Whether or not you will live past 70 depends on a seemingly infinite number of small variables: genes, lifestyle, whether or not you are hit by a bus, etcetera. Tied into that prediction have long been race and location. Black males in southern cities, for example, typically have a shorter life expectancy than white males in northern rural areas.

But a new study finds that on closer scrutiny these previously generalized principles begin to fall apart. By studying survival beyond 70 on a county-by-county basis, a team of researchers found that a combination of social factors, such as education, marital status and income, were much more predictive than race or geography alone. The research was published online April 17 in PLoS ONE.

"While there is an enormous survival difference between some counties and others, it is the social and environmental characteristics of a given county and its population that matter the most," Mark Cullen, chief of General Medical Disciplines at Stanford School of Medicine and co-author of the study, said in a prepared statement. "Once certain factors—such as the fraction of adults in the county who finish high school, the fraction with managerial or professional jobs, and the fraction of adults who live in two-parent households—are accounted for, even geography, such as being in the South, is moot." Out of the hundreds of counties studied, they didn't find even a single serious outlier.

These social factors, of course, vary by area and region, thus creating the familiar national health disparity maps. To test their hypothesis, the researchers created a health model of blacks and whites in which all of the 22 socioeconomic variables were the same. "Amazingly, almost all of the projected mortality differences evaporate," Cullen said. "In this hypothetical experiment, the chances of survival until 70 would be almost identical for whites and blacks." When the researchers repeated this model based on gender, however, they were unable to erase the survival differences, suggesting that these socioeconomic factors are playing an important role in these current lamentable racial differences in health.

The researchers focused primarily on counties with both substantial white and black populations and individuals who had lived past the age of 30 (to reduce issues of infant and youth mortality). Because the study was conducted on a population basis, they did not have specific details about the people who died (for example, whether someone who died before 70 in a high poverty county was, themselves, poor or who just lived in the area).

The findings suggest that new targets for improving overall health for many are beyond the traditional medical research realm and instead should be basic social issues, such as the levels of schooling and career achievement.