Does schooling protect health? What about race and income?

Which of the following would you assume is the greatest risk to your health: obesity, smoking, dropping out of high school, being African American (versus white) or being poor? Public health efforts have struggled for decades to reduce obvious threats to health amenable to change, such as smoking and obesity. As a result, smoking rates are decreasing nicely, though the jury is still out on obesity. Less attention has been placed on linking education and income to health, or the role of race.

A new paper in the American Journal of Public Health reminds us of the importance of underlying influencing factors on health, such as race, income and education. Researchers at Columbia University's Mailman School of Public Health used a complex statistical method known as decision analysis to crunch numbers from several massive, nationally representative datasets. Their goal was to place health risks in context by comparing how many years of perfect health are lost to various influencing factors. The most harmful? Poverty. The average low income person, defined as an income less than 200% of the US federal poverty level, loses 8.2 years of perfect health due to their income status. About a third of the US population is in this income bracket.

Smokers came in second, surrendering an average of 6.6 years of perfect health to their addiction; high school dropouts were third, losing 5.1 years; next came African Americans, who compared to comparable whites lose 4.7 years; and finally the obese, whose size-related health risks cost them 4.2 years. Unlike some previous studies, health insurance neither increased nor decreased the number of years of perfect health a person would enjoy.

Of particular interest was the following: If you adjusted for education, poverty still resulted in a loss of 6.4 years of perfect health. In other words, even the well educated poor are still plagued by health concerns. Adjusting race for poverty and income, however, reduced the amount of perfect health lost to 1.7 years. So well-educated, wealthy African Americans are still more likely to suffer ill-health, but much less so than their less educated, poorer counterparts.

This type of mathematical modeling has all sorts of shortcomings in terms of its ability to tell us exactly what's going on in the real world. But since it is based on real world data, and is often use to help make treatment decisions it cannot be completely discounted.

Do these figures surprise you? Assuming they are true, what solutions come to mind? Do they make you reconsider what our health priorities should be?