By David Brown
Washington Post Staff Writer
Wednesday, May 14, 2008
The difference in death rates between highly educated and poorly educated people in the United States is very wide and growing wider, according to new research.
For Americans with less than a high school education, the risk of dying prematurely is on the increase -- rising most quickly for white women in that category. In contrast, the risk of premature death among college graduates is falling -- fastest of all for black men.
White high school dropouts are four times as likely to die young as white college graduates, up from a threefold difference in the early 1990s. Among blacks, the trend is similar but less dramatic.
The study, published online yesterday, is the second this spring to reveal an ominous trend toward worsening health and earlier death in disadvantaged segments of the population, especially in certain groups of women.
"Socioeconomic disparity in mortality is pervasive, and it continues to increase," said Ahmedin Jemal, an epidemiologist at the American Cancer Society who lead the study.
A co-author, Robert N. Anderson of the National Center for Health Statistics, said the growing gap "says something about the overall health of our population." He added: "The haves seem to be doing quite well, and their mortality is going down. But those who don't have their resources are not doing so well."
The researchers compared mortality rates from 1993 to 2001. For technical reasons, researchers examined deaths of people age 25 to 64. All were premature deaths in the sense that they occurred at a lower age than the life expectancies of the groups studied.
Whether education levels also predict health and mortality in people 65 and older is unknown. Medicare, the federal health insurance program for seniors, may blunt the effects to some extent. Nevertheless, the new study does provide a window on what is happening at the ages when most people are working and raising families, and what it reveals is both discouraging and encouraging.
While there is growing hazard for poorly educated and presumably lower-income Americans, race in itself does not strongly predict health outcomes. The group with the steepest decline in premature deaths was college-educated African Americans.
White female high school dropouts had the greatest erosion in their health, with their mortality rate rising by a little more than 3 percent a year over the nine-year period. A greater number of accidents, heart attacks, and cases of emphysema and cancer were responsible for about half this increase.
White male high school dropouts also had an increase in their risk of dying prematurely rise, by about 1 percent a year. Increasing numbers of accidents, suicides and cancers were largely responsible.
White male high school dropouts were 4.4 times as likely to die prematurely as white male college graduates. For white women, mortality in dropouts was 3.8 times that of college graduates.
For African American males with college degrees, death rates fell dramatically -- an average of 6 percent a year over the nine-year period. Fewer deaths from heart disease, cancer and, especially, AIDS were the main reasons. For black women who had finished college, mortality fell about 3 percent a year, driven mostly by fewer strokes, heart attacks and cases of cancer.
In general, education had a stronger influence on mortality in men than in women. This was true for both blacks and whites.
The study did not explore the reasons for the disparities, but the researchers were willing to speculate. "The trend is best explained by tobacco, obesity and high blood pressure," Jemal said.
The prevalence of high blood pressure in men and women who did not finish high school rose in the 1990s, putting them at higher risk of strokes, heart attacks and kidney failure. Obesity rose markedly for both sexes, as well.
At the same time, better-educated and wealthier groups have been more likely to adopt more healthful behaviors. For example, the fraction of U.S. women without high school degrees who smoked fell from 37 percent to 25 percent from 1974 to 2004. For college graduates, it fell from 26 percent to 9 percent, a much steeper decline.
A study published in April found that in about 1,000 U.S. counties, the life expectancy of women fell between 1983 and 1999. The counties tended to be rural, with many in Appalachia, the South and the Mississippi River Valley. Increases in cancer, Type 2 diabetes and emphysema were largely responsible.
One of the authors of the new study said he thinks that one factor contributing to the drop in life expectancy in those places is a "selective loss" of young people with college educations.
"What is left is a concentration of people in the lower socioeconomic strata," said Anderson, who is chief of the mortality statistics branch at the National Center for Health Statistics. "There is no way to prove that is happening, but it wouldn't surprise me if that is what is happening."
The new study is published in PLoS One, a journal published by the Public Library of Science. It was based on vital statistics from 43 states and the District; seven states had incomplete education information and were excluded. The analysis was confined to non-Hispanic whites and blacks "because of problems with reporting other racial and ethnic subgroups," the researchers noted.