States and regions differ widely in longevity trends
January 10, 2018

The late Mayor Ed Koch of New York used to roll down the back window of his limousine when he was stopped at an intersection and yell out to nearby pedestrians: “How’m I doin’?” After an initial startle, many of them would yell back congratulations, complaints, or both. When Koch died, the New York Post published a memorable cover with his photo and the banner headline: “Ya did fine!”

Not perhaps the most sophisticated way to evaluate public policy, but we all should imitate his attitude. In public health, the best way to get a quick check on how we’re doing is to look at mortality rates—accounting for age, since an older population will almost inevitably have higher mortality rates than a young one.

NIA funds research on trends and differences in adult mortality rates, and other health indicators, looking for the risk factors that can be modified at any point in the lifecourse to improve outcomes.

Recent national and state estimates

Demographer Andrew Fenelon at the University of Maryland recently updated his analysis of trends in age-standardized death rates at older ages (55 and above) for the United States as a whole and separately for each state. An interactive visualization tool created by the Population Reference Bureau makes it easy to figure out how any state is doing in terms of health for older people.

The trend lines show that states in the Southern, Appalachian, and Old Midwest regions have been doing poorly. This problem predates the opioid epidemic, having persisted for decades, through multiple presidential administrations, health policy changes, and changes of power in Congress and in state legislatures. Older people in West Virginia, Mississippi, Kentucky, Oklahoma, and Alabama had relatively poor health in 1980 and have seen almost no improvement since then for women and very little for men.

State-level death rates at older ages give us a good, but not perfect, indicator of what’s happening to population health. Death isn’t the only health outcome that matters—although it’s certainly an important one. When death rates are high, other indicators usually look bad as well. And people do move around among states, but not as much as they used to, and not so much at older ages.

U.S. mortality rates in the global context

By world standards, improvement in mortality rates and life expectancy has been slow for the entire U.S., slower than in other rich countries and even some not-terribly-rich countries. We now rank 40th in the world by World Bank estimates in life expectancy at birth, behind countries like Costa Rica and Slovenia that spend vastly less than we do on the health sector … including biomedical research.

Why are we doing so poorly as a nation, and why in particular are some states stalling out and falling behind? What can be done about it?

Recommended reading

The story starts with smoking and obesity rates, but individual behavior is not a final nor a complete answer. Why have people in some states managed to control their health risks while those who live elsewhere have not?

For NIA, these are enduring concerns and we hope to see more research on them. A recent Request for Applications asked for studies to “identify mechanisms, explanations, and modifiable risk factors underlying recent trends of growing inequalities in morbidity and mortality by income, education, and geographic location at older ages in the United States.” That particular RFA has expired, but we still welcome new investigator-initiated applications.

The need for research on these life-and-death issues continues. We don’t have the answers yet, but NIA-funded research in recent years has sharpened the focus:

• R. Chetty et al., (2016) The Association Between Income and Life Expectancy in the United States, 2001-2014. JAMA 315: 1750-66. The association between household income and life expectancy (an average of mortality rates at all ages) exists everywhere, but is much stronger in some metropolitan areas than others.
• L. Dwyer-Lindgren et al., (2017) Inequalities in Life Expectancy among US Counties, 1980 to 2014: Temporal trends and key drivers. JAMA Internal Medicine, 177: 1003-1011. Geographic disparities in mortality at every age are large and growing over time, with county-level differences greatest at ages 65 to 85—ages when almost all Americans are covered by Medicare.
• J.K. Montez, A. Zajacova, and M.D. Hayward. (2017) Explaining Inequalities in Women’s Mortality across US States. Social Science and Medicine: Population Health 2: 561-571. State-level characteristics (such as tobacco taxes) accounted for 23 percent of the variation across states in men’s mortality rates at older ages, and 53 percent of women’s mortality.

As we begin 2018, if someone asked us how we’re doing, we’d all have to answer: “Not well.” In many states and counties, the answer would be, “Poorly.” Because we’re the National Institute on Aging, we seek to fund research that will contribute to reducing these disparities among regions of our nation. Please contact me or other program staff in our division if you’re interested in submitting an application for research in this area.