The U.S. Mortality Crisis as a Preston Curve Reversal

For decades, as Americans grew wealthier, they lived longer. This predictable relationship—where rising economic resources drive improvements in public health and medical technology—has served as a cornerstone of demographic understanding.

However, this pattern has recently broken down. Despite continued growth in real per capita income (the average income per person adjusted for inflation), U.S. life expectancy has stagnated or even declined during the 2010s. This disconnect suggests the country is no longer efficiently using its accumulated wealth to protect the health of its citizens.

A new study from the University of Pennsylvania, Yale, and Boston University characterizes this phenomenon as a "Preston curve reversal." The researchers argue that the U.S. mortality crisis is not just a story about specific causes of death, such as drug overdoses. Instead, it is a systemic failure to translate aggregate resources into broadly shared longevity gains.

The breakdown of the prosperity-longevity link

To understand this crisis, the authors rely on the Preston curve. This is a tool used to describe the ecological relationship (the connection between group-level traits) between a population's income and its life expectancy. Historically, as societies become richer, they build better infrastructure and public health networks. This process allows them to "translate" economic growth into extra years of life.

Until 2010, the United States followed this classic pattern. As seen in, both states and counties exhibited a simultaneous rightward and upward shift in their Preston curves.

As they became richer (moving right on the x-axis), they lived longer (moving up on the y-axis). Economic success fueled biological success.

The study finds that this loop snapped after 2010. While the U.S. continued to get richer, the curves failed to shift upward. The authors describe this as "decoupling." The engine of economic growth is still running, but it is no longer connected to the wheels of longevity.

Mapping the decoupling and divergence

The researchers employ a mathematical framework to quantify how this relationship has changed. They model life expectancy as a function of real per capita income using a second-order polynomial (a mathematical curve that can bend). This allows for the non-linear nature of how wealth affects health.

The authors identify two critical structural changes in the post-2010 era:

1. Decoupling: The curves shifted to the right (indicating more income) but failed to move up (indicating no longevity gain). This means additional resources are no longer yielding the expected dividends in lifespan.
2. Divergence: The curves became steeper. A steeper slope indicates that the relationship between income and longevity is becoming more unequal. High-income areas manage to maintain longevity, while lower-income areas experience sharper declines.

The robustness of this finding is a key part of the methodology. To ensure the results were not just a side effect of the Great Recession, the authors tested alternative "temporal anchors" (different starting points for comparison) .

They also substituted education for income. In both cases, the authors report that the decoupling and divergence remained clearly visible.

Evidence of a systemic translation failure

The study suggests this is a broad, structural issue. The authors report that this pattern of stagnation and decline appears across major population subgroups .

This includes both men and women, and both Black and White populations. This suggests the crisis is not a "compositional artifact" (a statistical illusion caused by changes in group makeup). It is a widespread shift in the national mortality regime.

The socioeconomic decomposition analysis provides further evidence. The researchers asked a counterfactual: if the relationships between socioeconomic factors and health in 2010 had simply continued into 2019, what would life expectancy look like?

The authors find that improvements in traditional markers—such as per capita income, wealth, education, and health insurance coverage—should have predicted higher life expectancy than was actually observed .

Essentially, the "inputs" for a long life improved, but the "output" failed to materialize. This reinforces the conclusion that the "translation" mechanism is failing.

Limits of the diagnostic approach

While the study provides a powerful diagnostic tool, it has limits. The authors explicitly state that their analysis is ecological and descriptive. They look at large groups rather than individuals. Consequently, they do not identify specific causal mechanisms. They do not prove that a specific policy caused the decline.

The paper does not adjudicate between competing theories of why this is happening. The authors suggest several possibilities for future research. Perhaps rising costs of living mean a dollar buys less "health" than it once did. Perhaps social alienation is undermining public health. Or perhaps the decline is driven by specific aging cohorts (groups of people born in the same era). Because the study focuses on identifying the pattern, it leaves the search for the trigger to future studies.

The verdict: A crisis of social organization

Is the U.S. mortality crisis solved by simply growing the economy? According to this research, the answer is no.

The study demonstrates that the U.S. has entered a period where the traditional link between prosperity and health has been severed. The "Preston curve reversal" shows a decoupling of wealth from well-being. It also shows a divergence in how that wealth is utilized across different geographies.

The findings suggest that focusing solely on income growth may be insufficient. If the "translation" mechanism is broken, adding more resources to a broken system will not yield the expected results. The challenge is not just a lack of resources, but a failure in the social organization of longevity.

Figures from the paper