American Scientist

As scientists from all disciplines know, where you look influences what you find.

By examining death rates county-by-county, public health researchers found that life spans, especially among women, fell in some United States counties. The numbers were small but such a reversal of fortune was, well, shocking.

"We started looking at disparity questions. This became arguably a bigger finding and a more depressing finding," said Majid Ezzati, an associate professor at the Harvard School of Public Health.

Ezzati, whose research results appeared in PLoS Medicine, is among scholars dicing and splicing mortality data to create more precise pictures of lifespan trends in the United States. The nation is not known as the life-expectancy leader in the developed world. The Centers for Disease Control and Prevention ranks the life expectancy of women in the U.S. 27th globally.

But on average, things look pretty good: Americans still live longer than they used to. From 1961 to 1999, the time period Ezzati studied, life expectancy increased nationally from 67 to 74 years for men and from 74 to 80 years for women.

Still, research published in 2006, also in PLoS Medicine, emphasized how profoundly the story changes when you start sorting out disparate groups. For example, a life-expectancy gap of 21 years has been documented between 5.5 million Asian females and 3.4 million urban black men considered at high risk of injury or chronic disease. Native Americans living in or near reservations in Arizona, Colorado, New Mexico and Utah, as a group, don't match the average lifespan. Nor do low-income whites living in Appalachia and the Mississippi Valley.

Ezzati and his team focused on 2,068 counties (or combined, adjacent counties when single county population numbers were too small) since reliable death data can be found for these government entities. And counties are a means to explore geographic variability in lifespan.

Most striking were differences among people on the extremes. Among counties that make up 2.5 percent of the U.S. population with the highest or lowest life spans, men lived nine years longer in 1983 but 11 years longer by 1999. For women, the gap grew from 6.7 years to 7.5 years.

Between 1983 and 1999, male and female life expectancies had a statistically significant decline in 11 counties for men, representing 0.5 percent of the population, and in 180 counties for women, representing 3 percent of the population. The average lifespan decline averaged 1.3 years in those counties for both the men and the women.

Researchers did not break down their findings by race, but they did detect a socioeconomic pattern. People who fared worse in life-expectancy measures lived in U.S. regions long associated with poverty, including Appalachia and near the Mississippi River in the Deep South.

Preventable health problems seemed to drive the downturn, particularly diseases affected by cigarette smoking and excess weight and obesity. Increases in lung cancer, chronic obstructive pulmonary disease and diabetes were observed in the lowest-scoring counties. There was also a slowdown in the decline of heart disease, something U.S. citizens experienced on a national scale beginning in the 1960s.

Ezzati is hopeful his results eventually will help people who find themselves living in counties with the lowest-end life spans. For one, his team's research identified where they live. On top of that, it documented a higher than average prevalence of preventable diseases in their environs. With the right outreach, public health teams might be able to turn things around and increase the likelihood that people who need it get treatment for hypertension, for instance, or help quitting cigarettes.

"Reducing blood pressure and tobacco use is something this country has done a very good job of dealing with at the national level," Ezzati said. "Why is it that what we know how to do is not getting to people who need it most?"

Declines in life expectancy normally are linked to calamitous changes in a country, say the impact of the collapse of the Soviet Union and its healthcare safety net on Russia or the firestorm spread of HIV and AIDS across parts of Africa.

But James Vaupel, executive director of the Max Planck Institute for Demographic Research in Germany, saw a certain inevitability in the Ezzati team findings in the United States.

"Disparities are getting wider. We are moving in the wrong direction. People should be paying more attention to help those who are least well off," said Vaupel, who is also on the faculty of Duke University.

Ezzati agrees that disparities in health status are more extreme in the United States than in other developed countries. But he doesn't rule out the chance that a similar type of study in other parts of the world could turn up similar discrepancies.

"Other countries could have periods like this or versions of this," he said. "It would have to be looked at empirically."—Catherine Clabby