• Women’s health and children’s health

    The following is a guest post from Bill Gardner, a psychologist who studies the mental health service system for children. Bill is an American living in Canada and a professor of pediatrics at Dalhousie University (Nova Scotia) and the Ohio State University. Bill blogs at Inequalities, and you can follow him on Twitter at @Bill_Gardner.

    There is a frightening graph in a recent article in Health Affairs by David Kindig and Erika Cheng. Kindig and Cheng looked at trends in male and female mortality rates from 1992–96 to 2002–06 in 3,140 US counties. What they found was that female mortality rates increased in 42.8% of counties (male mortality rates increased in only 3.4%). The counties are mapped below: red means that female mortality worsened. You can see a strong regional pattern: just about every county showed had worsened female mortality in several southern states, while no county showed such decline in New England. There are many questions about what explains this pattern. For example, did healthier women migrate out of the south from 1992 to 2006? Nevertheless, the map depicts a shocking pattern of female hardship, primarily in the southeast and midwest.

    KindigFigure1

    When I look at the graph, however, I am concerned not just about the women, but also about their children. The mental and physical health of mothers is a key determinant in children’s growth and development. What the map shows is that America has regions of communities with high concentrations of women experiencing substantial hardship. When women are not able to maintain their own health, how well can they nurture their children?

    This trend is amazing in a historical context. Overall US life expectancy had been increasing steadily over the decades. Before seeing data like these, I had the simple view that increasing life expectancy was part of a general increase in human well-being, powered by the steady growth in economic well-being. In fact, US GDP per capita increased from $24,400 in 1992 to $44,600 in 2006 (in current US $). This is a huge shot for the average American (although it was less for the median American). But a large subgroup of women was apparently left behind. I speculate, but do not have the expertise to test, that what we are seeing is that the widely discussed increase in economic inequality in late 20th century America is also an increase in geographic inequality. My guess is that not only are rich Americans rapidly pulling ahead of poor Americans, but that these groups are also increasingly segregated by region.

    It takes a family and a village to raise a child. What happens when the moms in the village all get crushed?

    Share
    Comments closed
     
    • Between 1990 and 2010, our nation’s maternal mortality rate worsened from 12 to 20 while the developed nations of the world as a group decreased from 26 to 16 ( TRENDS IN MATERNAL MORTALITY 1990 TO 2010 published by the United Nations in 2012 at http://www.unfpa.org). AMNESTY INTERNATIONAL USA published their analysis in 2010 regarding our nation’s maternal mortality rate. My take: until there is equitably accessible and culturally acceptable Primary Health Care for each citizen commnity by community, our healthcare industry will be neither equitably efficient nor reliably effective. Our nation’s maternal mortality rate, community by community, is the most prominent outcome from the paradign paralysis afflicting our nation’s healthcare.

      I am particularly saddened by the fact that the CDC stopped collecting maternal mortality information, state by state, in 2008. The last reports indicated that four states had achieved a 5 year, maternal mortality rate at the level of the best 10 nation’s or the world: Alaska, Indiana, Maine and Vermont.

      • Paul, thanks for the link. I am a big supporter of primary care, but FWIW Kindig & Cheng’s do not think that the solution is in the health care system.

        • The relationship between health care access and mortality is a bit tricky. In the US, for example, the causes of death which pull down male life expectancy are homicides and car accidents, and in old age the main causes of death are still of the type the health care sector cannot cure. So in some ways it’s not fair to expect a strong impact from health care directly to reduced mortality for those groups, because people are often already dead when brought to hospital after violence and car accidents and because we still can’t cure most kinds of cancer or completely cure cardiovascular diseases or strokes.

          In other cases it matters greatly, however. Early detection may prolong lives after cancer diagnosis, at least in some kinds of cancers, and screening may prevent cancer of the colon, for example. Finding a way to treat AIDS affects mortality figures a lot in younger age groups, and so on.

    • The resemblance to county-by-county election maps is uncanny.

      • Yes. I didn’t want to comment on that because I have no good explanation.

        • Well, there’s a perfectly good explanation: the Republican party lies to the poor and less-educated white portion of the population and persuades them to vote for policies that are extremely detrimental to them. The main trick is to blow the “I don’t want my taxes going to _them_” dog whistle.

          But this isn’t something that will pass this blog’s moderation, although it should be. (Dear moderator, you really ought to accept this.)

          • The same sort of pattern is explained very well in The Dictator’s Handbook and Why Nations Fail. Both texts illustrate and explain the consequences of inclusive and exclusive political systems (not to be confused with ideologies, which are merely the garments of various more or less inclusive regimes). The authors of The Dictator’s Handbook point to infant mortality, but I would suspect women’s health would not be vastly different.

        • I mean, you can even see the splits within the states. Florida’s liberal south versus conservative north. Virginia’s liberal north and conservative south. Conservative north-east California versus the coast. Urban Georgia (Atlanta and Athens) versus the rest of the state. I suppose I should thank Mitt Romney for Romneycare, seeing as all of Massachusetts is doing great. I guess the “gift” of healthcare can improve peoples lives, huh?

      • My sentiment EXACTLY.

    • The midwestern counties that worsened look mostly like rural counties, and most of the urban and suburban midwest looks blue to me. Many rural counties have seen an outmigration of their young people to cities and suburbs, so I wouldn’t be shocked if this is a compositional effect, where the population in rural areas has a higher mortality rate because it has become disproportionately elderly.

      This seems to be borne out somewhat by this map of median age by county:
      http://www.arcgis.com/home/item.html?id=fce0ca8972ae4268bc4a69443b8d1ef5

      • Alan,
        I agree that it would be important to look at that effect. See my sentence about migration in the post. Interestingly, though, the migration effect should influence both male and female mortality, since both genders would presumably move for jobs.

        • For such a dramatic increase in such a short time, the big killers probably have to be involved, don’t they? Tobacco, alcohol and obesity?

          There hasn’t been any dramatic increase in women’s obesity relative to men in this time period, but there has been in obesity in general. From what I’ve read, men lose more years to ischemic heart disease despite having comparable obesity levels – but maybe on very high obesity levels, women’s advantage disappear in this area?

          Smoking prevalence, as far as I know, hasn’t changed much between the genders since 1990 either. But looking a few more decades back, far more men than women smoked (although both genders smoked more). Women are definitively catching up in lung cancer deaths. Unlike with heart disease, where men who get it die much earlier than women on average, with lung cancer women die slightly earlier.

          Alcohol consumption has gone slightly down in the US since 1990, but is rising. I can’t find any stats on men vs. women and this annoys me, but I do believe it has become more equal, with men still consuming somewhat more. Since men have a higher tolerance for alcohol though (the CDC limits for harmful use are half for women what they are for men), a more equal consumption distribution may still mean far more health risks for women.

          • Would the change in smoking prevalence have to happen in the 1990s? People start smoking as adolescents, but the increased cancers appear mostly when they become middle aged or older. So there could be a complicated set of cohort dynamics occurring here.

      • I see that in areas I’m familiar with. In Ks and Oklahoma, the counties home to Wichita, Tulsa and Oklahoma City are blue.

    • County maps patterns don’t reflect population. It would help to see the relationship by per capita per state, urban rural, etc.

    • The other factor that needs to be considered in the frequency of single-parenting by women.

      • I thought that was going down, or at least was stable.

        More likely is Republican animosity towards women’s health care. Planned Parenthood does a lot more than abortions. Reduce affordable health care and mortality statistics get worse. No surprise whatsoever.

        (Again, dear moderator, this is something that needs to be said.)

        • David: “Republican animosity towards women’s health care” as a dominant cause of higher mortality? Are you for real? I’m no fan of the GOP generally, but the percentage of time any woman will spend accessing an abortion clinic is miniscule and would hardly account for that. Any thoughts on Republican California?

          I refer you to Bill’s repeated comment on this post urging people to consult Kindig’s data. There is probably a pretty nuanced story behind all this.

          • The person who questioned whether there was a link between the increase in women’s mortality in politically red states & Republican attempts to close Planned Parenthood did explicitly point out that Planned Parenthood does far, far more than provide abortions. When I was younger (i.e., before I had a job with insurance), I went to PP for all my gynecological care. Or more precisely, I went for birth control pills, but couldn’t get them until I had a Pap smear, blood pressure check, etc. For about 10 years, I saw a non-PP doctor not more than two times, but I did go to PP clinics at least twice a year due to the length of the pill prescription. If someone is shutting down PP clinics or keeping them from opening, then they are having an impact on women’s access to ob/gyn care.

            & since this increase in mortality for women is tied to an increase in the maternal mortality rate (it’s not just that women who make it to 70 no longer making it to 75), you have to wonder what kind of care childbearing women are getting, including counseling they are getting about their health prior to getting pregnant.

            • Thanks for doing my homework!

              See my statistical comment below: my thought here is that women’s medical problems are more responsive to improvements and degradations in quality of care during childbearing years than men’s would be in that same age bracket, so this worsening of life expectancy is quite reasonable to expect to be a side effect of political attacks on women’s health care providers. Medicare expansion could help here, but the same folks attacking PP are resisting Medicare expansion.

              Over here (Japan), maternal and infant mortality and morbidity are pretty much the lowest in the world. Everyone is insured, patient out of pocket costs are minimal, prenatal care is aggressive to the point of being intrusive, and it’s a major social issue when something goes wrong in even one case. The Japanese care about doing a good job in prenatal/postnatal care, and it shows in the statistics. IMHO, we could, too, if we tried. But we’re too busy playing politics.

    • The graph of increases in female mortality closely tracks county-by-county obesity rates (for both men and women):http://hyperplanes.blogspot.com/2013/04/sarah-kliff-on-female-mortality-rates.html

      I don’t have a theory for why the correlation would be so much stronger for women than men.

    • What age are the women dying? What is their economic status? You group in three categories, but what is the magnitude of each. So many questions. Im not sure you have sufficient data to support your hypothesis in this article. Perhaps the published papers do? If women out live men and there is a population boom, you would expect a large increase in male mortality followed by a large increase in female mortality much later. Finally, you state it went up in 46 percent of counties, compared to 3 percent for men. You dont state how much MORE the mortality rate was. There are more women than men. I guess i would like to understand the differential magnitude vs differential population. Is this mortality rate per women capita or per capita? One would expect a large difference in female mortality per capita in most counties because women outnumber men, withthis trend increasing im the near past. Am i missing something? please help. can you graph the magnitude difference normalized for the sex segmentaion?

    • Two things.
      First, to what extent does this reflect both increases in women working and decreases in working conditions for wage earners relative to upper class employees?
      Second, while a “by county” approach makes for a pretty map, counting Harris County, TX (Houston) against a larger-sized but almost unpopulated county in the Panhandle distorts the effects. Is there a way to visualize this data adjusted for population?

      • The first question is good and I do not know the answer. Remember: I am commenting on an article by Kindig. He has the data.

        Second, yes there are ways to distort geography so that the area reflects population size. This has been done with maps of US election results. They look weird and probably lead to other kinds of misinterpretation, but they do help counteract the problem that you and other commenters are (reasonably) concerned about.

      • This isn’t the first chart to use county-level data (Facebook released one recently) and it probably won’t be last. I’m not sure I understand our fascination with counties as they exist outside of presidential elections (in which they arguably mean something). Unfortunately, this chart does not present enough information to support its conclusion. As well, the original report is restricted for my access, so further validation isn’t possible.

        We don’t really understand the definitions used. “Worsening” may simply mean some negative score, regardless of how strongly negative it is. In some instances, a slight negative score may just be noise. In fact, the entire map may just be noise including how specific regions appear clustered. On this map, a state represents five dimensions: general latitude and longitude, height and width, and quantitative grading with color. Only one of these dimensions, color, is truly meant to communicate. The rest of these dimensions are incidental to the analysis–but that doesn’t stop our brains from trying to make meaning from them. Location, height, and width are all preattentive attributes that our brain process subconsciously (see the work by Colin Ware). A larger county with a jarring color will communicate a large quantitative value even as we tell ourselves that counties don’t correlate with population.

        Consider the year 2000 presidential election. One candidate could pick up what appears to be more geographical area than the other while losing still the popular vote. Likewise, health outcomes like mortality rates for women might fair far better in the aggregate but appear amplified when visualized at the county-level. What justification is there to use county-level data? If we can’t justify their use, our conclusions approach GIGO analysis.

        That isn’t to say I disagree Mr. Gardner’s concerns, which I found to be both poignant and alarming (as they exist outside of this map). And, as I’m sure others will rebut, perhaps a map like this, even if not entirely correct, does good by bringing attention to the subject. Perhaps. But from a practitioner’s perspective in the field of visualization, I think the map fails to bring the required rigor to support its conclusion.

        • Many sorts of data are collected by counties. That’s where births, deaths, marriages, and home ownership are recorded. I assume that’s why by-county is the way a lot of data are shown, regardless of population and other interesting variables.

          • But that’s exactly the problem isn’t it? Why aren’t those other “interesting” (more like, “necessary”) variables that might soften or amplify county-level data included in the analysis? It doesn’t really matter how the data are collected.

    • I can’t get the most recent issues of Health Affairs. Like others pointed out, you really should explain what mortality rates are referred to here. On the face of it, “mortality rate” should refer to the percentage of women dying per year. This depends most heavily on the age structure. (Here’s a factoid: developed countries tend to have higher mortality rates than many underdeveloped countries despite much higher life expectancy, precisely because of their age structure.) So the color of the county could mostly reflect demographic change. I trust the authors have thought of this and corrected for the effect but I have no way to know from the blog article.

    • Without having yet read the original article, I very much doubt that the increased mortality rates of women have much impact on women in the child-bearing ages, so the reference to children’s health probably does not apply. My guess is that the women die younger, but not in larger numbers at very young ages. More like at 69 rather than 81.

      As several people have mentioned, the numbers of women who live in those counties would be more useful than the map. I doubt that they are nearly half of all women, for example.

      The causes matter tremendously, of course, and more research into them is needed.

      • “My guess is that the women die younger, but not in larger numbers at very young ages. More like at 69 rather than 81.”

        My guess would be the opposite: one person dying at 25 has a much larger effect on total life expectancy than one person dying at 69. If life expectancy is 75, then dying at 25 reduces total years lived for the population (the numerator in the life expectancy number) by 50, whereas someone dying at 69 reduces said value by a mere 6 years.

    • Can I point out the giant elephant in the room? What do women do that men don’t? Have kids.

      Dying from pregnancy, childbirth, or abortion happens to young women and skews the life expectancy data much more than dying of age correlated diseases. Dying at 25 vs 65 instead of 70. So obesity numbers and smoking numbers would all be on the order of age 65. Maternal deaths on the order of age 25.

      These higher mortality regions correlate with many things, but one of those things is severe limits on bodily autonomy in the health care system. Plot this data vs. distance to nearest abortion provider and see if you’re still puzzled.

      It also addresses the urban/rural divide: Urban areas are easier places to get abortions because there are more providers, public transport, and more anonymity. The right to lifers talk about the baby’s right to life and forget that the mother has a right to life as well. And this is what happens. Or so I hypothesize.

      • Yes it is important to look at the healthcare of women in their childbearing years. I know I was shocked to discover that maternal mortality is so much higher in the US than in the EU. For example it is five times higher than Greece and three times higher than Ireland, neither of which are particuraly wealthy countries and certainly not compared to the US.

      • “Dying from pregnancy, childbirth, or abortion happens to young women and skews the life expectancy data much more than dying of age correlated diseases. Dying at 25 vs 65 instead of 70. So obesity numbers and smoking numbers would all be on the order of age 65. Maternal deaths on the order of age 25.”

        Except that obesity and pregnancy is a _bad_ combination. If I’m wrong in my claim that it’s Republican attacks on women’s health care that’s the problem, this will probably be the reason. (I.e. increased maternal mortality due to obesity-caused complications.)

        http://www.mayoclinic.com/health/pregnancy-and-obesity/MY01943

        Back in the 80s, a friend at a large US company (AT&T) reported that his boss was pregnant and not able to find an OBGYN who would take her. She was overweight, smoked, and married to a lawyer.

        • I’m both appalled and not surprised that your boss couldn’t find a doctor. At some point, self preservation is more important than someone else’s life… which is the basic premise behind abortion in the first place, but works elsewhere too.

          But I thought of another possible cause for the male mortality to improve while women’s goes down.
          1st, not having the numbers, I don’t know how much they’ve gone down. I’d like to trust that it is by more than the margin of error, but I don’t know. (Even so, 46% vs 3% – individually being on the wrong side of the margin of error vs. collectively still says something to the trend, but I digress.)

          Since the advent of potable water/sanitation, vaccines, antibiotics, and improved maternity outcomes, life expectancy shot through the roof. And for most developed world countries, women’s expectancy is 5-10 years higher than men’s. There are two broad areas where men die young in significant numbers: (1) just after the onset of puberty when all that new testosterone makes them risk prone, and (2) starting in their 50s from heart attacks. There have been vast strides made in men’s heart health in the last few decades, so many more men are living an extra decade or two. While some of this advancement has helped women, it really hasn’t helped them in anywhere near the numbers it has helped men.

          Also, the “nanny state” improvements on youth safety have likely kept a lot of teen boys from fatal accidents. Seat belts, airbags, helmet laws, etc… keep a lot of boys from dying of youthful folly. A few deaths saved here would, like maternal deaths in women, make a bigger impact on overall life expectancy. And while they also save women, they save more guys because guys do more stuff that gets them killed early. In fact a group of archaeologists figured out that they could estimate age of puberty of boys in historical records by when their teen “death spike” is. For women, they can approximate it by average age of first birth, which has more diligent records than when a guy’s voice gets deeper. So reducing the severity of the teenaged danger years will have a disproportionate effect on boys over girls.

          But I think it’s all of those things – bad/no sex ed, reduced access to contraception and abortion AND improved survivability of teenaged hijinks and heart health improvements that primarily help men.

    • Trying to figure out why you cite the increase in nominal GDP per capita rather than something else.

      Real GDP per capita increased by around 33% from 1992-2006.

      The poorest quintile of U.S. households received 3.8% of U.S. income in 1992 ($11,396 per family, in 2011 dollars). In 2006 they received 3.4% of U.S. income ($12,663 per family, in 2011 dollars).

      Meanwhile, the richest 5% of households received 18.6% of U.S. income in 1992 ($227,123 per family, in 2011 dollars) and 22.3% of U.S. income in 2006 ($331,756 per family, in 2011 dollars).

      (All data from U.S. Bureau of the Census)

      • You are right, GDP shouldn’t be given in nominal dollars. I _thought_ I was presenting GDP in current dollars. At least that is what the World Bank page said it was.

    • The distortional effects of looking at the data on a county basis have already been noted; it would be valuable to look at raw numbers as well. It’s also puzzling why rural counties in Nevada, New Mexico, Colorado, and Arizona would show marked improvement, while rural counties in Wyoming, Idaho, and Montana would show worsening. I wonder if the apparently large differentials are caused by relatively small differences in numbers, given the small population densities in many of these western counties.

      What’s also interesting is that the counties that include Detroit, Cleveland, Chicago, Milwaukee, Denver, and Los Angeles show improvement, even though all of those cities have substantial disadvantaged and/or impoverished minority populations. Does this mean in spite of the well-publicized issues facing major urban areas that adequate health care is more likely to be available there than in rural counties?

      • Mr. H. as someone who grew up in one of the “cow counties” of Nevada, one of the big issues is availability of health care, along with ofther issues duiscussed (wealth/proverty) (medical insurance) etc. My town is the county seat of a county with a declining population through loss of jobs (mining and the federal govt being the major employers). Keeping or attracting professional care is almost impossible, even kids who grew up there, got a medical degree and still love the town, can only afford to travel in one day a week to see patients. The nearest fully staffed surgical hospital is 130 miles away, Other places such as Mississippi and Louisiana have the same problems. Its not all lack of health care in the US, its’s the distribution that may be part of the problem.

    • Rising per capita GDP does not mean rising prosperity for the “average person. Has the “pool” of uninsured people who aren’t poor enough for medicaid, and who aren’t rich enough to buy health insurance increased?

    • The article says that health care has nothing to do with these mortality rates, yet a 2009 Harvard study concluded that 45,000 Americans die of preventable illness each year due to lack of health insurance.

      It can’t be both, can it?

    • For an explanation of how social service administration varies locally in parallel with political conservatism, see Disciplining the Poor, by Soss, Fording, and Schram. If the pattern they observe in TANF administration also applies to health services, it might provide an explanation for the similarity of this health map to the electroal one.

      Also, I would guess that the difference between the pattern for men and women has less to do with reproductive health than with violence, which affects men’s health much more.

    • PRWORA replaced AFDC with TANF and ended entitlement to cash assistance for low-income families, meaning that eligible families may be denied aid even if they are eligible. Under TANF, states have broad discretion to determine who is eligible for benefits and services.

      The Personal Responsibility Work Opportunity Reconciliation Act of 1996 (PRWORA) was passed by the House of Representatives on July 31, 1996 and by the Senate on August 1, 1996. On August 22, 1996 President Bill Clinton signed the bill into law. (PL 104-193)

      The legislation created the block grant cash assistance program known as Temporary Assistance for Needy Families (TANF).

      This system was intentionally designed to prevent assistance to women with children by the complicated language and requirements (you’d need lawyer representation in the application process) that have not been included in this brief synopsis of TANF. However, the evidence of higher female mortality rates best summarizes Clinton’s PWORA outcome of poverty and increase mortality rates of women (esp. single women raising children) in a society that takes advantage of women’s unpaid Labor (double meaning).

      TANF major provisions include: Able bodied adult cash assistance recipients must work or be in work activities (job training, subsidized employment, job search and job readiness assistance, etc.) after two years of receiving assistance. This provision is subject to good cause exemptions on a limited basis.
      Receipt of cash assistance under Temporary Assistance is restricted to a lifetime limit of five years.
      States are required to meet certain levels of recipient work participation or face losing a portion of their TANF block grant: Twenty-five percent of a state’s eligible welfare population (families that include an adult or minor child head of household receiving cash assistance) must be participating in approved work activities (outlined in the bill) in FY 1997. This level of participation rises by 5 percent each year until FY 2002, when it becomes 50 percent thereafter. The work participation rate for two-parent families is required to be at 75 percent in FY 1997 -1998 and 90 percent in 1999 and beyond.
      Individuals receiving cash assistance (unless exempt) must work a minimum number of hours per week (averaged over a month) to be counted toward meeting the work participation rate

      From Wikipedia:Temporary Assistance for Needy Families: The reform granted states wide discretion of how to distribute TANF entitlements. States also have the authority to eliminate payments to recipients altogether.

      While employment of TANF recipients increased in the early years of reform, it declined in the later period after reform, particularly after 2000. From 2000-2005, employment among TANF recipients declined by 6.5 percent. Among welfare leavers, it was estimated that close to two-thirds worked at a future point in time. About 20 percent of welfare leavers are not working, without a spouse, and without any public assistance. Leavers who left welfare because of sanctions (time limits or failure to meet program requirements) fared comparably worse than those who left welfare voluntarily. Sanctioned welfare recipients have employment rates that are, on average, 20 percent below those who left for reasons other than sanctions.

      While the participation of many low-income single parents in the labor market has increased, their earnings and wages remained low, and their employment was concentrated in low-wage occupations and industries. Over three quarters (78 percent) of employed low-income single parents were concentrated in 4 typically low-wage occupations: service; administrative support and clerical; operators, fabricators, and laborers; and sales and related jobs.While the average income among TANF recipients increased over the early years of reform, it has become stagnant in the later period; for welfare leavers, their average income remained steady or declined in the later years.Studies that compared household income (includes welfare benefits) before and after leaving welfare find that between one-third and one-half of welfare leavers had decreased income after leaving welfare.

      During the 1990s, poverty among single-mother and their families declined rapidly from 35.4 percent in 1992 to 24.7 percent in 2000, a new historic low. However, due to the fact that low-income mothers who left welfare are likely to be concentrated in low-wage occupations, the decline in public assistance caseloads has not translated easily into reduction in poverty. The number of poor female-headed families with children dropped from 3.8 million to 3.1 million between 1994 and 1999, a 22 percent decline compared to a 48 percent decline in caseloads. As a result, the share of working poor in the U.S. population rose, as some women left public assistance for employment but remained poor. Most studies have found that poverty is quite high among welfare leavers. Depending on the source of the data, estimates of poverty among leavers vary from about 48 percent to 74 percent

      TANF requirements have led to massive drops in the number of people receiving cash benefits since 1996, but there has been little change in the national poverty rate during this time.

      However, the Nordic model compensates women’s labor unlike the USA’s patriarchal system. Denmark, Finland, Iceland, Norway and Sweden are the forerunners in designing family-friendly policies. They guarantee generous parental leave with high compensation rates for mothers to take considerable time out of work in connection with childbirths, and to return to their previous jobs afterwards with publicly funded daycare. It’s a model of family-friendly policies, and family welfare that celebrates and values women’s roles in society. Perhaps, one day the United States will begin compensating women’s labor, rather than impoverishing them for the rest of their lives.

    • The five year limit on cash assistance alone made TANF one of the worst public policies ever enacted. To this day, my older daughter is angry at Clinton for signing TANF. I don’t blame her.

    • With regard to Japan’s low infant mortality rate mentioned in the comments: I guess the fact that there are few unmarried mothers, most births occur between 25-29, low number of crack babies, etc. have nothing to do with Japan’s low infant mortality rate.

      One question, how can the country that has the best low birth weight infant survival (which is where a vast bulk of infant mortalities occur) have such a poor record in overall infant survival? Is it the healthcare system or something else?

      From the abstract: “These findings suggest that improving health outcomes across the United States will require increased public and private investment in the social and environmental determinants of health—beyond an exclusive focus on access to care or individual health behavior.”

    • I cam here from Jesse’s Cafe Americain. These charts end up saying more about the pundits who analyze them. Hmmm, lowering of woen’s life expectancy relative to men. Could it be the rise in obesity combined with improvement to medication for heart conditiosn which affect mena bit mroe than women? I know the self back patting urge and tsk tsking of the right is too much for most libs to resist, but god forbid people put the cheese doodles and big gulps down and take a daily walk. Access to health care has nothing to do with becoming 30 lbs overweight.

      Here’s the unspoken thign that no one will mention: I bet that graph also lines up well with lower income whites and the general black + hispanic population. The CDC mentions that obesity hits women, minorities adn the disadvantaged much more than others, and New England is nearly snow white and has relatively high average incomes.

      Impulse control, accountabiltiy, being responsible for yourself. I know these are old concepts, and that no one is ever at fault for their actions (except Wall St.), but spare me the smug lefty behavior.

      • Since the largest swaths of blue are in heavily hispanic areas (Los Angeles, New Mexico, Southern Texas, and Miami), I’d say race isn’t a factor. I’d also say obesity isn’t a factor since it is more likely to cause men heart attacks than women. Improved heart health, very likely. The rest of it, no. Just no.

    • I see a lot of assumption going on. We don’t know what the pattern was before 1992. Was there perfection before? Why assume, without proof, that the cause is political? Let’s not jump to conclusions.

    • My speculation: it is the food system of GMOs, HFCS and CAFO meats. Also our increasingly toxic chemical environment. We should adopt the EU rules for food and body care- lotions, shampoos, anything put on the skin.

      Women are more susceptible to these toxins compared to men because of body fat indexes. French GMO study shows increase of all cancers by 400%, organ damage and infertility.

      Purdue University study shows a 50% abortion rate in cattle fed GMOs as well as a new pathogen associated with GMO crops that has never been studied as Dr. Huber was ‘retired’ after his discovery.

      We are seeing all of these results in the American population.

    • I wonder if single older women are considered….I understand this classification of women are among the poorest in the nation.

    • @ Louis Wheeler: I think you have phrased your comment incorrectly, Mr. Wheeler. Poor people can’t afford medical care, so they don’t see doctors for problems that can easily be fixed in a modern nation. Poor people don’t have a personal doctor, because their money goes into putting shoes on their kids and just getting by. Poor people don’t live in areas where there ARE doctors. Poor people go to emergency rooms when something is really bad, overloading our trauma system. Poor people don’t get follow-up visits, or take their medicine (because they can’t afford the inflated prices of non-generic drugs ) or eat a good diet, because the part of the country they live is a food wasteland because they don’t know what to eat, don’t have the money to buy it, and its not available. How about I sum it up for you? “It sucks to be poor.”

    • I would like to see the comparison between aging populations, especially in the midwest and southeast, as these communities are generally those with the greatest mortality…

      When the population is impacted by lack of jobs, etc, has very little inflow of younger immigrant families, as may be the case in many of these impacted communities, it is not unusual to have greater mortality. Many of these counties might be consider all to be rural, but with dramatic ranges in terms of populations and services.

      Having grown up in rural North Dakota (actually all of ND is considered rural by many standards), there was a tremendous difference in health services in a county of 6,000 vs a county of 60,000. Apples and oranges, people.