Bitter clingers die young

The map of counties with decreasing life expectancy among women looks a lot like the map of counties that voted more Republican in 2008 than they had in 2004.

Life expectancies increase over time. That’s pretty much a law of nature. Absent an epidemic, seeing them move in the wrong direction is a strong diagnostic of social dislocation (e.g., Russian male death rates before and after Gorbachev).

So this graphic from the LA Times, showing declining female life expectancy in some U.S. counties over the decade 1997-2007, is pretty scary:

Does the map remind you of anything?


Now, maps are tricky things; the areas don’t match perfectly (West Virginia has a big vote swing but no increasing-mortality counties) and it may be that the mortality-rate increases weren’t from the same social strata as the move toward voting Republican. But something is happening in that area, and it really ain’t good.

It would be nice if our political and media class worried about real problems like this one and real threats like global warming rather than mostly imaginary problems and threats having to do with public finance.

Update: Here’s the picture for males, over a 20-year rather than a 10-year timespan. To my eyeball, the match to the vote-swing map is even closer. Is Tea Partying a mortality risk?

Underlying study here. Not cheerful reading.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact:

30 thoughts on “Bitter clingers die young”

  1. Weird feature: the difference between southern OK and north TX. Possible explanation (good for that area only): different levels of Native American population, plus alcohol.

  2. The declines in life expectancy in Southwestern Virginia had already made the local news when they were still preliminary. These declines are broadly entrenched throughout Appalachia. If I had to guess, it is what I used to call the troika because I saw it so often in disability cases I worked on: diabetes, hypertension and obesity, combined with difficulty making and/or accessing lifestyle and medical interventions.

    I do wonder whether the generation that is behind mine will live as long — as obesity and diabetes become more common at a younger age, the notion that people will start making lifestyle changes they could not manage when they were healthy strikes me as a chimera. Even with great drugs, these diseases are really hard to manage and put all kinds of stress on many different organs.

    SD and OK changes are almost certainly related to Native American populations, which are notoriously vulnerable to the troika.

  3. 1197-2007. Was that the decade when the calendar changed from Julian to Gregorian and a number of centuries went missing as a result?

  4. “Life expectancy” has serious problems with definition and sampling. Not only do the maps not overlap, but the concept of “life expectancy” in a specific area allows bizarre results due to migration, with no real effect on general human longevity.

  5. It would be interesting to investigate why the decline in life expectancy stops at the PA border. It’s pretty widely accepted that western PA, where I grew up, is culturally not that different from WV. Could the difference be related to PA gov’t programs?

    As an aside, Clearlake, CA, is a patch of Appalachian afflictions (the troika + meth), thus the red spot on the map there.

  6. What is the emigration rate out of these areas? If those young and healthy are exiting to go to the coastal cities it might explain the result as those being left behind.

  7. Redwave, which do you find especially funny: shorter life expectancies, or the risk of global catastrophe? The Red team has always had an admirable sense of humor.

  8. If I were paranoid, I’d suggest that being a political enemy of the Administration wasn’t very healthy… Unsurprisingly, when the government is dispensing a huge fraction of the nation’s wealth, not supporting that government can be a bad decision, economically. And being poor is the largest risk factor for premature death.

    Instead I’ll suggest that the data are too crude to draw any conclusions at all from. They collected a lot of data, but by combining them into “average life expectancy”, they threw most of it away. And they’re very bad at explaining what their charts are supposed to be displaying.

  9. I’m sorry, Brett, perhaps I missed something there? Being a political enemy of which adminstration, from 1997-2007? Was there some secret Clinton revenge pogrom put in place while he was being impeached for a blowjob, that lasted through the Bush years?

  10. Brett apparently can’t read the dates covered by the study, despite the fact that they have already been discussed in comments.

    Note for Brett: for 70% of the study time frame, the enemies of the administration were in completely different parts of the country.

  11. According to the study, the six lowest male life expectancy counties are: Mississippi (Holmes (81% Obama), Tunica (76% Obama), Quitman (67% Obama)) then West Virginia (McDowell (53% Obama)) then Mississippi (Sharkey (68% Obama), Humphrey (71% Obama)(tie)).

    You may want to consult this map, which agrees with the exception of Appalachia and places with high Native American population.

  12. Poor rural whites, poor rural blacks, poor rural Indians. Meth. Diabetes. Tea Party. These are symptoms of a dysfunctional economy for many parts of the country. Did you read the description of Cairo, IL during the floods? There is barely anything there worth saving. No one has a clue on how to fix it. No one.

  13. “West Virginia has a big vote swing but no increasing-mortality counties”

    Huh? The map shows at least a dozen counties with declining life expectancies in WV.

    Taking a quick look at their spreadsheet, all but 2 of these counties lost (female) population over the period 1987-2007. I’d be surprised if population movement wasn’t part of the explanation of these nationwide trends–younger and healthier people are presumably more likely to move away from economically depressed areas.

  14. I suspect the map would look a bit different on both ends of the life expectancy range if you threw out the low population counties. There are counties in Texas where no one is born or dies for decades at a time. And there are quite a few counties around the country that are experiencing population crashes.

  15. 1. Why would per county be a good way to track these figures? 1a. What should we make of the fact that most of these appear to be low population and low population density counties?

    2. How do you control for emigration from these counties to richer counties? Could this be a largely self selection artifact (more educated people move out of these counties for example)?

    3. Why aren’t we using the age adjusted figures which would reveal whether or not this is a mortality at birth issue (where nearly all dramatic life expectancy changes occur) or adult mortality figure? The two have VERY different policy implications.

  16. Actually, Charles, if you read the study you find that they joined counties of low population together with adjacent counties of similar demographics, to create a minimum county population floor, to deal with that problem.

    But, of course, that doesn’t deal with the issue of low population density, which I think is a big issue here: For a lot of causes of death, survival rates are strongly driven by how fast you can get to emergency care. Back in the 80’s my dad had a massive heart attack. He happened, by pure chance, to be visiting friends down in the city, mere blocks from a good hospital, and lived another couple of decades. If he’d been back on the farm, he would have been dead before we could have gotten him to an emergency room. Not because we were poor, or Republicans, because it would have just taken too long to get treatment.

    But it’s not like you can plop fully staffed emergency rooms on a 20 mile grid across the entire country. Even if it weren’t for the cost, there aren’t that many doctors. Not a problem in many countries with higher population densities. Perhaps some system of high speed transport, or telepresence for doctors would help.

    Sebastian: Agreed on your point 3, that’s part of what I was getting at, by saying that in boiling it all down to life expectancy, they threw away most of the information they’d collected. What are they dying of, at what age?

  17. Mark, I’ll bet that it’s less the bitter clingers dying, and more the poor people f*cked by the policies of the bitter clingers. I remember the maps from one of the ‘Red State-Blue State’ articles, which showed the state-level breakdown of Democratic/GOP voting by income levels. The map for the bottom third income group was blue for almost all states.

  18. Brett, the problem with seeing low density as a lurking variable that might make the data look disproportionately worse is that the effects are being compared WITHIN as well as between geographic locations. In other words, the inability to access ER care as a function of the rural nature of the county or counties is either unchanged or changed: If it’s unchanged, then it is more likely that the health of people has changed for the worse. If an ER or hospital has closed, then reduced access to health care services might indeed explain at least some of the change. The latter would at least be more amenable to straightforward policy changes.

    There might also be an intermediate explanation: younger people who in the past might have been present to intervene and provide faster or better access to services or other healthy behaviors (nutrition, taking medicine as prescribed, etc.) have moved away. This does seem to be happening in many rural counties, leaving a mostly aging population to make their own arrangements.

  19. Barry at 5:26 makes a big point — the purple map shows the higher-income voters that cause the mess, the red map at the top shows the impact on the victims. Also, those poor, rural counties are indeed losing all of their young people. In Appalachia, most of those who can get out, do.

  20. Mark,
    How did the authors define “life expectancy”? How do you? People usually intend “life expectancy at birth”, which, for people born today, is a projection. If this is a projection from real-world data, you face two enormous issues: 1. huge assumptions about he evolution of treatment and 2. sampling of deaths (who gets counted, and where?).

  21. Stupidity (defined as a particular basket of cultural traits, not as something you are born with) is the underlying causal variable in pretty much all these sorts of correlations.

    Good luck, academics, in trying to phrase this in a non-contentious way that won’t be recognized and picked up upon by the leaders of this self-same stupid class! You can hardly call it “Southern Syndrome” or “Christian Overload Disease”. And if you try to be clever (“Albion’s Syndrome”) you’ll be understood anyway. I think the best you can do is label this “Factor A”.
    Even then, of course, how do you fix the problem? You can hardly get away with telling this society “destroy your crappy culture and adopt a better one” — these aren’t Arabs or Greeks or Russians, after all.

  22. I’ll repost it here:

    According to the study, the six lowest male life expectancy counties are: Mississippi (Holmes (81% Obama), Tunica (76% Obama), Quitman (67% Obama)) then West Virginia (McDowell (53% Obama)) then Mississippi (Sharkey (68% Obama), Humphrey (71% Obama)(tie)).

    You may want to consult this map, which agrees with the exception of Appalachia and places with high Native American population.

  23. “You can hardly get away with telling this society “destroy your crappy culture and adopt a better one” — these aren’t Arabs or Greeks or Russians, after all.”

    Got that right. And crappy cultures aren’t limited to white poor people, either. What do you think that “legacy of slavery’ liberals are always complaining about is, anyway?

  24. Mark, aside from the most obvious point that correlation, even if you bothered to claim one statistically, says little or nothing about causation, I find it mildly hilarious that you would be trying to make a political point about these maps. Therefore, since I know you are an intelligent person, I gave you the benefit of the doubt and determined this was your attempt at (sarcastic) humor. If not, I really have to wonder. Maybe since females have teded to vote for D’s in recent years, the counties where they are dieing a little faster are shifting a bit to the R column, since the D’s are not as efficient as getting out the “dead vote” as we conservatives claim. In any event, I expect the gender gap peaked a while ago so maybe you can relax on this.

    Look, if you can expound a ludicrous theory, why can’t I?

  25. Data really needs to control for net migration into and out of the counties before one can start making generalizations. Interesting though. Why only women one has to ask.

  26. At best guess, you’ve got increased social pressure from a combination of economic dislocation and cuts to general ‘welfare’ budgets (pre natal clinics, etc.). Falling further behind the median American family.

    The dovetail with the ‘Tea Party’ is of course these people are angry, trapped– they know something is wrong.

    skilled politicians with strong media strategies have tapped that into an inchoate rage against government. Britain’s poorest boroughs oscillate between hard left and right wing groups like the English Defence League and the ex National Front British National Party.

    It’s a potent force, though, that sense of rage and dislocation. It’s not a reliable force for right wing politicians, just as militant trade unionism became the bane of British Labour governments and the Labour Party itself. It’s quite easy to become a captive of your own base, and wreck the party trying to free it: ask Harold Wilson, James Callaghan and Neil Kinnock– the last 3 Labour leaders before T Blair (excepting the brief reign of the late John Smith). Tony Benn was our Michelle Bachmann for decades, harrying from the Left. And a number of local governments (Liverpool!) were taken over by extreme leftists (Militant Tendency). Middle class Britain fled Labour in droves. De-toxifying the Labour brand was the work of nearly 2 decades pre Blair’s first PMership in 1997.

    Interesting to know what is killing those women. We know some of the pathologies with black women (diabetes and blood pressure in particular) but some of those counties are very white? Late diagnosis and treatment of ovarian and breast cancer? A sign of cutbacks in public health programmes? General rise in obesity?

    Note the Appalachians/ western parts of the ‘Old South’ are genetically closest in the USA to Scotland. And Scotland has one of the lowest life expectancies in Europe, due to smoking, alcohol abuse (not just amount drunk, but the fact that it is drunk in binges), poor diet (the home of the deep fried Mars bar) and an ‘x’ factor which might be vitamin D deficiency (Scotland is 55-60 degrees N, very little sunlight in winter, and a generally cloudy and wet climate in which the inhabitants tend to stay indoors). But that pathology is a male thing particularly.

Comments are closed.