• The news on mortality isn’t getting any better

    In the latest issue of Health Affairs, “Mortality Under Age 50 Accounts For Much Of The Fact That US Life Expectancy Lags That Of Other High-Income Countries“:

    Life expectancy at birth in the United States is among the lowest of all high-income countries. Most recent studies have concentrated on older ages, finding that Americans have a lower life expectancy at age fifty and experience higher levels of disease and disability than do their counterparts in other industrialized nations. Using cross-national mortality data to identify the key age groups and causes of death responsible for these shortfalls, I found that mortality differences below age fifty account for two-thirds of the gap in life expectancy at birth between American males and their counterparts in sixteen comparison countries. Among females, the figure is two-fifths. The major causes of death responsible for the below-fifty trends are unintentional injuries, including drug overdose—a fact that constitutes the most striking finding from this study; noncommunicable diseases; perinatal conditions, such as pregnancy complications and birth trauma; and homicide. In all, this study highlights the importance of focusing on younger ages and on policies both to prevent the major causes of death below age fifty and to reduce social inequalities.

    Then, if you’re not feeling bad enough, “Even As Mortality Fell In Most US Counties, Female Mortality Nonetheless Rose In 42.8 Percent Of Counties From 1992 To 2006“:

    Researchers increasingly track variations in health outcomes across counties in the United States, but current ranking methods do not reflect changes in health outcomes over time. We examined trends in male and female mortality rates from 1992–96 to 2002–06 in 3,140 US counties. We found that female mortality rates increased in 42.8 percent of counties, while male mortality rates increased in only 3.4 percent. Several factors, including higher education levels, not being in the South or West, and low smoking rates, were associated with lower mortality rates. Medical care variables, such as proportions of primary care providers, were not associated with lower rates. 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.

    Mortality went up for women in more than 40% of counties. Seriously. Here’s the map:

    Change In Female Mortality Rates From 1992–96 To 2002–06 In US Counties

    Change In Female Mortality Rates From 1992–96 To 2002–06 In US Counties

    Think all of those red areas are going to invest in improving “social or environmental determinants of health” anytime soon?


    • Red states and blue states… kind of matches up with politics. I don’t think red states invest in health or women or the environment. (24 words)

    • Anyone with access to the gated article know if the authors corrected for demographic variations within the counties over time?

      The overlap between this map and the net shift of people and jobs away from the northeast, rust-belt, and California to the South between ~1990-and-2012 makes me wonder if underlying changes in the composition of many counties hasn’t played a role in modulating the mortality statistics more than any changes in clinical efficacy, public health provision, etc. If nothing else, it could seems like it could significantly confound the above analysis.


      Do you really think the red/blue state analysis holds? There are quite a few red counties within blue states.

      • I use the “squint test”. If you squint your eyes, things blur together and a larger pattern appears. Yes, there are red counties in blue states and blue counties in red states. For California, the red counties are rural, sparsely populated, conservative areas… Republican strongholds in a liberal state. Same for Nevada and Utah (but not liberal states).

        • Well – datasets that map statistical associations that aren’t sufficient to derive conclusions about causality generally wind up as ideological rorschach tests that we tend to project our preconceived ideas onto.

          I couldn’t find an ungated copy, but I did find a summary that presented the actual magnitudes of the changes under consideration here:

          “Women dying younger than expected in the whole country – accounting for every county – dropped from 324 to 318 per 100,000. However, in a little under half of the counties it actually increased from 317 to 333 per 100,000. ”

          So we’re talking about an increase of 16 premature deaths per 100,000 per year on average in the cases of the counties that saw their mortality stats worsening.

          One factor that the authors may or may not have completely controlled for is that the variance in sparsely populated areas will be significantly greater than that in areas with 10-100X as many persons, and I suspect that this effect will hold even over a 5 year sampling interval if the populations is sufficiently small relative to the average.

          Having said all of that – I suspect that the phenomenon that they are measuring *is* real and that premature mortality really is worsening amongst the cohort that they identify – poor women with relatively little education living in rural settings.

          My hunch is that smoking, obesity, etc are driving a good chunk of this change, but after hearing daily updates from a front line ER that treats lots of poor white women with low educations I can’t help but think that increasing deaths to presciption painkiller overdose are also driving a good chunk of the premature mortality. The overlap between the original maps above and the maps at the site below aren’t perfect, but it’s close enough to be plausible.


      • JayB, the article is currently ungated. IIRC, new Health Affairs are ungated for a short time (maybe for the month they are published?). Then they get gated. I think a lot of old ones are ungated. You can actually email the author and ask very nicely for a copy of the article, and they will often give them away.

        The authors controlled in their multivariate analysis for these demographics: ethnicity, median household income, % of high school and BA grads, % of single parent households, and % of children living in poverty. The authors also controlled for some medical care factors (e.g. density of primary care providers) and health behaviors (% of adults obese or smokers). The map represents just raw change in mortality.

        % of Hispanic residents was significantly associated with a reduction in mortality at the county level (as was % of BA grads, unsurprisingly). In epidemiology, there is a Hispanic paradox: despite having lower SES and poorer health status, Hispanics have health outcomes (e.g. mortality) that are comparable to Whites (and sometimes better). To date, I don’t believe there has been a satisfactory explanation (but see the Wikipedia article for some hypotheses).


        In other words, if you are thinking immigration, it looks like it has not been a big factor in mortality disparities at the county level. Still, immigrants have disparities in accessing health services that we need to address.

    • Considering the death rate of young people in Chicago due to violence does anyone think that blue area of the country doesn’t have poorer life expectancy?

      • That’s the comment you bring to the table after looking at that graph? Why would you mistrust that data? Does it just not fall in line with your “gut”?

        I’m more interested in what happened to Kentucky / West Virginia. Yikes.

        • Brian, I simply brought up a real time consideration. Are you saying that you think Chicago has a better life expectancy? If not what are you trying to say about Chicago?

          • I’m saying that it seemed odd to me that you’d immediately challenge the validity of the data without offering one iota of evidence. To me, I look at that graph and arrive at the conclusion that Chicago is getting better.

            Besides, that graph is noting trends – not absolute references. Chicago is trending towards improvements in life expectancy. Much of the south, particularly Kentucky, West Virginia, and Tennessee are trending in the other direction. You rather miss the point of the graph if you switch the conversation to talk about relative comparisons instead of which way those data points are moving.

            Given that I live near those areas (western PA), I was just wondering if someone had more details as to what is driving those trends.

            • Brian, the graph has to do with female mortality rates up till 2006. I discussed deaths due to violence in Chicago and most of them would be males. That would give at least a part of the explanation for the first study “Mortality Under Age 50 Accounts For Much Of The Fact That US Life Expectancy Lags That Of Other High-Income Countries“.

    • According to the World Health Organization report (“Trends in Maternal Mortality 1990-2010”), the maternal mortality rate for the developed countries of the world decreased from 26 in 1990 to 16 in 2010. For the United States, it INCREASED from 12 to 21. Really, is there anything else about our nation’s healthcare industry that is more important than those statistics?

    • The major causes of death responsible for the below-fifty trends are unintentional injuries, including drug overdose—a fact that constitutes the most striking finding from this study; noncommunicable diseases; perinatal conditions, such as pregnancy complications and birth trauma; and homicide.

      It looks like cultural and genetic factors that have nothing to do with healthcare access or delivery. I remember reading somewhere that poor Americans have as much access to health as poor people in the other industrialized countries.

      People in Scotland will not live as long as their English neighbours, new statistics show.

      The average Scottish man will live to the age of 75.9 while male Englishmen will live to 78.6.

      The Scottish have the same socialized medical care as do the English so this would imply a Genetic or cultural difference rather than a difference in medical care. BTW Many people in the USA are descended form Scottish immigrants.

      As far as the counties that would most likely be because the more successful people are leaving those counties.

      • Also if the black population is growing faster than the white population in a county one would expect a falling life expectancy because whites live significantly longer than blacks. That is just a demographic effect.