Those are among the conclusions of the study by a team of researchers that has spent years teasing apart the regional and demographic differences in longevity in the United States. It sketches a picture of widening inequality among regions and is likely to add urgency to the debate over health-care reform and spending.
Led by Christopher J. L. Murray and Sandeep C. Kulkarni of the Institute of Health Metrics and Evaluation at the University of Washington, the study covers the years 1987 to 2007 and updates an earlier analysis of U.S. life expectancy through the 1990s. That study identified areas of the country where life expectancy was stagnant or declining; this report found that not only was that still the case, but more places are experiencing a decline. The study will be published Wednesday in the journal Population Health Metrics.
As they had previously, Murray and his colleagues found huge variation in life expectancy in the United States, with some of biggest extremes right around Washington. Fairfax County’s men, for example, had the longest life expectancy in the country, 81.1 years in 2007. In the city of Petersburg, 25 miles south of Richmond, life expectancy was 14 years less, and at 66.9 years among the lowest for men in the country.
The District is one of the places that has seen a big recent jump in life expectancy for a particular demographic group — black men, from 61.7 years in 1997 to 68.9 in 2007, a huge jump in demographic terms. The reason for the big increase isn’t entirely clear.
The region where life expectancy is lowest, and in some places declining, begins in West Virginia, runs through the southern Appalachian Mountains and west through the Deep South into North Texas. Places of high life expectancy are more scattered. In addition to Northern Virginia they include counties in Colorado, Minnesota, Utah, California, Washington state and Florida.
Although the research didn’t look for causes, there are several possible reasons for the slowing of longevity in parts of in the United States. The rising rate of obesity and plateauing of the smoking cessation rate among women are two. Poorly controlled blood pressure and a shortage of primary-care physicians are two others.
What surprised Murray and his team was that despite increased consciousness about disparities and per capita spending on health care that is at least 50 percent higher than European countries, the United States is falling farther behind them with each passing year.
“My expectation was that in the last decade we would at least be keeping up in terms of the pace of progress. But that’s not what’s happening,” said Murray.
Among the places with the biggest increases in life expectancy in the last two decades is New York City. The arrival of triple-drug antiretroviral therapy for HIV infection, along with dramatic declines in homicide and infant mortality, caused great gains in the late 1990s.
“The critical insight this work underscores is something that we’ve known for years — that both health and health care are produced locally,” said Elliott Fisher, a physician at Dartmouth Medical School who studies regional variations.
Life expectancy is an abstract concept that summarizes the health and threats to longevity that exist at a particular moment in history. It is not an actual measure of how long people are living.
To calculate life expectancy, researchers imagine a large group of people — say, 100,000 — born in a particular year, such as 2007. They then statistically move that “birth cohort” through life, subjecting it to the risk of dying that exists for every five-year period (0-5 years old, 6-10 years old, 11-15 years old, etc.) in that particular year. A few members of the group will die as infants, a few as children, a few as young adults, some in middle age, most in old age, and a few in extreme old age. The average age at death is this artificial population’s life expectancy.
The actual risk of dying at each age is determined from death certificates, which also include information about where the deceased person lived most of his or her life. Because of that, life expectancy can be calculated for geographical areas as small as counties. However, in the new study Murray and his team collapsed the nation’s 3,147 counties and large cities into about 2,400, merging some if their populations were too small to produce statistically meaningful populations.
In 2007, life expectancy for American men was 75.6 years, compared with 73.7 years a decade earlier. For women it was 80.8 years in 2007, up from 79.6 in 1997. Both rank the United States as 37th in the world in 2007.
The county-based life expectancy for men ranged from 65.9 years in Holmes County, Miss., to 81.1 years in Fairfax County. Women had the longest life expectancy in Collier County , Fla., which includes Naples (86 years) and the shortest in Holmes County, Miss. (73.5 years).
Among states, Mississippi fares the worst. Of its 82 counties, 26 had life expectancies for men in 2007 that were roughly 66 to 69 years.
The most striking trend was the slowing of life expectancy in women. From 1987 to 1997, there were 314 counties with either a loss of life expectancy or no growth in it. From 1997 to 2007 there were 860 counties in which that was the case.