A century-long study led by Yale School of Public Health researchers reveals vast disparities in life expectancy across U.S. states.
A groundbreaking new study from the Yale School of Public Health uncovers major differences in life expectancy across the United States. Over the past 100 years, where you live has played a huge role in how long you can expect to live.
Researchers analyzed more than 179 million death records from 1969 to 2020. Instead of looking at deaths year by year, they focused on birth cohorts, or groups of people born around the same time. This approach gives a clearer picture of how early life experiences and long-term social conditions shape health over a lifetime.
Published in JAMA Network Open, the study found that some states, especially in the Northeast and on the West Coast, saw big improvements in life expectancy. But others, particularly in the South, showed little to no progress over the entire century.
Credit: Ephemia Nicolakis“For females born in some Southern states, life expectancy increased by less than three years from 1900 to 2000,” said Theodore R. Holford, the Susan Dwight Bliss Professor Emeritus of Biostatistics at YSPH and lead author of the study. “That’s a staggering contrast when you consider that in states like New York and California, life expectancy rose by more than 20 years over the same period.”
States in the Northeast and West, along with the District of Columbia, recorded the greatest gains. Notably, the nation’s capital had the lowest life expectancy for the 1900 birth cohort but improved by 30 years for females and 38 years for males by 2000.
The Role of Socioeconomic and Policy Differences
By contrast, states like Mississippi, Alabama, and Kentucky saw minimal gains, particularly among women, suggesting that systemic factors — like socioeconomic disadvantages, limited access to health care, and weaker public health initiatives — have lasting impacts on mortality, researchers say.
“These trends in mortality and life expectancy reflect not only each state’s policy environment but also their underlying demographics as well,” said Jamie Tam, an assistant professor of health policy and management at YSPH and co-author of the study. “It’s not surprising that states with fewer improvements to life expectancy also have higher rates of poverty for example.”
The researchers used an age-period-cohort model, allowing them to disentangle the effects of aging, historical events, and generational influences on mortality. This method captured how early-life exposures — such as access to sanitation, vaccinations, or tobacco — shaped health trajectories later in life.
“Looking at mortality trends by cohort gives us a more accurate reflection of the lived experiences of populations,” Holford said. “It shows the long-term impact of policies and social conditions affecting the life course of populations that might otherwise be invisible in year-by-year comparisons of mortality rates from different generations.”
Measuring Healthy Aging and State-Level Variations
The team also examined the rate at which mortality increased after age 35, which is summarized by the number of years it takes for an individual’s risk of death to double. Longer doubling times indicate healthier aging. Again, researchers observed stark differences between states: New York and Florida showed slower mortality increases, while Oklahoma and Iowa saw faster mortality escalations.
Holford emphasized that these patterns aren’t just a reflection of the past. “The disparities we see today are the result of decades of cumulative effects — on smoking rates, health care access, environmental exposures, and public health investments,” he said. “Without conscious policy changes, these gaps will likely persist or even widen.”
The researchers pointed to California and Kentucky as examples of how local health policies can influence mortality rates. California was an early adopter of smoke-free workplace policies in 1995, which led to entire generations of young people growing up in smoke-free environments, as well as generations of working-age adults who were prompted to quit, the research said.
In contrast, Kentucky essentially made no effort to control cigarette smoking, resulting in higher cigarette use and, therefore, higher mortality compared with California. Similar patterns for smoking and higher mortality were observed in West Virginia, Oklahoma, Arkansas, Tennessee, Louisiana, Mississippi, and Alabama, the study said. Lower socioeconomic status is also associated with higher mortality risk; states that are socioeconomically advantaged appeared to have improved mortality rates more quickly than other states, according to the study.
The researchers hope their findings encourage greater focus on interventions — like tobacco control, health care access, and environmental protections — that can transform life expectancy outcomes for future generations.
“This research highlights the importance of viewing health through a generational lens,” Tam said. “The benefits of health interventions ripple across lifetimes.”
As Holford put it, “Where you are born shouldn’t determine how long you live. But in America, it still does.”
Reference: “All-Cause Mortality and Life Expectancy by Birth Cohort Across US States” by Theodore R. Holford, Lisa McKay, Jamie Tam, Jihyoun Jeon and Rafael Meza, 28 April 2025, JAMA Network Open.
DOI: 10.1001/jamanetworkopen.2025.7695
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.
29 Comments
Boring.
Where the hell are these numbers from? American life expectancy in 1900 was about 47. Nowhere neeeeear the much higher ranges shown in the bar graphs here.
States that did better invested in science and education. The laggard states remained stuck in their 0 AD mindsets.
BS. First you have to just believe the researchers conclusions and the validity of their data. For example they compare California and Kentuck and point to the fact that California outlawed smoking in the workplace in 1995 as an explanation of the differences in life expectancy between the 2. But this is extremely weak. Kentucky outlawed smoking in the workplace in 2003, just 8 years after California. And smoking in Kentucky (and long term smoking we have been told has a major impact in life expectancy ) has decreased among adults in Kentucky since 1965 by 80% and among youth by 73% and we are to believe that Kentucky life expectancy since 1900 has NOT improved.
The study claims California’s life expectancy between 1900 and 2020 increased by 20 years. California current life expectancy is 80 meaning that life expectancy in California back in 1900 was 60 years making it one of the lowest of any State. Kentucky’s life expectancy is currently 75 years so the life expectancy in 1900 if in fact there has been virtually NO CHANGE was close to 75 years. And again this no change in life expectancy in Kentucky since 1900 to present has occurred despite the enormous decrease in smoking rates in Kentucky. But the researchers attribute the difference in life expectancy between California and Kentucky is the 8 year difference in the states enacting smoke free workplace environment. Also not addressed in the research report is the fact that Kentucky (like West Virginia, Alabama and Arkansas with Kentucky are 4 of the 5 states with the lowest life expectancy which is only a few years difference from the top 5 States) is a coal mining State as are 4 of the 5 States with the lowest life expectancy.
So that your arrogance and elitism and stuff it.
Intelligence, ambition, regional cuisine, climate, genetics, etc. also would be factors.
All factors you mention (minus climate) influence longevity – even within each state. Climate and politics are factors that may explain the strong correlation to state boundaries.
Its peoples diets and jobs, mining, working in bad conditions on farms. All these play a major role in like longevity.
If I were working in a mine or on a farm, I’d try my best so that the next generation doesn’t have to work there. I’d vote for better education and to attract newer, cushier tech jobs. I’d vote for unionization of mining and farm jobs. The residents of the laggard states mentioned consistently vote against all three.
Your assessment sounds like a rationalization of support for your personal hobby horses. Do you have independent sources that you can cite that demonstrate a strong correlation between investment in “science” (whatever it is that you mean by that generalization), and the increase in longevity at any level of granularity? If you do, can you demonstrate that it isn’t a spurious correlation?
Would you expect any correlation between the various causes of death such as felony violence, child mortality from disease, adult mortality from aging, infectious diseases, diet, access to medical care, and accidents in the workplace and home? If so, do you have a study with a co-correlation matrix to break out the individual contributions to total mortality?
I am glad you have laid down some standards. I have to prove that states like California and Massachusetts have more investments in science than states like Mississippi and West Virginia. Tough job! But I should do that.
But … you need to stick to your own standards. First you prove your statement: “Your assessment sounds like a rationalization of support for your personal hobby horses.” There are many other sentences for which I will hold you to the same standard, so do get on with it.
Yes, you should backup any bold claims made about reality. Instead, you segued into a distraction to attempt to make me support my claims. However, unlike you, I made no generalized claim about how things are. I said, “sounds like,” which is a personal opinion for which it is not necessary to provide proof; the fact that I said it is self-evident proof that it is what you have led me to believe based on your unsupported remarks.
It isn’t just a matter of proving that “states like California and Massachusetts have more investments in science than states like Mississippi and West Virginia.” One also has to prove that it isn’t a spurious correlation or just coincidence, and that you have the cause and effect correct. I suspect that if abundant gold had been found in Mississippi in the 1850s it would be one of the states with greater longevity. Although, the enervating humidity may have served as a drag on the economy before air conditioning had been invented.
You have made several statements without the qualifier that they are your personal opinion. You have expressed them as though they are common knowledge or that you can readily provide support for them, which you have not done.
States with the worst outcomes continue to elect Republicans that have no interest in average citizens. But they will require posting the 10 Commandments on classroom walls.
This study is flawed and this summary plays up those flaws without any comment. The paper itself does state that the results should not be used for prediction. For example, it projects that females born in Washington DC in 2020 will live to age 93 on average. Is this credible? It’s an example of extrapolating from the steep improvement slope from 1900 (where WDC was worst) to later data (where it was online with other regions), but on up. Any linear model will have this problem … improvement (or decline) from the worst will end up exaggerating the future.
This synopsis focuses on the nonsensical results.
That said, many of the conclusions about causes of qualitative differences, seem reasonable enough.
“I’d vote for better education and to attract newer, cushier tech jobs.”
It sounds like you have already decided what factors contribute most to longevity, in the absence of other rigorous, in-depth studies proving it. That isn’t science, that is politics.
Please enlighten me why you think: “It sounds like you have already decided what factors contribute most to longevity” – I said what I’d vote for, not what contributes to longevity.
The implication of you being willing to vote for a particular ‘remedy’ is that you believe it would solve the problem. That is a weak attempt to side-step my remark.
I don’t see any mention of a few other things such as …
1. Population mobility change from 1900-2000.
2. Deaths from criminal sources.
3. Natural disasters.
4. How automobile related deaths influence results since many youth now never reach adulthood.
5. Immigration legal and illegal, affect on overall results as they tend to have shorter life spans than natural born populations.
And I’m sure there are many more things that could skew results.
Education and functional public health efforts, revered in those educated blue regions and reviled in the dropout South. MATTER.
Everytime I see one of these “studies”, the same states are battered and yet in many rural areas, people grow there own food and eat less fast or over processed food. It may be due to choice or for financial reasons but it is healthier. I read a lot of supposition and opinion but little actual data.
At least the air is cleaner in rural areas – due to lack of pollutants. Less availability of over-processed food also likely results in healthier diet.
You have apparently never lived through a major dust storm in an arid rural region. Many rural areas down-wind of cities share the smog of the upwind cities. Did you read the article published today, https://scitechdaily.com/the-surprising-way-plants-may-be-polluting-the-air-you-breathe/ ? Terpenes have long been known to contribute to poor air quality, especially in areas such as the Great Smoky Mountains of Appalachia. Furthermore, forested rural areas are more prone to seasonal wildfires such as are currently being experienced in Canada and shared with the USA.
Do you actually think that people who live in “rural areas” universally prepare their own food like some kind of off-the-grid ‘Preppers?’ Twinkies and their equivalent are available readily where they buy gasoline and diesel for their vehicles, if not the grocery stores they frequent to buy groceries.
Once again you have made broad-brush statements as though they they are common knowledge that don’t need support. You are just sharing your ignorance. You come across as an inexperienced, young urbanite who knows little about life outside the city you live in.
What justifies you denigrating the South by making the claim that “education and functional public health efforts” are “reviled” in the “dropout South?”
Do you honestly believe that people in the South enjoy poor health and short lives? Or, are you expressing your disdain for people beneath you in “fly over” regions? I am personally offended by your arrogance in suggesting that political affiliation is somehow related to education and intelligence.
Such clear bias without even an attempt to justify the remarks has no place in a website dedicated to science and technology.
No surprise. Blue states are more generous to their people as far as healthcare and even SNAP and other programs. They use the tax dollars for their people. They follow CDC guidelines. They trust the experts as we all should. Here is an example , I once needed to get the ACA. In Pennsylvania it was $70/mt with a $1,000 deductible. When I went back to Florida that same year it went up to $450/mt. and a much higher deductible. It was around $5,000. Also, Blue states invest in education, new technology, clean energy and clean drinking water. Red states do not. It’s always been that way. Democrats have always supported the middle class and those that need a hand back up. Red states support the wealthy only and don’t care about their people, only their bank account. They support corporate America and vote in their favor cause they have to, they take millions from their lobbyists. Those votes are not good for their constituents as companies are deregulated and cause more pollution and increases cancer risk. They prey on their uneducated voters and use a few cards in their deck to get them like the gun card, abortion card, and racism card. Also Fox news who lies for a living is very popular in those uneducated red states. They are an easy target to brainwash.
“Where you are born shouldn’t determine how long you live. But in America, it still does.”
It also does in the world. Japan, and pockets in other countries, where longevity puts the NE USA and West Coast to shame, demonstrating that it isn’t just “education and functional public health efforts” that determine longevity when looked at more broadly. For those who brag that their enlightened politics are responsible for their, and their similarly enlightened cohorts longevity, they have to explain why we find pockets of centenarians throughout the world that don’t fit their preconceptions.
Just the other day I was watching a documentary that says these big fancy expensive colleges like Stanford Yale Harvard they built and got started with them opium money and I’m talking way way back in the day as well as the John Perkins for the blind was also opium built and funded back in the 1800s sounds like a big money laundering thing even back then
Leland Stanford is characterized as an industrialist, attorney, and politician by Wikipedia: https://en.wikipedia.org/wiki/Leland_Stanford . However, he probably made most of this money working as a merchant selling goods to gold miners at inflated prices, which was common when goods were in short supply and gold was plentiful. He may have sold opium, as it was not illegal at the time, and probably wanted by Chinese miners and railroad workers. Opium and cocaine were not unheard of in unregulated medicines intended to treat the symptoms of various common diseases. I think that you are just sensationalizing what was a different way of life 175 years ago.
you realize that 2 of those 3 schools you list were founded before the american revolution. so i’m unsure what point you’re trying to make.
As an academic research physician, these conclusions are embarrassing for medicine. This is a classic case of fitting the data to your biases. You think moving people from Mississippi and Louisiana to New York would change their life expectancy?!
These comments make me laugh. I’m 73. Think I’ll live longer?