To call Albert DeLardi of Westport active is something of an understatement. The 91-year-old retired salesman volunteers at the town’s playhouse, is a member of a local men’s group and is a regular at the Westport Center for Senior Activities. On a recent morning, he can’t talk long because he has a workout class to attend at the senior center.
“It depends on the individual if you’re active, you get involved and you do things,” the retired salesman says of longevity.
In Westport, people tend to be active and live longer than people in many other parts of the state. The life expectancy at birth in the affluent, coastal Connecticut town ranges from the low to high 80s, well above the state and U.S. average. The life expectancy for one area is 89.1, the highest for any neighborhood in Connecticut, and more than two decades greater than some neighborhoods in the state.
That’s according to estimates released last year by the U.S. Centers for Disease Control and Prevention as part of its U.S. Small-Area Life Expectancy Project. The never-before-tabulated, hyper-local estimates are based on mortality data compiled at the Census tract or neighborhood level. The project allows residents across the U.S. and most of Connecticut to look up their neighborhood’s life expectancy by address. The tool is available here. The data reveals the sometimes dramatic disparities between neighborhoods and within cities and towns across the state as well as the U.S. as a whole, and in so doing paints a compelling picture of how our personal health is influenced by our address.
“Health outcomes are largely not just dependent on a person’s biology or on their capacity to make good health behavior decisions, which we like to believe,” says Wizdom Powell, director of the UConn Health Disparities Institute and associate professor of psychiatry at UConn Health. “Your decisions around the health behaviors that you take on are constrained by your environment. You could be a person who has every intention of eating the right foods, or going to school and getting a quality education, but if those resources are not available in your immediate geography, or in the context of where you live, then you’re not going to be able to follow through, even with those best intentions.”
Disparities in outcomes
The impact a neighborhood has on health is particularly apparent when comparing cities to the suburbs, or when looking at vastly different neighborhoods within cities. As the Connecticut Health Foundation notes, a neighborhood in northeast Hartford has the lowest life expectancy in the state at 68.9. This is a full 15 years lower than in nearby West Hartford center. According to the foundation, “in New Haven’s Newhallville neighborhood, life expectancy is 71.7 years. In the neighborhood next door, Prospect Hill, life expectancy is more than a decade longer: 82.3 years.” Wide ranges in life expectancy are seen in cities across the state: New London has a 13½-year difference in longest- and shortest-lived neighborhoods; Norwalk has a gap of 11.6 years; Torrington has a range of 14 years.
In one Bridgeport neighborhood, State Street serves as a stark dividing line. The life expectancy for those who live on the south side of the street bordered by Park Avenue to the east, Lisbia Street to the west and the railroad tracks to the south is just 70.4 years. Across the road on the north side of State Street, life expectancy jumps either five or eight years. In the other direction, a few blocks farther south, closer to the water, life expectancy increases by more than a decade to 81.4.
There are also surprising disparities within and between suburban towns. For instance, the Westport neighborhood with the highest life expectancy in the state — an area on the western side of the Saugatuck River between routes 1 and 15 — outshines its neighbors within Westport by between three and seven years. One part of Avon has a life expectancy that is almost seven years greater than any other area in Avon or the surrounding area. Pinpointing what accounts for these smaller differences is difficult, but we should not dismiss them out of hand, experts say. Powell says that variations of one to two years might be too small to really mean much, but “when you start getting into a difference of like seven or six years, my antenna goes up.”
One possibility is that there are more women in a given neighborhood, which would skew the life expectancy higher, as women live several years longer than men on average, Powell says. But it’s also possible that there is something about these specific neighborhoods that is encouraging health. “It is true that where you live matters when it comes to your life expectancy,” Powell says. “Why that matters is still an issue for scientists to really, truly disentangle.”
Dr. Brita Roy, assistant professor of medicine at the Yale School of Medicine and director of Population Health for Yale Medicine, studies disparities in health outcomes and the way positive psychology can prevent disease. She says pinpointing the reasons for disparate outcomes in life expectancy within towns like Westport and Avon “is an interesting conversation,” because “typically we focus on differences between really high- and low-income tracts.” She adds that it could be “things like the social connectedness of a neighborhood, for example, and the walkability of a neighborhood is also really important because that often influences regular physical activity. … Just being able to do your errands by foot or by bike has been more strongly associated with longer life expectancy, even compared to frequency of going to the gym.”
Another factor that gets overlooked, Roy says, is commute times. “Longer commute times, especially commuting alone, has also been associated with lower life expectancy.” But Roy stresses that all of this is speculation without a closer study of the area, and she points out that there’s a greater margin of error when looking at life expectancy within smaller areas.
The power of place
Debarchana Ghosh, a professor of geography at UConn who studies health geography and lifespan, says research indicates that only about 10 or 15 percent of your life expectancy is accounted for by genetics. The rest is influenced by what is called the exposome, all the external influences a person encounters throughout his or her life. “A person living in a very green neighborhood may or may not like to walk, and we see that. But my choice is not restricted,” Ghosh says. However, a person in a neighborhood without parks, or without safe parks, doesn’t have the option of walking. And where we live can also influence whether we feel like walking or not. “I used to live on the border of Manchester and Glastonbury,” Ghosh says. “When I used to come to Glastonbury, more people are active, more people are running down the Main Street near the Whole Foods.” In one town she’s subtly being influenced to exercise, in the other she isn’t. That could explain the variation in life expectancy between the neighboring towns. In Glastonbury, the life expectancy in every neighborhood is more than 80 years, while in Manchester it is below 80 for more than half of the neighborhoods.
Ghosh says that consulting this type of data, along with other factors and considerations, before a move makes sense, even though it would be far from a perfect indicator of how long one individual might live. For instance, if a wealthy person moved to a poor neighborhood with a low life expectancy, she says that person’s children would likely have better health outcomes than their neighbors, even if they couldn’t use their wealth. Ghosh hypothesizes this would be the case, “because their parents did not face those exposures, those detrimental exposures living in that neighborhood, or a similar neighborhood with low expectancy.”
These various influences on lifespan are part of why Lloyd Mueller, principal investigator and senior epidemiologist with Connecticut’s Department of Public Health, cautions against reading too much into these numbers at the individual level.
“I think it’s good as a snapshot of a population. It isn’t necessarily good for tracking life-course experience,” he says. “There’s residential mobility throughout the U.S. and the smaller the geographic unit you’re looking at the more likely people are to not have been there 10 years ago.” This can muddy the waters when looking at a small area and trying to measure life expectancy. In addition, he says there can often be more than meets the eye to statistics of any kind. “The reason that emphysema rates are high in Arizona is not because there’s something about the environment of Arizona. It’s because people move there for relief. So it’s counterintuitive.”
Reading between the numbers
The U.S. Small-Area Life Expectancy Project was launched in partnership between the CDC and the nonprofit Robert Wood Johnson Foundation and the National Association for Public Health Statistics and Information. It took three years to compile information from more than 65,000 Census tracts nationwide, including more than 700 in Connecticut.
Elizabeth Arias, director of the U.S. Life Table program, was the lead researcher for the project. She and fellow researchers looked at deaths that occurred during a six-year period between 2010 and 2015 in various age groups and then matched those deaths with specific addresses. Because they were looking at areas with small populations, they sometimes did not have enough deaths within a specific year for a specific age group to make an estimation. To remedy that they would pool the years together for the six-year period. When that still didn’t result in a usable number, they would predict death rates based on areas with similar demographics. “Overall, the percentage of the estimates that were based on predictive values were not that large,” she says. “We didn’t want to model everything. We wanted to have a gauge of what mortality looked like based on observed data as opposed to all modeled data.”
Even so, because of the modeling used for this estimate, Mueller, from Connecticut’s Department of Public Health, says the life expectancy estimates may not be as accurate as those based on large populations. “There is only about 2 percent of all Connecticut’s tracts where no modification was made,” he says. He adds that there are often geographical mistakes made at the local level in Connecticut that may not have been accounted for.
What residents say
Beyond the statistics, visiting the Westport neighborhood with the highest life expectancy makes one thing clear: this is a nice place to live. There are several people out jogging and walking, and the green spaces in the area include a natural playground at the Leonard Schine Preserve, which abuts the property of the headquarters of Bridgewater Associates, the world’s largest hedge fund. But as nice as the area is, it is no nicer-seeming than other parts of town with lower, but still higher-than-average life expectancies. Residents throughout Westport appear to enjoy similar access to health care, good schools, grocery stores and critically acclaimed restaurants with farm-to-table ingredients. Appearances, of course, can be deceptive, but town residents were also at a loss to explain the differences in outcomes within different areas of Westport.
Lou Mall, a town representative for a district that includes parts of the area with the highest life expectancy, says, “I don’t think there’s anything different in the water we drink or the air we breathe.” He adds that what seems to make the place special is “its wonderful neighborhoods and wonderful neighbors. … It’s just a nice, safe place to raise your children.”
DeLardi, the active 91-year-old Westport resident, agrees. “We’re neighborly,” he says, though “we’re not over on each other’s porch all the time.” He lives just outside the area in town with the highest life expectancy, but he is within walking distance to the train station. When he used to commute to New York City several days a week, he’d make the one-mile trek there and back. Maybe it was something subtle in his neighborhood that encouraged those walks.