively through a system which has a track record of neglecting its patients or using them as test subjects without their consent.


Healthcare is an issue, but its just one of the many issues on reservations.

Indian reservations have issues with poverty, unemployment, alcoholism, substandard housing, violence, and suicide.


Fair enough. Nonetheless I don't think the point of the article is to compare the US to Europe but simply to reflect on the wide range that exists within the US.

You'd need to also do the exact same thing in Europe and look at what kind of spreads exist inside of Europe as well.

Let's see the worst life expectancy areas of Russia for example. You're nearly dipping into the 50s for life expectancy in that case, and you're going to see a 30+ year gap from top to bottom across peak Europe vs the lowest areas. This has recently come to the fore with the Russian Government's move to push the retirement age so low that it nearly clips the average man off of it.

"A stroll around the Zapadnoye Cemetery reveals how few Novgorod men actually make it to old age. It’s easy to find graves of young men who died in accidents, wars or gangster feuds—many with epitaphs such as “came to a tragic end.” For the rest, the typical age of death is around 62. Graves of men who lived beyond 70, however, are pretty hard to find."

https://www.bloomberg.com/news/features/2018-08-27/putin-s-m...


Because the problems that are correlated with lower life expectancy are highly localized.

You can't compare France (83) to Ukraine (71), you're talking 15x gdp (and just being there would show you how wildly different they are).

There are still places in the US where electricity isn't used, the primary languages aren't European, and the gdp per person is that of a third world country. If it's fair to compare those places to New York or San Francisco, France vs Ukraine doesn't seem terribly far-fetched.

It's kind of disappointing, but I'm not at all shocked to see the South having the largest area in the red, and the North East and the West coast doing well. I'd really like to see this map over time, though. I think that would be a really interesting addition to the data, as it would give a really good picture of which areas are getting better or worse.

I built this viz a few years back to visualize the top causes of death by state : https://www.infino.me/mortality/usmap

Still interesting to me how bad people in Alaska have it.




Applications are open for YC Winter 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: